CN116993827A - A relocation method, device and electronic equipment - Google Patents

Abstract

Translated fromChinese

本申请实施例提供了一种重定位方法、装置及电子设备，涉及图像处理技术领域。基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定目标机器人的初始参考位姿；获取图像采集设备所采集的多帧参考图像，并基于初始参考位姿，确定每帧参考图像对应的参考位姿；利用多帧参考图像和每帧参考图像对应的参考位姿，构建区域局部地图，并确定与区域局部地图模板匹配的目标子地图；对图像采集设备所采集的当前图像与目标子地图所关联的各个基准图像进行相似度匹配，确定与当前图像所匹配的目标基准图像，并根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿。应用本申请实施例提供的方案，可以对移动机器人进行重定位。

Embodiments of the present application provide a relocation method, device and electronic equipment, which relate to the technical field of image processing. Based on the first frame of images collected by the image acquisition device mounted on the target robot, determine the initial reference pose of the target robot; obtain multiple frames of reference images collected by the image acquisition device, and determine the corresponding reference image of each frame based on the initial reference pose reference pose; use multi-frame reference images and the reference pose corresponding to each frame of reference image to construct a regional local map, and determine the target sub-map that matches the regional local map template; compare the current image collected by the image acquisition device with the target Each reference image associated with the submap performs similarity matching to determine the target reference image that matches the current image, and based on the target pose associated with the target reference image, determine the relocation pose of the target robot relative to the prior map. By applying the solutions provided by the embodiments of this application, the mobile robot can be repositioned.

Description

Translated fromChinese

一种重定位方法、装置及电子设备A relocation method, device and electronic equipment

技术领域Technical field

本申请涉及图像处理技术领域，特别是涉及一种重定位方法、装置及电子设备。The present application relates to the field of image processing technology, and in particular to a relocation method, device and electronic equipment.

背景技术Background technique

随着人工智能技术的不断发展，移动机器人被广泛应用于各类场景中。例如，工厂中，利用巡检机器人进行巡检；家庭中，利用扫地机器人清扫地面等。With the continuous development of artificial intelligence technology, mobile robots are widely used in various scenarios. For example, in factories, inspection robots are used to conduct inspections; in homes, sweeping robots are used to clean the floors.

通常，在移动机器人首次进入某个区域时，可以首先对该区域的区域探索，并构建该区域的探索地图，其中，该探索地图也可以称为先验地图；这样，在该区域中执行指定任务的过程中，移动机器人可以实时确定自身相对于上述先验地图的位姿。其中，所谓位姿是指移动机器人的位置和姿态，并且，上述移动机器人确定自身相对于先验地图的位姿的过程，可以称为移动机器人的定位过程。Usually, when a mobile robot enters a certain area for the first time, it can first explore the area and construct an exploration map of the area. The exploration map can also be called a priori map; in this way, the specified During the task, the mobile robot can determine its own position relative to the above-mentioned prior map in real time. Among them, the so-called pose refers to the position and attitude of the mobile robot, and the process of the above-mentioned mobile robot determining its own pose relative to the a priori map can be called the positioning process of the mobile robot.

然而，在很多情况下，移动机器人会发生定位丢失的现象，即移动机器人无法确定自身相对于上述先验地图的位姿。例如，手动将在指定区域中工作的移动机器人搬运至该指定区域的其他位置，导致该移动机器人定位丢失；又例如，移动机器人停机之后再次启动，发生定位丢失等。However, in many cases, the mobile robot will suffer from positioning loss, that is, the mobile robot cannot determine its position relative to the above-mentioned a priori map. For example, if a mobile robot working in a designated area is manually moved to another location in the designated area, the positioning of the mobile robot is lost; another example is if the mobile robot is stopped and then started again, positioning loss occurs.

基于此，如何对发生定位丢失的移动机器人进行重定位是当前亟待解决的问题，而所谓重定位是指在移动机器人发生定位丢失时，重新确定移动机器人相对于先验地图的位姿。Based on this, how to reposition a mobile robot that has lost its positioning is an urgent problem that needs to be solved. The so-called relocation refers to re-determining the position and posture of the mobile robot relative to the prior map when the mobile robot loses its positioning.

发明内容Contents of the invention

本申请实施例的目的在于提供一种重定位方法、装置及电子设备，以对发生定位丢失的移动机器人进行重定位。具体技术方案如下：The purpose of the embodiments of the present application is to provide a relocation method, device and electronic equipment to reposition a mobile robot that has lost its location. The specific technical solutions are as follows:

第一方面，本申请实施例提供了一种重定位方法，所述方法包括：In a first aspect, embodiments of the present application provide a relocation method, which method includes:

基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定所述目标机器人的初始参考位姿；Determine the initial reference pose of the target robot based on the first frame of images collected by the image acquisition device mounted on the target robot;

获取所述图像采集设备所采集的多帧参考图像，并基于所述初始参考位姿，确定所述目标机器人在采集每帧参考图像时的参考位姿，得到每帧参考图像对应的参考位姿；Obtain multiple frames of reference images collected by the image acquisition device, and based on the initial reference pose, determine the reference pose of the target robot when collecting each frame of reference image, and obtain the reference pose corresponding to each frame of reference image. ;

利用所述多帧参考图像和每帧参考图像对应的参考位姿，构建所述目标机器人当前所在的区域局部地图，并在预设的先验子地图中，确定与所述区域局部地图模板匹配的目标子地图；其中，所述先验子地图是利用预设的布局子地图对所述目标机器人所属空间的先验地图进行分割得到的，所述布局子地图是按照区域分割线对所述目标机器人所属空间的空间布局图进行分割得到的；Using the multi-frame reference images and the reference pose corresponding to each frame of reference image, construct a local map of the area where the target robot is currently located, and determine the match with the regional local map template in the preset a priori sub-map The target sub-map; wherein, the a priori sub-map is obtained by dividing the a priori map of the space to which the target robot belongs using a preset layout sub-map, and the layout sub-map is obtained by dividing the a priori map according to the area dividing line. It is obtained by segmenting the spatial layout of the space where the target robot belongs;

对所述图像采集设备所采集的当前图像与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像，并根据所述目标基准图像所关联的目标位姿，确定所述目标机器人相对于所述先验地图的重定位位姿。Perform similarity matching between the current image collected by the image acquisition device and each reference image associated with the target sub-map, determine the target reference image that matches the current image, and determine the target reference image based on the associated The target pose determines the relocation pose of the target robot relative to the a priori map.

可选的，一种具体实现方式中，所述图像采集设备为双目相机，所述基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定所述目标机器人的初始参考位姿，包括：Optionally, in a specific implementation, the image acquisition device is a binocular camera, and the initial reference pose of the target robot is determined based on the first frame image collected by the image acquisition device mounted on the target robot, include:

基于目标机器人所搭载图像采集设备所采集到第一帧同步的左目图像和右目图像，确定所述目标机器人的初始参考位姿。Based on the synchronized left eye image and right eye image of the first frame collected by the image acquisition equipment mounted on the target robot, the initial reference pose of the target robot is determined.

可选的，一种具体实现方式中，所述获取所述图像采集设备所采集的多帧参考图像，包括：Optionally, in a specific implementation, the obtaining of multiple frames of reference images collected by the image acquisition device includes:

获取所述图像采集设备所采集到的多个关键帧，作为多帧参考图像；Obtain multiple key frames collected by the image acquisition device as multi-frame reference images;

其中，所述多个关键帧包括：按照预设时间间隔所采集的多帧图像、按照所述目标机器人的预设移动距离所采集的多帧图像，或者，包括预设图像特征的多帧图像。Wherein, the plurality of key frames include: multiple frames of images collected according to a preset time interval, multiple frames of images collected according to a preset movement distance of the target robot, or multiple frames of images including preset image features. .

可选的，一种具体实现方式中，所述先验子地图的构建方式，包括：Optionally, in a specific implementation, the construction method of the prior submap includes:

获取所述先验地图和所述布局子地图；Obtain the a priori map and the layout submap;

在所述先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的先验子地图。In the a priori map, template matching is performed on each layout submap to obtain a priori submap matched by the layout submap.

可选的，一种具体实现方式中，所述在所述先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的先验子地图，包括：Optionally, in a specific implementation, in the a priori map, template matching is performed on each layout submap to obtain a priori submap matched by the layout submap, including:

在所述先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的初始子地图；In the a priori map, template matching is performed on each layout submap to obtain an initial submap matched by the layout submap;

获取每个布局子地图对应的连接关系和每个初始子地图对应的可通行关系，其中，每个布局子地图对应的连接关系为：每个布局子地图所表征区域与该布局子地图所表征区域的可通行通道的连接关系，每个初始子地图对应的可通行关系为：基于构建所述先验地图时，所述目标机器人的移动轨迹确定的，每个初始子地图与相邻初始子地图的可通行关系；Obtain the connection relationship corresponding to each layout submap and the passability relationship corresponding to each initial submap, where the connection relationship corresponding to each layout submap is: the area represented by each layout submap and the area represented by the layout submap The connection relationship of the passable channels in the area. The passability relationship corresponding to each initial sub-map is: based on the movement trajectory of the target robot when constructing the a priori map, each initial sub-map is connected to the adjacent initial sub-map. The accessibility relationship of the map;

针对每个布局子地图，利用该布局子地图对应的连接关系以及该布局子地图对应的初始子地图对应的可通行关系，对该布局子地图对应的初始子地图进行校正，得到该布局子地图所匹配的先验子地图。For each layout submap, use the connection relationship corresponding to the layout submap and the passability relationship corresponding to the initial submap corresponding to the layout submap to correct the initial submap corresponding to the layout submap to obtain the layout submap The matched prior submap.

可选的，一种具体实现方式中，所述在预设的先验子地图中，确定与所述区域局部地图模板匹配的目标子地图，包括：Optionally, in a specific implementation, determining the target submap that matches the regional local map template in the preset a priori submap includes:

针对每个先验子地图，对该先验子地图和所述区域局部地图进行模板匹配，并对该先验子地图所匹配的布局子地图和所述区域局部地图进行模板匹配，得到匹配结果；For each prior sub-map, template matching is performed on the prior sub-map and the regional local map, and template matching is performed on the layout sub-map matched by the prior sub-map and the regional local map to obtain a matching result. ;

基于所述匹配结果，在预设的先验子地图中，确定目标子地图。Based on the matching result, the target submap is determined in the preset a priori submap.

可选的，一种具体实现方式中，所述对所述图像采集设备所采集的当前图像与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像，包括：Optionally, in a specific implementation, the current image collected by the image acquisition device is matched with each reference image associated with the target sub-map, and the image that matches the current image is determined. Target baseline images, including:

对所述图像采集设备所采集的最近的关键帧与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像。Similarity matching is performed between the most recent key frame collected by the image acquisition device and each reference image associated with the target sub-map to determine the target reference image that matches the current image.

可选的，一种具体实现方式中，每个基准图像所关联的位姿为：所述图像采集设备采集该基准图像时，所述图像采集设备相对于所述先验地图的位姿；Optionally, in a specific implementation manner, the pose associated with each reference image is: the pose of the image capture device relative to the a priori map when the image capture device captures the reference image;

所述根据所述目标基准图像所关联的目标位姿，确定所述目标机器人相对于所述先验地图的重定位位姿，包括：Determining the relocation pose of the target robot relative to the a priori map based on the target pose associated with the target reference image includes:

根据所述图像采集设备与所述目标机器人的相对位姿，对所述目标基准图像所关联的目标位姿进行位姿转换，得到所述目标机器人相对于所述先验地图的重定位位姿。According to the relative pose of the image acquisition device and the target robot, pose conversion is performed on the target pose associated with the target reference image to obtain the repositioned pose of the target robot relative to the a priori map. .

第二方面，本申请实施例提供了一种机器人，所述机器人包括：In a second aspect, embodiments of the present application provide a robot, which includes:

图像采集设备，用于采集关于所述机器人所在区域的图像；Image acquisition equipment, used to collect images of the area where the robot is located;

处理器，用于基于所述图像采集设备所采集的图像，执行本申请实施例的第一方面任一所述的重定位方法实施例的步骤。A processor, configured to perform the steps of the relocation method embodiment described in any one of the first aspects of the embodiments of this application based on the images collected by the image acquisition device.

第三方面，本申请实施例提供了一种重定位装置，所述装置包括：In a third aspect, embodiments of the present application provide a relocation device, which includes:

初始位姿确定模块，用于基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定所述目标机器人的初始参考位姿；An initial pose determination module is used to determine the initial reference pose of the target robot based on the first frame of images collected by the image acquisition device mounted on the target robot;

参考位姿确定模块，用于获取所述图像采集设备所采集的多帧参考图像，并基于所述初始参考位姿，确定所述目标机器人在采集每帧参考图像时的参考位姿，得到每帧参考图像对应的参考位姿；A reference pose determination module is used to obtain multiple frames of reference images collected by the image acquisition device, and based on the initial reference pose, determine the reference pose of the target robot when collecting each frame of reference image, and obtain each frame of reference image. The reference pose corresponding to the frame reference image;

局部地图构建模块，用于利用所述多帧参考图像和每帧参考图像对应的参考位姿，构建所述目标机器人当前所在的区域局部地图，并在预设的先验子地图中，确定与所述区域局部地图模板匹配的目标子地图；其中，所述先验子地图是利用预设的布局子地图对所述目标机器人所属空间的先验地图进行分割得到的，所述布局子地图是按照区域分割线对所述目标机器人所属空间的空间布局图进行分割得到的；A local map construction module for constructing a local map of the area where the target robot is currently located using the multi-frame reference image and the reference pose corresponding to each frame of reference image, and in the preset a priori sub-map, determine the The target sub-map matched by the regional local map template; wherein the prior sub-map is obtained by segmenting the prior map of the space to which the target robot belongs using a preset layout sub-map, and the layout sub-map is Obtained by dividing the spatial layout diagram of the space where the target robot belongs according to the area dividing lines;

重定位模块，用于对所述图像采集设备所采集的当前图像与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像，并根据所述目标基准图像所关联的目标位姿，确定所述目标机器人相对于所述先验地图的重定位位姿。A relocation module, configured to perform similarity matching between the current image collected by the image acquisition device and each reference image associated with the target sub-map, determine the target reference image that matches the current image, and determine the target reference image based on the The target pose associated with the target reference image is used to determine the relocation pose of the target robot relative to the a priori map.

可选的，一种具体实现方式中，所述图像采集设备为双目相机，所述初始位姿确定模块，具体用于：Optionally, in a specific implementation, the image acquisition device is a binocular camera, and the initial pose determination module is specifically used for:

基于目标机器人所搭载图像采集设备所采集到第一帧同步的左目图像和右目图像，确定所述目标机器人的初始参考位姿。Based on the synchronized left eye image and right eye image of the first frame collected by the image acquisition equipment mounted on the target robot, the initial reference pose of the target robot is determined.

可选的，一种具体实现方式中，所述参考位姿确定模块，具体用于：Optionally, in a specific implementation, the reference pose determination module is specifically used for:

获取所述图像采集设备所采集到的多个关键帧，作为多帧参考图像；Obtain multiple key frames collected by the image acquisition device as multi-frame reference images;

其中，所述多个关键帧包括：按照预设时间间隔所采集的多帧图像、按照所述目标机器人的预设移动距离所采集的多帧图像，或者，包括预设图像特征的多帧图像。Wherein, the plurality of key frames include: multiple frames of images collected according to a preset time interval, multiple frames of images collected according to a preset movement distance of the target robot, or multiple frames of images including preset image features. .

可选的，一种具体实现方式中，所述装置还包括先验子地图构建模块，所述先验子地图构建模块，包括：Optionally, in a specific implementation manner, the device further includes a priori sub-map building module, and the prior sub-map building module includes:

获取子模块，用于获取所述先验地图和所述布局子地图；Obtain sub-module, used to obtain the a priori map and the layout sub-map;

先验子地图构建子模块，用于在所述先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的先验子地图。The a priori submap construction submodule is used to perform template matching on each layout submap in the a priori map to obtain the a priori submap matched by the layout submap.

可选的，一种具体实现方式中，所述先验子地图构建子模块，具体用于：Optionally, in a specific implementation manner, the a priori submap constructs a submodule, specifically used for:

在所述先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的初始子地图；In the a priori map, template matching is performed on each layout submap to obtain an initial submap matched by the layout submap;

获取每个布局子地图对应的连接关系和每个初始子地图对应的可通行关系，其中，每个布局子地图对应的连接关系为：每个布局子地图所表征区域与该布局子地图所表征区域的可通行通道的连接关系，每个初始子地图对应的可通行关系为：基于构建所述先验地图时，所述目标机器人的移动轨迹确定的，每个初始子地图与相邻初始子地图的可通行关系；Obtain the connection relationship corresponding to each layout submap and the passability relationship corresponding to each initial submap, where the connection relationship corresponding to each layout submap is: the area represented by each layout submap and the area represented by the layout submap The connection relationship of the passable channels in the area. The passability relationship corresponding to each initial sub-map is: based on the movement trajectory of the target robot when constructing the a priori map, each initial sub-map is connected to the adjacent initial sub-map. The accessibility relationship of the map;

针对每个布局子地图，利用该布局子地图对应的连接关系以及该布局子地图对应的初始子地图对应的可通行关系，对该布局子地图对应的初始子地图进行校正，得到该布局子地图所匹配的先验子地图。For each layout submap, use the connection relationship corresponding to the layout submap and the passability relationship corresponding to the initial submap corresponding to the layout submap to correct the initial submap corresponding to the layout submap to obtain the layout submap The matched prior submap.

可选的，一种具体实现方式中，所述局部地图构建模块，具体用于：Optionally, in a specific implementation, the local map building module is specifically used for:

针对每个先验子地图，对该先验子地图和所述区域局部地图进行模板匹配，并对该先验子地图所匹配的布局子地图和所述区域局部地图进行模板匹配，得到匹配结果；For each prior sub-map, template matching is performed on the prior sub-map and the regional local map, and template matching is performed on the layout sub-map matched by the prior sub-map and the regional local map to obtain a matching result. ;

基于所述匹配结果，在预设的先验子地图中，确定目标子地图。Based on the matching result, the target submap is determined in the preset a priori submap.

可选的，一种具体实现方式中，所述重定位模块，具体用于：Optionally, in a specific implementation manner, the relocation module is specifically used for:

对所述图像采集设备所采集的最近的关键帧与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像。Similarity matching is performed between the most recent key frame collected by the image acquisition device and each reference image associated with the target sub-map to determine the target reference image that matches the current image.

可选的，一种具体实现方式中，每个基准图像所关联的位姿为：所述图像采集设备采集该基准图像时，所述图像采集设备相对于所述先验地图的位姿；Optionally, in a specific implementation manner, the pose associated with each reference image is: the pose of the image capture device relative to the a priori map when the image capture device captures the reference image;

所述重定位模块，具体用于：The relocation module is specifically used for:

根据所述图像采集设备与所述目标机器人的相对位姿，对所述目标基准图像所关联的目标位姿进行位姿转换，得到所述目标机器人相对于所述先验地图的重定位位姿。According to the relative pose of the image acquisition device and the target robot, pose conversion is performed on the target pose associated with the target reference image to obtain the repositioned pose of the target robot relative to the a priori map. .

第四方面，本申请实施例提供了一种电子设备，包括：In a fourth aspect, embodiments of the present application provide an electronic device, including:

存储器，用于存放计算机程序；Memory, used to store computer programs;

处理器，用于执行存储器上所存放的程序时，实现上述任一方法实施例的步骤。The processor is used to implement the steps of any of the above method embodiments when executing a program stored in the memory.

第五方面，本申请实施例提供了一种计算机可读存储介质，所述计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现上述任一方法实施例的步骤。In a fifth aspect, embodiments of the present application provide a computer-readable storage medium. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of any of the above method embodiments are implemented.

第六方面，本申请实施例还提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述任一方法实施例的步骤。In a sixth aspect, embodiments of the present application also provide a computer program product containing instructions that, when run on a computer, cause the computer to perform the steps of any of the above method embodiments.

本申请实施例有益效果：Beneficial effects of the embodiments of this application:

以上可见，应用本申请实施例提供的方案，为了对目标机器人进行重定位，可以预先按照区域分割线对目标机器人所属空间的空间布局图进行分割，得到布局子地图，之后，利用上述布局子地图对目标机器人所属空间的先验地图进行分割，得到先验子地图。这样，在对目标机器人进行重定位时，可以首先基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定目标机器人的初始参考位姿，然后，可以获取图像采集设备所采集的多帧参考图像，并基于上述初始参考位姿，确定目标机器人在采集每帧参考图像时的参考位姿，得到每帧参考图像对应的参考位姿；接着，可以利用多帧参考图像和每帧参考图像对应的参考位姿，构建目标机器人当前所在的区域局部地图，并在预设的先验子地图中，确定与区域局部地图模板匹配的目标子地图；进而，可以对图像采集设备所采集的当前图像与目标子地图所关联的各个基准图像进行相似度匹配，确定与当前图像所匹配的目标基准图像，从而，可以根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿。It can be seen from the above that by applying the solution provided by the embodiment of the present application, in order to reposition the target robot, the spatial layout map of the space where the target robot belongs can be divided in advance according to the area dividing lines to obtain a layout sub-map. After that, the above layout sub-map is used Segment the prior map of the space where the target robot belongs to obtain the prior sub-map. In this way, when relocating the target robot, the initial reference pose of the target robot can be determined based on the first frame of images collected by the image acquisition device mounted on the target robot, and then, multiple frames collected by the image acquisition device can be obtained. Reference image, and based on the above initial reference pose, determine the reference pose of the target robot when collecting each frame of reference image, and obtain the reference pose corresponding to each frame of reference image; then, you can use multiple frame reference images and each frame reference image The corresponding reference pose is used to construct a local map of the area where the target robot is currently located, and in the preset a priori sub-map, the target sub-map that matches the regional local map template is determined; furthermore, the current image collected by the image acquisition device can be The image is matched with each reference image associated with the target sub-map to determine the target reference image that matches the current image. Therefore, the position of the target robot relative to the prior map can be determined based on the target posture associated with the target reference image. Reposition pose.

基于此，应用本申请实施例提供的方案，可以实现对目标机器人的重定位。并且，通过区域局部地图与先验子地图的模板匹配可以提高区域定位的准确性，从而，可以降低因区域误定位而导致的重定位失败的可能性，提高重定位的准确性；通过当前图像与目标子地图的各个基准图像的相似度匹配，可以进一步的提高重定位的准确性。此外，仅需将当前图像与目标子地图的各个基准图像进行相似度匹配，而无需遍历完整的先验地图所关联的全部基准图像进行相似度匹配，可以减少重定位的计算量，进而，提高重定位效率。Based on this, by applying the solutions provided by the embodiments of this application, the target robot can be repositioned. Moreover, the accuracy of regional positioning can be improved through template matching between the regional local map and the prior submap, thereby reducing the possibility of relocation failure caused by regional mislocalization and improving the accuracy of relocation; through the current image Similarity matching with each reference image of the target submap can further improve the accuracy of relocation. In addition, it is only necessary to perform similarity matching between the current image and each reference image of the target submap, without traversing all reference images associated with the complete prior map for similarity matching, which can reduce the calculation amount of relocation, thereby improving Relocation efficiency.

当然，实施本申请的任一产品或方法并不一定需要同时达到以上所述的所有优点。Of course, implementing any product or method of the present application does not necessarily require achieving all the above-mentioned advantages simultaneously.

附图说明Description of the drawings

为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，还可以根据这些附图获得其他的实施例。In order to explain the embodiments of the present application or the technical solutions in the prior art more clearly, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only These are some embodiments of the present application. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings.

图1为本申请实施例提供的一种移动机器人的示意图；Figure 1 is a schematic diagram of a mobile robot provided by an embodiment of the present application;

图2为本申请实施例提供的一种先验子地图的构建方式的流程示意图；Figure 2 is a schematic flowchart of a method for constructing a priori submaps provided by an embodiment of the present application;

图3为本申请实施例提供的一种重定位方法的流程示意图；Figure 3 is a schematic flowchart of a relocation method provided by an embodiment of the present application;

图4为本申请实施例提供的重定位方法的一种具体实例的流程示意图；Figure 4 is a schematic flowchart of a specific example of the relocation method provided by the embodiment of the present application;

图5为本申请实施例提供的一种机器人的结构示意图；Figure 5 is a schematic structural diagram of a robot provided by an embodiment of the present application;

图6为本申请实施例提供的一种重定位装置的结构示意图；Figure 6 is a schematic structural diagram of a relocation device provided by an embodiment of the present application;

图7为本申请实施例提供的一种电子设备的结构示意图。FIG. 7 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

具体实施方式Detailed ways

下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员基于本申请所获得的所有其他实施例，都属于本申请保护的范围。The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art based on this application fall within the scope of protection of this application.

在很多情况下，移动机器人会发生定位丢失的现象，即移动机器人无法确定自身相对于上述先验地图的位姿。基于此，如何对发生定位丢失的移动机器人进行重定位是当前亟待解决的问题，而所谓重定位是指在移动机器人发生定位丢失时，重新确定移动机器人相对于先验地图的位姿。In many cases, the mobile robot will lose its positioning, that is, the mobile robot cannot determine its position relative to the above-mentioned a priori map. Based on this, how to reposition a mobile robot that has lost its positioning is an urgent problem that needs to be solved. The so-called relocation refers to re-determining the position and posture of the mobile robot relative to the prior map when the mobile robot loses its positioning.

为了解决上述技术问题，本申请实施例提供了一种重定位方法。In order to solve the above technical problems, embodiments of the present application provide a relocation method.

其中，该方法可以适用于需要对移动机器人进行重定位的各种应用场景，例如，对在工作过程中，被搬运至其他位置的移动机器人进行重定位；对停机之后再次启动的清洁机器人进行重定位等。Among them, this method can be applied to various application scenarios that require relocation of mobile robots. For example, relocation of mobile robots that are transported to other locations during the work process; relocation of cleaning robots that are started again after shutdown. Positioning etc.

并且，该方法可以应用于搭载有图像采集设备的移动机器人自身，例如，该移动机器人搭载有数据处理模块，在利用自身所搭载的图像采集设备采集图像后，利用自身所搭载的数据处理模块执行该方法；也可以应用于搭载于目标机器人上且具有数据处理功能的图像采集设备，例如，该图像采集设备搭载有数据处理模块，在采集图像后，利用自身所搭载的数据处理模块执行该方法；还可以应用于能够与搭载有图像采集设备的移动机器人进行通信，并为移动机器人提供重定位服务的各类电子设备，例如，管理平台等；并且，在本方法的执行主体为电子设备时，该电子设备可以是独立的电子设备，也可以是由多台电子设备构成的设备集群，以下简称电子设备。Moreover, this method can be applied to a mobile robot itself equipped with an image acquisition device. For example, the mobile robot is equipped with a data processing module. After collecting images with the image acquisition device it is equipped with, the mobile robot is executed using the data processing module it is equipped with. This method can also be applied to an image acquisition device mounted on the target robot and having a data processing function. For example, the image acquisition device is equipped with a data processing module. After acquiring the image, it uses its own data processing module to execute the method. ; It can also be applied to various electronic devices that can communicate with mobile robots equipped with image acquisition equipment and provide relocation services for mobile robots, such as management platforms, etc.; and, when the execution subject of this method is an electronic device , the electronic device may be an independent electronic device or a device cluster composed of multiple electronic devices, hereinafter referred to as electronic devices.

基于此，本申请实施例不对该方法的应用场景和执行主体进行具体限定。Based on this, the embodiments of this application do not specifically limit the application scenarios and execution subjects of this method.

本申请实施例提供的一种重定位方法，可以包括如下步骤：A relocation method provided by the embodiment of the present application may include the following steps:

基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定所述目标机器人的初始参考位姿；Determine the initial reference pose of the target robot based on the first frame of images collected by the image acquisition device mounted on the target robot;

获取所述图像采集设备所采集的多帧参考图像，并基于所述初始参考位姿，确定所述目标机器人在采集每帧参考图像时的参考位姿，得到每帧参考图像对应的参考位姿；Obtain multiple frames of reference images collected by the image acquisition device, and based on the initial reference pose, determine the reference pose of the target robot when collecting each frame of reference image, and obtain the reference pose corresponding to each frame of reference image. ;

利用所述多帧参考图像和每帧参考图像对应的参考位姿，构建所述目标机器人当前所在的区域局部地图，并在预设的先验子地图中，确定与所述区域局部地图模板匹配的目标子地图；其中，所述先验子地图是利用预设的布局子地图对所述目标机器人所属空间的先验地图进行分割得到的，所述布局子地图是按照区域分割线对所述目标机器人所属空间的空间布局图进行分割得到的；Using the multi-frame reference images and the reference pose corresponding to each frame of reference image, construct a local map of the area where the target robot is currently located, and determine the match with the regional local map template in the preset a priori sub-map The target sub-map; wherein, the a priori sub-map is obtained by dividing the a priori map of the space to which the target robot belongs using a preset layout sub-map, and the layout sub-map is obtained by dividing the a priori map according to the area dividing line. It is obtained by segmenting the spatial layout of the space where the target robot belongs;

对所述图像采集设备所采集的当前图像与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像，并根据所述目标基准图像所关联的目标位姿，确定所述目标机器人相对于所述先验地图的重定位位姿。Perform similarity matching between the current image collected by the image acquisition device and each reference image associated with the target sub-map, determine the target reference image that matches the current image, and determine the target reference image based on the associated The target pose determines the relocation pose of the target robot relative to the a priori map.

以上可见，应用本申请实施例提供的方案，为了对目标机器人进行重定位，可以预先按照区域分割线对目标机器人所属空间的空间布局图进行分割，得到布局子地图，之后，利用上述布局子地图对目标机器人所属空间的先验地图进行分割，得到先验子地图。这样，在对目标机器人进行重定位时，可以首先基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定目标机器人的初始参考位姿，然后，可以获取图像采集设备所采集的多帧参考图像，并基于上述初始参考位姿，确定目标机器人在采集每帧参考图像时的参考位姿，得到每帧参考图像对应的参考位姿；接着，可以利用多帧参考图像和每帧参考图像对应的参考位姿，构建目标机器人当前所在的区域局部地图，并在预设的先验子地图中，确定与区域局部地图模板匹配的目标子地图；进而，可以对图像采集设备所采集的当前图像与目标子地图所关联的各个基准图像进行相似度匹配，确定与当前图像所匹配的目标基准图像，从而，可以根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿。It can be seen from the above that by applying the solution provided by the embodiment of the present application, in order to reposition the target robot, the spatial layout map of the space where the target robot belongs can be divided in advance according to the area dividing lines to obtain a layout sub-map. After that, the above layout sub-map is used Segment the prior map of the space where the target robot belongs to obtain the prior sub-map. In this way, when relocating the target robot, the initial reference pose of the target robot can be determined based on the first frame of images collected by the image acquisition device mounted on the target robot, and then, multiple frames collected by the image acquisition device can be obtained. Reference image, and based on the above initial reference pose, determine the reference pose of the target robot when collecting each frame of reference image, and obtain the reference pose corresponding to each frame of reference image; then, you can use multiple frame reference images and each frame reference image The corresponding reference pose is used to construct a local map of the area where the target robot is currently located, and in the preset a priori sub-map, the target sub-map that matches the regional local map template is determined; furthermore, the current image collected by the image acquisition device can be The image is matched with each reference image associated with the target sub-map to determine the target reference image that matches the current image. Therefore, the position of the target robot relative to the prior map can be determined based on the target posture associated with the target reference image. Reposition pose.

基于此，应用本申请实施例提供的方案，可以实现对目标机器人的重定位。并且，通过区域局部地图与先验子地图的模板匹配可以提高区域定位的准确性，从而，可以降低因区域误定位而导致的重定位失败的可能性，提高重定位的准确性；通过当前图像与目标子地图的各个基准图像的相似度匹配，可以进一步的提高重定位的准确性。此外，仅需将当前图像与目标子地图的各个基准图像进行相似度匹配，而无需遍历完整的先验地图所关联的全部基准图像进行相似度匹配，可以减少重定位的计算量，进而，提高重定位效率。Based on this, by applying the solutions provided by the embodiments of this application, the target robot can be repositioned. Moreover, the accuracy of regional positioning can be improved through template matching between the regional local map and the prior submap, thereby reducing the possibility of relocation failure caused by regional mislocalization and improving the accuracy of relocation; through the current image Similarity matching with each reference image of the target submap can further improve the accuracy of relocation. In addition, it is only necessary to perform similarity matching between the current image and each reference image of the target submap, without traversing all reference images associated with the complete prior map for similarity matching, which can reduce the calculation amount of relocation, thereby improving Relocation efficiency.

下面为了便于理解本申请实施例所提供的一种重定位方法，首先对本申请实施例中涉及到的先验地图、空间布局图、先验子地图进行说明。In order to facilitate understanding of a relocation method provided by the embodiment of the present application, the a priori map, spatial layout map, and a priori submap involved in the embodiment of the present application will be described below.

通常，移动机器人在首次进入某个区域时，可以通过对该区域进行探索，构建该区域的先验地图。Usually, when a mobile robot enters a certain area for the first time, it can construct a priori map of the area by exploring the area.

其中，上述构建先验地图的过程，可以称为SLAM(Simultaneous LocalizationAnd Mapping，即时定位与地图构建)过程。也就是说，移动机器人在首次进入某个区域时，可以将当前所在位置作为在该区域中的初始位置，之后，移动机器人可以从该初始位置出发，在该区域中移动进行区域探索。在进行区域探索的过程中，移动机器人可以利用自身所搭载的图像采集设备采集关于该区域的多帧图像，并且，针对每帧图像，可以确定图像采集设备在采集该帧图像时，目标机器人的位姿，作为该帧图像对应的位姿。Among them, the above-mentioned process of constructing a priori maps can be called the SLAM (Simultaneous Localization And Mapping, instant positioning and map construction) process. That is to say, when a mobile robot enters a certain area for the first time, it can use its current location as its initial position in the area. After that, the mobile robot can start from this initial position and move in the area to explore the area. During the process of area exploration, the mobile robot can use its own image acquisition device to collect multiple frames of images about the area, and for each frame of image, it can be determined that the image acquisition device collects the image of the target robot. pose, as the pose corresponding to this frame of image.

然后，在对该区域完成区域探索后，可以利用图像采集设备所采集的多帧图像以及每帧图像对应的位姿，构建关于该区域的探索地图，作为该区域的先验地图。Then, after completing the regional exploration of the area, you can use the multiple frames of images collected by the image acquisition device and the pose corresponding to each frame of image to construct an exploration map of the area as a priori map of the area.

其中，上述移动机器人所搭载的图像采集设备可以是单目相机，也可以是双目相机，这都是合理的，在本申请实施例中不做具体限定。The image acquisition device mounted on the mobile robot may be a monocular camera or a binocular camera, which are both reasonable and are not specifically limited in the embodiments of this application.

此外，由于上述先验地图是根据多帧图像以及每帧图像对应的位姿所构建的，从而，该先验地图的每个地图区域，均关联有多帧图像，并且，每帧图像均关联有移动机器人的位姿。In addition, since the above-mentioned prior map is constructed based on multiple frame images and the pose corresponding to each frame image, each map area of the prior map is associated with multiple frame images, and each frame image is associated with There are poses of mobile robots.

其中，所谓每帧图像所关联的移动机器人的位姿是指：在移动机器人所搭载的图像采集设备采集该帧图像时，该移动机器人的位姿。The so-called pose of the mobile robot associated with each frame of image refers to the pose of the mobile robot when the image acquisition device mounted on the mobile robot collects the image of the frame.

并且，所谓移动机器人的位姿是指该移动机器人的位置和姿态。其中，移动机器人的位置是指移动机器人在区域中的具体位置，而移动机器人的姿态是指移动机器人的朝向。Furthermore, the posture of the mobile robot refers to the position and posture of the mobile robot. Among them, the position of the mobile robot refers to the specific position of the mobile robot in the area, and the posture of the mobile robot refers to the orientation of the mobile robot.

由于每帧图像均是由图像采集设备所采集的，从而，基于每帧图像可以确定图像采集设备的位姿，又由于图像采集设备是搭载在移动机器人上的，从而，图像采集设备的位置和移动机器人的位置是相同的，并且，该图像采集设备的姿态和移动机器人的姿态可以通过该图像采集设备的安装位置确定，这样，上述图像采集设备的位姿和移动机器人的位姿具有相关性，即上述图像采集设备和移动机器人具有相对位姿。Since each frame of image is collected by the image acquisition device, the pose of the image acquisition device can be determined based on each frame of image. And since the image acquisition device is mounted on the mobile robot, the position and orientation of the image acquisition device can be determined based on the image acquisition device. The position of the mobile robot is the same, and the attitude of the image acquisition device and the attitude of the mobile robot can be determined by the installation position of the image acquisition device. In this way, the attitude of the image acquisition device and the attitude of the mobile robot are relevant. , that is, the above image acquisition device and the mobile robot have relative postures.

示例性的，如图1所示，在以移动机器人的移动方向为该移动机器人的前方时，安装在该移动机器人上的双目相机，可以对该移动机器人的后方区域进行图像采集。其中，该双目相机的采集视角与水平面的夹角为α，且该双目相机的垂直视场角为β。也就是说，在移动机器人在指定区域的地面上移动时，安装在该移动机器人上的双目相机可以以垂直视场角β对位于移动机器人后方的相机垂直视场范围内的墙体、门窗等进行图像采集。For example, as shown in Figure 1, when the moving direction of the mobile robot is in front of the mobile robot, a binocular camera installed on the mobile robot can collect images of the rear area of the mobile robot. Wherein, the angle between the collection angle of the binocular camera and the horizontal plane is α, and the vertical field of view angle of the binocular camera is β. That is to say, when the mobile robot moves on the ground in a designated area, the binocular camera installed on the mobile robot can view the walls, doors and windows within the vertical field of view of the camera behind the mobile robot at a vertical field of view β. Wait for image collection.

针对每个区域，该区域内可以包括各类障碍物，例如，墙壁、家具、人物、动物等。从而，移动机器人对该区域进行区域探索所得到的先验地图中可以包括该区域中的各类障碍物。For each area, the area can include various types of obstacles, such as walls, furniture, people, animals, etc. Therefore, the prior map obtained by the mobile robot's regional exploration of the area can include various obstacles in the area.

在构建上述先验地图后，移动机器人可以存储上述先验地图。这样，移动机器人在该区域内移动时，可以实时确定自身相对于上述先验地图中的位姿，即利用上述先验地图，确定自身在该区域中的位姿以及所在位置周围的障碍物信息。After constructing the above-mentioned prior map, the mobile robot can store the above-mentioned prior map. In this way, when the mobile robot moves in the area, it can determine its own position relative to the above-mentioned a priori map in real time, that is, use the above-mentioned a priori map to determine its own position and posture in the area and the obstacle information around the location. .

然后，在移动机器人发生定位丢失时，即该移动机器人无法确定自身相对于上述先验地图的位姿时，便需要对该移动机器人进行重定位，以重新找回移动机器人在该先验地图中的位姿。这样，在重定位自身相对于先验地图的位姿后，该移动机器人便可以继续利用上述先验地图，确定自身在该区域中的位姿以及所在位置周围的障碍物信息，进而，利用上述先验地图增强该移动机器人的避障能力和定位鲁棒性。Then, when the mobile robot loses its positioning, that is, when the mobile robot cannot determine its position relative to the above-mentioned a priori map, the mobile robot needs to be repositioned to regain the position of the mobile robot in the a priori map. posture. In this way, after relocating its posture relative to the prior map, the mobile robot can continue to use the above-mentioned prior map to determine its own posture in the area and the obstacle information around the location, and then use the above-mentioned The prior map enhances the obstacle avoidance ability and positioning robustness of the mobile robot.

其中，由于先验地图是基于移动机器人所搭载的图像采集设备所采集的多帧图像以及每帧图像所对应的位姿构建的，而在图像采集设备的采集频率、图像采集设备的采集范围以及移动机器人自身的形状尺寸等因素的影响下，所构建的先验地图可能是不完整的。Among them, since the prior map is constructed based on multiple frames of images collected by the image acquisition equipment mounted on the mobile robot and the pose corresponding to each frame of image, the collection frequency of the image acquisition equipment, the collection range of the image acquisition equipment, and Under the influence of factors such as the shape and size of the mobile robot itself, the constructed prior map may be incomplete.

例如，移动机器人由于自身形状限制不能到达该区域中的部分位置，从而，该移动机器人所搭载的图像采集设备便无法采集到该区域中各个位置的图像，这样，利用图像采集设备所采集的关于该区域的多帧图像所构建的关于该区域的先验地图是不完整的。For example, a mobile robot cannot reach some locations in the area due to its own shape limitations. Therefore, the image acquisition equipment mounted on the mobile robot cannot collect images of various locations in the area. In this way, the image acquisition equipment collected by the image acquisition equipment cannot collect images of various locations in the area. The a priori map of the area constructed from multiple frames of images of the area is incomplete.

基于此，为了提高移动机器人重定位的定位准确性，可以利用上述区域的空间布局图对上述先验地图进行丰富和补充，进而，得到较为完整的先验地图。之后，可以利用该较为完整的先验地图，对移动机器人进行重定位。Based on this, in order to improve the positioning accuracy of mobile robot relocation, the spatial layout map of the above area can be used to enrich and supplement the above-mentioned prior map, and then obtain a more complete prior map. Afterwards, this relatively complete prior map can be used to reposition the mobile robot.

其中，上述区域的空间布局图是指用于表征该区域的空间布局的图像，该空间布局图中可以包括该区域内的墙壁、门窗等空间布局结构。并且，可以将上述墙壁、门窗可以作为该区域的区域分割线，并利用该区域分割线将该区域划分为多个子区域。相应的，可以将上述空间布局图按照该空间布局图中的区域分割线进行划分，得到多个布局子地图。The spatial layout diagram of the above-mentioned area refers to an image used to represent the spatial layout of the area. The spatial layout diagram may include spatial layout structures such as walls, doors and windows in the area. Moreover, the above-mentioned walls, doors and windows can be used as area dividing lines of the area, and the area dividing lines can be used to divide the area into multiple sub-areas. Correspondingly, the above-mentioned spatial layout map can be divided according to the area dividing lines in the spatial layout map to obtain multiple layout sub-maps.

示例性的，在利用清洁机器人对房屋进行清洁时，该房屋的空间布局图可以是该房屋的户型图，在该户型图中标注有该房屋内的各个墙壁和各个门窗的位置，并且，将上述各个墙壁和各个门窗作为区域分割线，可以将上述房屋划分为客厅、卧室等不同子区域。For example, when a cleaning robot is used to clean a house, the spatial layout diagram of the house may be a floor plan of the house, in which the positions of each wall and door and window in the house are marked, and, Each of the above-mentioned walls and each door and window serve as area dividing lines, and the above-mentioned house can be divided into different sub-areas such as living rooms and bedrooms.

并且，上述空间布局图中还可以包括区域间的可通行通道，例如，房屋内的门框等；基于此，上述空间布局图还可以包括每个布局子地图所表征区域与该布局子地图所表征区域的可通行通道的连接关系，，作为该布局子地图的连接关系，例如，上述空间布局图可以包括布局子地图S所表征的厨房与该厨房中的门框的连接关系。Moreover, the above-mentioned spatial layout map may also include passable passages between areas, such as door frames in houses, etc. Based on this, the above-mentioned spatial layout map may also include the areas represented by each layout sub-map and the areas represented by the layout sub-map. The connection relationship of the traversable channels of the area is used as the connection relationship of the layout sub-map. For example, the above-mentioned spatial layout map may include the connection relationship between the kitchen represented by the layout sub-map S and the door frame in the kitchen.

并且，由于上述可通行通道可以连接两个不同的区域，例如，房间A与房间B可以通过门C连接，从而，可选的，针对每个布局子地图的连接关系还可以用于表征该布局子地图与另一个布局子地图之间通过该布局子地图所表征区域的可通行通道建立的连接关系，即包括该可通行通道的两个区域各自对应的布局子地图之间的连接关系，例如：布局子地图D与布局子地图E通过可通行通道F建立的连接关系等。Moreover, since the above-mentioned accessible passage can connect two different areas, for example, room A and room B can be connected through door C, thus, optionally, the connection relationship for each layout submap can also be used to characterize the layout. The connection relationship between a submap and another layout submap is established through the accessible passage of the area represented by the layout submap, that is, the connection relationship between the corresponding layout submaps of the two areas including the accessible passage, for example : The connection relationship established between layout submap D and layout submap E through the accessible channel F, etc.

可选的，两个不同局部子地图之间的连接关系也可以表征连接这两个局部子地图的可通行通道。例如，第j个布局子地图与第j+1个布局子地图之间通过门M连接，从而，可以将第j个布局子地图与第j+1个布局子地图之间的连接关系表示为L(j，j+1)。这样，L(j，j+1)既可以表征第j个布局子地图与第j+1个布局子地图之间存在连接关系，又可以表征连接第j个布局子地图与第j+1个布局子地图的门M。Optionally, the connection relationship between two different local sub-maps can also represent the passable channel connecting the two local sub-maps. For example, the jth layout submap and the j+1th layout submap are connected through the gate M. Therefore, the connection relationship between the jth layout submap and the j+1th layout submap can be expressed as L(j,j+1). In this way, L(j,j+1) can not only represent the connection relationship between the jth layout submap and the j+1th layout submap, but also can represent the connection between the jth layout submap and the j+1th layout submap. Layout the door M of the submap.

此外，可选的，在获取上述先验地区后，可以将上述先验地图处理为平面栅格地图；并且，由于上述空间布局图通常是三维地图，从而，在获取上述空间布局图后，可以对上述空间布局图进行识别，得到该空间布局图的平面图。之后，便可以利用该空间布局图的平面图对上述平面栅格地图进行补充。In addition, optionally, after obtaining the above-mentioned a priori region, the above-mentioned a priori map can be processed into a flat raster map; and, since the above-mentioned spatial layout map is usually a three-dimensional map, after obtaining the above-mentioned spatial layout map, the above-mentioned spatial layout map can be The above-mentioned spatial layout diagram is identified to obtain the floor plan of the spatial layout diagram. The flat grid map can then be supplemented with a plan view of the spatial layout.

通常，在某一区域内，对移动机器人进行重定位是通过遍历该区域所包括的多帧图像数据实现的，并且，整体区域所包括的图像数据是多于该整体区域的某个子区域所包括的图像数据的，这样，遍历该整体区域所包括的各个图像数据的重定位方式所需耗费的计算资源，明显高于遍历该整体区域内的子区域所包括的各个图像数据的重定位方式所需耗费的计算资源。Usually, in a certain area, the mobile robot is repositioned by traversing multiple frames of image data included in the area, and the image data included in the overall area is more than included in a certain sub-area of the overall area. image data, in this way, the computing resources required to traverse the relocation of each image data included in the overall area are significantly higher than the relocation method of traversing each image data included in the sub-region within the overall area. The computing resources required.

进而，为了节约计算资源，在对移动机器人进行重定位时，可以首先在完整的先验地图中，确定该移动机器人的当前位置所属的子地图，之后，可以利用所确定的子地区所包括的图像数据确定该移动机器人相对于该子地图的位姿。这样，确定移动机器人在较小的子区域中的位姿的重定位方式，相比于直接确定移动机器人在完整的先验地图的位姿的重定位方式，计算量更小，效率更高。Furthermore, in order to save computing resources, when relocating a mobile robot, the sub-map to which the mobile robot's current position belongs can be determined first in the complete prior map, and then, the sub-map included in the determined sub-region can be used. The image data determines the pose of the mobile robot relative to the submap. In this way, the relocation method of determining the position and posture of the mobile robot in a smaller sub-area requires less calculation and is more efficient than the relocation method of directly determining the position and posture of the mobile robot in the complete a priori map.

基于此，在对移动机器人进行重定位之前，可以首先将上述先验地图，划分为多个先验子地图，这样，在对移动机器人进行重定位时，可以首先确定该移动机器人所在的先验子地图，然后，利用所确定的先验子地图，确定该移动机器人的重定位位姿。Based on this, before relocating the mobile robot, the above-mentioned prior map can be divided into multiple prior sub-maps. In this way, when relocating the mobile robot, the prior map where the mobile robot is located can be first determined. sub-map, and then use the determined prior sub-map to determine the relocation pose of the mobile robot.

下面，对先验子地图的构建方式，进行说明。Next, the construction method of the prior submap is explained.

可选的，一种具体实现方式中，如图2所示，先验子地图的构建方式，可以包括如下步骤S201-S202：Optionally, in a specific implementation manner, as shown in Figure 2, the construction method of the a priori submap may include the following steps S201-S202:

S201：获取先验地图和布局子地图；S201: Obtain the prior map and layout sub-map;

S202：在先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的先验子地图。S202: In the prior map, perform template matching on each layout sub-map to obtain the prior sub-map matched by the layout sub-map.

在本具体实现方式中，在构建任一空间的先验子地图时，可以首先获取该空间的先验地图和布局子地图。In this specific implementation, when constructing a priori submap of any space, the priori map and layout submap of the space can be obtained first.

其中，如前所述，上述先验地图是基于移动机器人的区域探索得到的地图，每个先验地图中包括该空间的地图信息和移动机器人的位姿信息；而上述布局子地图是按照该空间的空间布局图中的区域分割线对该空间的空间布局图进行分割所得到的子地图。Among them, as mentioned above, the above-mentioned prior map is a map obtained based on the area exploration of the mobile robot. Each prior map includes the map information of the space and the pose information of the mobile robot; and the above-mentioned layout sub-map is based on the The submap obtained by dividing the spatial layout map of the space by the area dividing lines in the spatial layout map of the space.

由于先验地图和布局地区所表征的是同一空间，从而，针对每个布局子地图，均可以在先验地图中，确定与该布局子地图表征同一子区域的先验子地图。基于此，在获取该空间的先验地图和布局子地图后，针对每个布局子地图，可以在该先验地图中，对该布局子地图进行模板匹配，从而，在该先验地图中，确定与每个布局子地图模板匹配的先验子地图。这样，通过对每个局部子地图进行模板匹配，便可以得到该空间的多个先验子地图。Since the prior map and the layout area represent the same space, for each layout sub-map, the prior sub-map representing the same sub-region as the layout sub-map can be determined in the prior map. Based on this, after obtaining the a priori map and layout submap of the space, for each layout submap, template matching can be performed on the layout submap in the a priori map, so that in the a priori map, Determine a priori submaps that match each layout submap template. In this way, by performing template matching on each local submap, multiple prior submaps of the space can be obtained.

其中，所谓模板匹配是指：在先验地图中，确定与上述布局子地图模板匹配的最小地图区域。The so-called template matching refers to: in the prior map, determining the minimum map area that matches the above-mentioned layout sub-map template.

示例性的，在构建某一房屋的先验子地图时，可以首先获取该房屋的先验地图和布局子地图。之后，针对用于表征该房屋的厨房区域的布局子地图，可以将该布局子地图与上述先验地图进行模板匹配，从而，在先验地图中，确定与该布局子地图模板匹配的地图区域，作为与该布局子地图相匹配的先验子地图，即得到用于表征厨房区域的先验子地图。For example, when constructing a priori submap of a certain house, the priori map and layout submap of the house can be obtained first. Afterwards, for the layout submap used to characterize the kitchen area of the house, the layout submap can be template matched with the above-mentioned prior map, thereby determining the map area matching the layout submap template in the prior map. , as the prior submap matching the layout submap, that is, the prior submap used to characterize the kitchen area is obtained.

在一些情况下，各个布局子地图的模板可能较为相似，然而，如前所述，每个局部子地图所对应的连接关系是不同的。从而，针对每个局部子地图，还可以结合该子地图所对应的连接关系，在先验地图中，确定与该局部子地图相匹配的先验子地图。In some cases, the templates of each layout submap may be relatively similar. However, as mentioned above, the connection relationships corresponding to each local submap are different. Therefore, for each local submap, the connection relationship corresponding to the submap can also be combined to determine the a priori submap matching the local submap in the prior map.

基于此，可选的，可以获取先验地图、布局子地图以及每个布局子地图对应的连接关系，之后，针对每个布局子地图，在先验地图中，对该布局子地图进行模板匹配，并利用该布局子地图对应的连接关系，确定该布局子地图所匹配的先验子地图。其中，每个布局子地图对应的连接关系为：每个布局子地图所表征区域与该布局子地图所表征区域的可通行通道的连接关系。Based on this, optionally, you can obtain the prior map, layout sub-map and the connection relationship corresponding to each layout sub-map. Then, for each layout sub-map, perform template matching on the layout sub-map in the prior map. , and use the connection relationship corresponding to the layout submap to determine the prior submap matched by the layout submap. Wherein, the connection relationship corresponding to each layout sub-map is: the connection relationship between the area represented by each layout sub-map and the passable channel of the area represented by the layout sub-map.

此外，在先验地图中，也可以根据目标机器人的移动轨迹，确定在先验地图中的各个子地图之间的可通行关系，而在布局子地图与某一先验子地图相匹配时，该布局子地图的连接关系，与该布局子地图所匹配的先验子地图的可通行关系也是一致的。In addition, in the prior map, the traversable relationship between each sub-map in the prior map can also be determined based on the movement trajectory of the target robot. When the layout sub-map matches a certain prior sub-map, The connection relationship of the layout sub-map is also consistent with the accessibility relationship of the prior sub-map matched by the layout sub-map.

基于此，可选的，一种具体实现方式中，上述步骤S202，在先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的先验子地图，可以包括如下步骤11-13：Based on this, optionally, in a specific implementation, in the above step S202, in the a priori map, template matching is performed on each layout submap to obtain the a priori submap matched by the layout submap, which may include the following Steps 11-13:

步骤11：在先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的初始子地图；Step 11: In the prior map, perform template matching on each layout submap to obtain the initial submap matched by the layout submap;

步骤12：获取每个布局子地图对应的连接关系和每个初始子地图对应的可通行关系，其中，每个布局子地图对应的连接关系为：每个布局子地图所表征区域与该布局子地图所表征区域的可通行通道的连接关系，每个初始子地图对应的可通行关系为：基于构建先验地图时，目标机器人的移动轨迹确定的，每个初始子地图与相邻初始子地图的可通行关系；Step 12: Obtain the connection relationship corresponding to each layout submap and the passability relationship corresponding to each initial submap. Among them, the connection relationship corresponding to each layout submap is: the area represented by each layout submap and the area represented by the layout submap. The connection relationship of the accessible channels in the area represented by the map. The accessible relationship corresponding to each initial sub-map is: based on the movement trajectory of the target robot when constructing the prior map, each initial sub-map and the adjacent initial sub-map passability relationship;

步骤13：针对每个布局子地图，利用该布局子地图对应的连接关系以及该布局子地图对应的初始子地图对应的可通行关系，对该布局子地图对应的初始子地图进行校正，得到该布局子地图所匹配的先验子地图。Step 13: For each layout submap, use the connection relationship corresponding to the layout submap and the passability relationship corresponding to the initial submap corresponding to the layout submap to correct the initial submap corresponding to the layout submap to obtain the The prior submap that the layout submap matches.

在本具体实现方式中，在获取先验地图和布局子地图后，可以在先验地图中，对每个布局子地图进行模板匹配，从而，得到该布局子地图所匹配的初始子地图；In this specific implementation, after obtaining the prior map and layout sub-map, template matching can be performed on each layout sub-map in the prior map, thereby obtaining the initial sub-map matched by the layout sub-map;

之后，可以确定每个布局子地图对应的连接关系和每个初始子地图对应的可通行关系，其中，每个布局子地图对应的连接关系为：每个布局子地图所表征区域与该布局子地图所表征区域的可通行通道的连接关系；每个初始子地图对应的可通行关系为：基于构建先验地图时，目标机器人的移动轨迹确定的，每个初始子地图与相邻初始子地图的可通行关系。After that, the connection relationship corresponding to each layout sub-map and the passability relationship corresponding to each initial sub-map can be determined, where the connection relationship corresponding to each layout sub-map is: the area represented by each layout sub-map and the area represented by the layout sub-map. The connection relationship of the accessible channels in the area represented by the map; the accessible relationship corresponding to each initial sub-map is: based on the movement trajectory of the target robot when constructing the prior map, each initial sub-map and the adjacent initial sub-map accessible relationship.

进而，可以获取上述每个布局子地图对应的连接关系和上述每个初始子地图对应的可通行关系。接着，针对每个布局子地图，可以利用该布局子地图对应的连接关系以及该布局子地图对应的初始子地图对应的可通行关系，对该布局子地图所对应的初始子地图进行校正，并将校正后的初始子地图，作为该布局子地图所匹配的先验子地图。Furthermore, the connection relationship corresponding to each of the above layout sub-maps and the passability relationship corresponding to each of the above-mentioned initial sub-maps can be obtained. Then, for each layout submap, the connection relationship corresponding to the layout submap and the passability relationship corresponding to the initial submap corresponding to the layout submap can be used to correct the initial submap corresponding to the layout submap, and The corrected initial submap is used as the prior submap matched by the layout submap.

由于先验子地图中包括多张相机位姿图像，而布局子地图中可以包括该空间布局中的可通行通道，从而，在确定与布局子地图相匹配的先验子地图时，可以建立布局子地图中的可通行通道与先验子地图中的相机位姿的对应关系。Since the prior submap includes multiple camera pose images, and the layout submap can include passable channels in the spatial layout, the layout can be established when determining the prior submap that matches the layout submap. The correspondence between the passable channels in the submap and the camera pose in the prior submap.

这样，在构建关于该区域的先验子地图后，在移动机器人在该区域内发生定位丢失，并可以利用预先构建的该区域的先验子地图，对该移动机器人进行重定位。In this way, after constructing a priori sub-map about the area, if the mobile robot loses its positioning in the area, the mobile robot can be repositioned using the pre-constructed a priori sub-map of the area.

下面，结合附图，对本申请实施例提供的一种重定位方法进行具体说明。Below, a relocation method provided by the embodiment of the present application will be described in detail with reference to the accompanying drawings.

图3为本申请实施例提供的一种重定位方法的流程示意图，如图3所示，该方法可以包括如下步骤S301-S304。Figure 3 is a schematic flowchart of a relocation method provided by an embodiment of the present application. As shown in Figure 3, the method may include the following steps S301-S304.

S301：基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定目标机器人的初始参考位姿；S301: Based on the first frame of images collected by the image acquisition device mounted on the target robot, determine the initial reference pose of the target robot;

在目标机器人发生定位丢失时，需要确定该目标机器人相对于先验地图的重定位位姿时，可以首先获取该目标机器人所搭载的图像采集设备所采集的第一帧图像，之后，基于上述第一帧图像，确定该目标机器人的初始参考位姿。When the target robot loses its positioning and needs to determine the relocation pose of the target robot relative to the a priori map, the first frame of images collected by the image acquisition device mounted on the target robot can be first obtained, and then based on the above-mentioned third One frame of image determines the initial reference pose of the target robot.

其中，上述目标机器人的初始参考位姿是该目标机器人相对于以采集该第一帧图像的位置为起点，且包括给位置的地图的位姿。Wherein, the initial reference pose of the above-mentioned target robot is the pose of the target robot relative to a map that takes the position where the first frame image is collected as a starting point and includes the given position.

在获取到上述第一帧图像后，可以对该第一帧图像进行位姿分析，从而，得到图像采集设备采集该第一帧图像时，目标机器人的位姿。After the above-mentioned first frame of image is acquired, the pose of the first frame of image can be analyzed, thereby obtaining the pose of the target robot when the image acquisition device collects the first frame of image.

可选的，可以利用预设的位姿分析算法，确定在图像采集设备采集该第一帧图像时，目标机器人的位姿。其中，上述位姿分析算法可以是BA(Bundle Adjustment，光束法平差)优化算法，也可以是其他算法，这都是合理的，在本申请实施例中不做具体限定。Optionally, a preset pose analysis algorithm can be used to determine the pose of the target robot when the image acquisition device collects the first frame of images. The above pose analysis algorithm may be a BA (Bundle Adjustment) optimization algorithm or other algorithms. This is reasonable and is not specifically limited in the embodiments of this application.

在上述目标图像采集设备为双目相机时，可能存在双目相机的左目相机和右目相机不同步，使得所得到的左目相机所采集的左目图像与右目相机所采集的右目图像不同步的情况。例如，左目相机与右目相机的图像采集时刻可能是不同步的，从而，所得到的左目图像与右目图像是不同步的。When the target image acquisition device is a binocular camera, the left-eye camera and the right-eye camera of the binocular camera may be out of sync, causing the resulting left-eye image captured by the left-eye camera to be out of sync with the right-eye image captured by the right-eye camera. For example, the image acquisition times of the left-eye camera and the right-eye camera may be asynchronous, and thus the obtained left-eye images and right-eye images are asynchronous.

而在左目图像与右目图像不同步时，该左目图像与右目图像不是同一时刻下的图像，从而，利用左目图像和右目图像计算得到的视差和深度值，其结果是不准的，进而，利用上述左目图像和右目图像确定目标机器人的位姿，所得到的位姿存在较大误差。基于此，为了提高定位准确性，在上述图像采集设备为双目相机时，在利用双目相机所采集的第一帧图像确定目标机器人的初始参考位姿之前，可以首先校验该双目相机所采集的左目图像和右目图像是否同步。When the left-eye image and the right-eye image are out of sync, the left-eye image and the right-eye image are not images at the same time. Therefore, the disparity and depth values calculated using the left-eye image and the right-eye image are inaccurate. Furthermore, using The above left eye image and right eye image determine the pose of the target robot, and there is a large error in the resulting pose. Based on this, in order to improve the positioning accuracy, when the above-mentioned image acquisition device is a binocular camera, before using the first frame of images collected by the binocular camera to determine the initial reference pose of the target robot, the binocular camera can first be verified Whether the collected left eye image and right eye image are synchronized.

基于此，可选的，一种具体实现方式中，图像采集设备为双目相机，上述步骤S301，可以包括如下步骤21：Based on this, optionally, in a specific implementation manner, the image acquisition device is a binocular camera, and the above step S301 may include the following step 21:

步骤21：基于目标机器人所搭载图像采集设备所采集到第一帧同步的左目图像和右目图像，确定目标机器人的初始参考位姿。Step 21: Determine the initial reference pose of the target robot based on the synchronized left eye image and right eye image of the first frame collected by the image acquisition device mounted on the target robot.

在本具体实现方式中，在图像采集设备为双目相机时，可以首先获取目标机器人所搭载的双目相机的左目相机所采集的左目图像以及双目相机的右目相机所采集的右目图像，之后，判断上述左目图像和右目图像是否同步。In this specific implementation, when the image acquisition device is a binocular camera, the left eye image collected by the left eye camera of the binocular camera mounted on the target robot and the right eye image collected by the right eye camera of the binocular camera can be first acquired, and then , determine whether the above left-eye image and right-eye image are synchronized.

进而，在第一次判断出上述左目图像和右目图像同步时，则该同步的左目图像和右目图像，即为可以获取上述双目相机所采集到的第一帧同步的左目图像和右目图像，从而，可以利用上述第一帧同步的左目图像和右目图像，确定目标机器人的初始参考位姿。Furthermore, when it is determined for the first time that the left-eye image and the right-eye image are synchronized, the synchronized left-eye image and right-eye image are the synchronized left-eye images and right-eye images of the first frame collected by the above-mentioned binocular camera. Therefore, the initial reference pose of the target robot can be determined using the synchronized left-eye image and right-eye image of the first frame.

其中，在上述判断过程中，在每次判断出上述左目图像和右目图像不同步时，可以对上述双目相机进行调整，之后，获取调整后的双目相机所采集到的左目图像和右目图像，并再次判断所采集到的左目图像和右目图像是否同步。直至第一次判断出双目相机所采集到的左目图像和右目图像同步，便可以将同步的左目图像和右目图像，作为双目相机所采集到的第一帧同步的左目图像和右目图像，并利用第一帧同步的左目图像和右目图像，确定目标机器人的初始参考位姿。Wherein, in the above judgment process, every time it is judged that the above left eye image and the right eye image are out of sync, the above binocular camera can be adjusted, and then the left eye image and right eye image collected by the adjusted binocular camera are obtained. , and judge again whether the collected left eye image and right eye image are synchronized. Until it is judged for the first time that the left-eye image and the right-eye image collected by the binocular camera are synchronized, the synchronized left-eye image and right-eye image can be regarded as the first synchronized left-eye image and right-eye image captured by the binocular camera. And use the synchronized left eye image and right eye image of the first frame to determine the initial reference pose of the target robot.

可选的，可以通过判断上述双目相机的左目相机和右目相机是否同步，来确定上述双目相机所采集的左目图像和右目图像是否同步。若上述双目相机的左目相机和右目相机同步，则上述双目相机所采集的左目图像和右目图像同步；若上述双目相机的左目相机和右目相机不同步，则上述双目相机所采集的左目图像和右目图像也是不同步的。Optionally, it can be determined whether the left-eye image and the right-eye image collected by the above-mentioned binocular camera are synchronized by determining whether the left-eye camera and the right-eye camera of the above-mentioned binocular camera are synchronized. If the left-eye camera and the right-eye camera of the above-mentioned binocular camera are synchronized, the left-eye image and the right-eye image collected by the above-mentioned binocular camera are synchronized; if the left-eye camera and the right-eye camera of the above-mentioned binocular camera are not synchronized, then the images collected by the above-mentioned binocular camera are synchronized. The left eye image and the right eye image are also out of sync.

这样，由于同步的左目图像和右目图像之间的误差较小，从而，利用第一帧同步的左目图像和右目图像，所确定的目标机器人的初始参考位姿是较为准确的。In this way, since the error between the synchronized left-eye image and the right-eye image is small, the initial reference pose of the target robot determined using the synchronized left-eye image and right-eye image in the first frame is relatively accurate.

S302：获取图像采集设备所采集的多帧参考图像，并基于初始参考位姿，确定目标机器人在采集每帧参考图像时的参考位姿，得到每帧参考图像对应的参考位姿；S302: Obtain multiple frames of reference images collected by the image acquisition device, and based on the initial reference pose, determine the reference pose of the target robot when collecting each frame of reference image, and obtain the reference pose corresponding to each frame of reference image;

在确定目标机器人的初始参考位姿后，可以以该初始参考位姿为参考，控制目标机器人移动，并且，在该目标机器人移动的过程中，该目标机器人所搭载的图像采集设备可以采集多帧图像。这样，电子设备便可以获取图像采集设备所采集的多帧参考图像。After determining the initial reference pose of the target robot, the target robot can be controlled to move using the initial reference pose, and during the movement of the target robot, the image acquisition device mounted on the target robot can collect multiple frames. image. In this way, the electronic device can acquire multiple frames of reference images collected by the image acquisition device.

其中，可选的，可以获取图像采集设备所采集到的每帧图像，并将上述每帧图像作为参考图像，从而，得到多帧参考图像。Optionally, each frame of image collected by the image acquisition device can be acquired, and each frame of image can be used as a reference image, thereby obtaining multiple frames of reference images.

由于图像采集设备所采集的图像数量较多，将图像采集设备所采集的每帧图像作为参考图像，并确定每帧参考图像的参考位姿，所需耗费的运算资源较多。基于此，为了节约计算资源，可以在图像采集设备所采集的图像中确定关键帧，并将所确定的关键帧作为参考图像。Since the image acquisition device collects a large number of images, it requires a lot of computing resources to use each frame of image collected by the image acquisition device as a reference image and determine the reference pose of each frame of reference image. Based on this, in order to save computing resources, key frames can be determined in the images collected by the image acquisition device, and the determined key frames can be used as reference images.

可选的，一种具体实现方式中，上述S302，可以包括如下步骤31：Optionally, in a specific implementation manner, the above S302 may include the following step 31:

步骤31：获取图像采集设备所采集到的多个关键帧，作为多帧参考图像；Step 31: Obtain multiple key frames collected by the image acquisition device as multi-frame reference images;

其中，多个关键帧包括：按照预设时间间隔所采集的多帧图像、按照目标机器人的预设移动距离所采集的多帧图像，或者，包括预设图像特征的多帧图像。The multiple key frames include: multiple frames of images collected according to a preset time interval, multiple frames of images collected according to a preset movement distance of the target robot, or multiple frames of images including preset image features.

在本具体实现方式中，电子设备可以获取图像采集设备所采集的多个关键帧，并将所获取的多个关键帧，作为多帧参考图像。In this specific implementation manner, the electronic device can acquire multiple key frames collected by the image acquisition device, and use the multiple acquired key frames as multi-frame reference images.

其中，上述多个关键帧可以包括图像采集设备按照预设时间间隔所采集的多帧图像、图像采集设备按照目标机器人的预设移动距离所采集的多帧图像，或者，包括预设图像特征的多帧图像。The multiple key frames may include multiple frames of images collected by the image acquisition device according to preset time intervals, multiple frames of images collected by the image acquisition device according to the preset movement distance of the target robot, or including preset image features. Multiple frame images.

并且，上述预设时间间隔、预设移动距离以及预设图像特征均可以按照实际需要进行设定，例如，上述预设时间间隔可以是3秒、5秒等；上述预设移动距离可以是1米、5米等；而上述预设图像特征可以是障碍物等，这都是合理的，在本申请实施例中不做具体限定。Moreover, the above-mentioned preset time interval, preset movement distance and preset image characteristics can be set according to actual needs. For example, the above-mentioned preset time interval can be 3 seconds, 5 seconds, etc.; the above-mentioned preset movement distance can be 1 meters, 5 meters, etc.; and the above-mentioned preset image features can be obstacles, etc., which are reasonable and are not specifically limited in the embodiments of this application.

此外，可选的，图像采集设备可以按照预设时间间隔采集图像，并将所采集的多帧图像发送给电子设备，这样，电子设备便可以获取图像采集设备按照预设时间间隔所采集的多个关键帧，并将多个关键帧，作为多帧参考图像；In addition, optionally, the image acquisition device can collect images at preset time intervals and send the collected multiple frames of images to the electronic device. In this way, the electronic device can acquire multiple frames of images collected by the image acquisition device at preset time intervals. key frames, and use multiple key frames as multi-frame reference images;

可选的，电子设备可以接收图像采集设备所采集的每帧图像，并将所获取的多帧图像中包括预设图像特征的图像，作为关键帧，从而，将所确定的多个关键帧，作为多帧参考图像。Optionally, the electronic device can receive each frame of image collected by the image acquisition device, and use the image including the preset image characteristics among the acquired multiple frame images as a key frame, thereby using the determined multiple key frames, as a multi-frame reference image.

电子设备在获取多帧参考图像后，可以基于上述初始参考位姿，确定在图像采集设备采集每帧参考图像时，上述目标机器人的参考位姿。之后，针对每帧参考图像，将图像采集设备采集该帧参考图像时，目标机器人的参考位姿，确定为该帧参考图像所对应的参考位姿，从而，得到每帧参考图像所对应的参考位姿。After acquiring multiple frames of reference images, the electronic device can determine, based on the above-mentioned initial reference pose, the reference pose of the target robot when the image acquisition device collects each frame of reference image. After that, for each frame of reference image, when the image acquisition device collects the reference image of the frame, the reference pose of the target robot is determined as the reference pose corresponding to the reference image of the frame, thereby obtaining the reference pose corresponding to each frame of reference image. Posture.

S303：利用多帧参考图像和每帧参考图像对应的参考位姿，构建目标机器人当前所在的区域局部地图，并在预设的先验子地图中，确定与区域局部地图模板匹配的目标子地图；S303: Use the multi-frame reference images and the reference pose corresponding to each frame of reference image to construct a local map of the area where the target robot is currently located, and determine the target sub-map that matches the regional local map template in the preset a priori sub-map. ;

其中，先验子地图是利用预设的布局子地图对目标机器人所属空间的先验地图进行分割得到的，布局子地图是按照区域分割线对目标机器人所属空间的空间布局图进行分割得到的；Among them, the prior submap is obtained by dividing the prior map of the space where the target robot belongs by using the preset layout submap, and the layout submap is obtained by dividing the spatial layout map of the space where the target robot belongs according to the area dividing line;

在得到上述多帧参考图像，并确定每帧参考图像的参考位姿后，可以利用上述多帧参考图像以及每帧参考图像的参考位姿，构建目标机器人当前所在的区域局部地图。After obtaining the above-mentioned multi-frame reference images and determining the reference pose of each frame of reference image, the above-mentioned multi-frame reference images and the reference pose of each frame of reference image can be used to construct a local map of the area where the target robot is currently located.

在得到上述目标机器人当前所在的区域局部地图后，电子设备可以获取目标机器人所属空间的预设的先验子地图。After obtaining the above-mentioned local map of the area where the target robot is currently located, the electronic device can obtain the preset prior sub-map of the space to which the target robot belongs.

其中，上述先验子地图可以是利用预设的布局子地图对目标机器人所属空间的先验地图进行分割得到的，而上述布局子地图则如前所述，可以是按照区域分割线对目标机器人所属空间的空间布局图进行分割得到的。Among them, the above-mentioned prior sub-map can be obtained by dividing the prior map of the space where the target robot belongs by using a preset layout sub-map, and the above-mentioned layout sub-map can be obtained by dividing the target robot according to the area dividing line as mentioned above. It is obtained by dividing the spatial layout of the space it belongs to.

并且，上述先验子地图的构建方式如前所述，此处不再赘述。Moreover, the construction method of the above-mentioned prior sub-map is as mentioned above and will not be described again here.

可选的，在上述先验子地图为栅格地图时，为了便于将上述区域栅格地图与上述先验子地图进行模板匹配，可以将上述区域局部地图处理为栅格地图。Optionally, when the above-mentioned a priori sub-map is a raster map, in order to facilitate template matching between the above-mentioned regional raster map and the above-mentioned a priori sub-map, the above-mentioned regional local map can be processed into a raster map.

这样，在获取预设的先验子地图后，电子设备可以在上述先验子地图中，确定与区域局部地图模板匹配的目标子地图。In this way, after acquiring the preset a priori submap, the electronic device can determine the target submap that matches the regional local map template in the above a priori submap.

可选的，可以分别计算区域局部地图与每个先验子地图的模板相似度，并将最大模板相似度所对应的先验子地图，确定为与区域局部地图模板匹配的目标子地图。Optionally, the template similarity between the regional local map and each prior submap can be calculated separately, and the prior submap corresponding to the maximum template similarity can be determined as the target submap that matches the regional local map template.

此外，由于上述先验子地图均存在与该先验子地图相匹配的布局子地图，从而，在确定区域局部地图与每个先验子地图的匹配关系时，可以将区域局部地图与每个先验子地图进行模板匹配，并将区域局部地图与每个先验子地图相匹配的布局子地图的进行模板匹配，从而，提高模板匹配的可靠性。In addition, since the above-mentioned prior sub-maps all have layout sub-maps that match the prior sub-map, when determining the matching relationship between the regional local map and each prior sub-map, the regional local map and each prior sub-map can be determined. The prior sub-map performs template matching, and the regional local map is template matched with the layout sub-map that matches each prior sub-map, thereby improving the reliability of template matching.

基于此，可选的，一种具体实现方式中，上述步骤S303，在预设的先验子地图中，确定与区域局部地图模板匹配的目标子地图，可以包括如下步骤41-42：Based on this, optionally, in a specific implementation, the above step S303, determining the target submap that matches the regional local map template in the preset a priori submap, may include the following steps 41-42:

步骤41：针对每个先验子地图，对该先验子地图和区域局部地图进行模板匹配，并对该先验子地图所匹配的布局子地图和区域局部地图进行模板匹配，得到匹配结果；Step 41: For each prior sub-map, perform template matching on the prior sub-map and the regional local map, and perform template matching on the layout sub-map and regional local map matched by the prior sub-map to obtain the matching result;

步骤42：基于匹配结果，在预设的先验子地图中，确定目标子地图。Step 42: Based on the matching results, determine the target submap in the preset a priori submap.

在本具体实现方式中，在构建目标机器人当前所在的区域局部地图后，针对每个先验子地图，可以将上述区域局部地图与该先验子地图进行模板匹配，并将该先验子地图所匹配的布局子地图和区域局部地图进行模板匹配，得到匹配结果。In this specific implementation, after constructing a local map of the area where the target robot is currently located, for each prior sub-map, the above-mentioned regional local map can be template matched with the prior sub-map, and the prior sub-map can be The matched layout submap and the regional local map are template matched to obtain the matching result.

之后，在上述匹配结果表征区域局部地图与该先验子地图模板匹配，且该区域局部地图与该先验子地图所对应的布局子地图模板匹配时，可以将该先验子地图，确定为区域局部地图所对应的目标子地图。Afterwards, when the local map of the above-mentioned matching result representation area matches the prior sub-map template, and the regional local map matches the layout sub-map template corresponding to the prior sub-map, the prior sub-map can be determined as The target submap corresponding to the regional local map.

在确定与该区域局部地图模板匹配的目标子地图后，便可以确定目标机器人在先验地图中所在的子区域。其中，由于上述确定与目标机器人当前所在的区域局部地图的目标子地图的过程，可以确定目标机器人当前所在区域在先验地图中所对应的子区域，而不能确定该目标机器人相对于先验地图的位姿，从而，可以将确定与目标机器人当前所在的区域局部地图的目标子地图的过程，称为目标机器人的粗定位过程。After determining the target sub-map that matches the local map template of the area, the sub-area where the target robot is located in the prior map can be determined. Among them, due to the above-mentioned process of determining the target sub-map of the local map of the area where the target robot is currently located, the sub-area corresponding to the area where the target robot is currently located in the a priori map can be determined, but it cannot be determined that the target robot is relative to the a priori map. Therefore, the process of determining the target sub-map of the local map of the area where the target robot is currently located can be called the rough positioning process of the target robot.

此外，由于先验子地图是基于布局子地图划分得到的，从而，针对任一先验子地图，与该先验子地图相匹配的区域局部地图，可以与该先验子地图所对应的布局子地图相匹配。基于此，为了提高所确定的布局子地图的准确性，在预设的先验子地图中，确定区域局部地图相匹配的目标子地图后，还可以通过确定该区域局部地图与该目标子地图相匹配的布局子地图是否匹配，来验证上述区域局部地图与目标子地图的匹配关系。In addition, since the prior sub-map is divided based on the layout sub-map, for any prior sub-map, the regional local map matching the prior sub-map can be the same as the layout corresponding to the prior sub-map. submap to match. Based on this, in order to improve the accuracy of the determined layout submap, after determining the target submap that matches the regional local map in the preset a priori submap, it is also possible to determine the regional local map and the target submap. Check whether the matching layout submaps match to verify the matching relationship between the local map of the above area and the target submap.

由于先验地图和区域局部地图均是由目标机器人探索得到的，并且，所得到的先验地图以及区域局部地图中可能存在边缘模糊、扭曲等现象，从而，可以确定先验地图与布局地图的全局偏差，之后，利用该全局偏差对区域局部地图进行校正。Since both the prior map and the regional local map are explored by the target robot, and the obtained prior map and the regional local map may have edge blurring, distortion, etc., the relationship between the prior map and the layout map can be determined. The global deviation is then used to correct the regional local map.

基于此，可选的，可以获取先验地图和空间布局图，之后，将上述先验地图和空间布局图进行外形匹配，确定上述先验地图和空间布局图的全局偏差，并利用该全局偏差，对区域局部地图进行校正。然后，在各个布局子地图中，确定与目标子地图相匹配的布局子地图，并将校正后的区域局部地图与所确定的布局子地图进行匹配。进而，若该区域局部地图与所确定的布局子地图相匹配，则该区域局部地图与目标子地图相匹配；若该区域局部地图与所确定的布局子地图不匹配，则可以输出用于表征上述区域局部地图匹配失败的通知消息。这样，通过将区域局部地图与目标子地图相匹配的布局子地图分别进行匹配，可以对区域局部地图和目标子地图的匹配关系进行校验，从而，提高重定位的准确性。Based on this, optionally, a priori map and spatial layout map can be obtained, and then the above-mentioned prior map and spatial layout map can be matched in appearance to determine the global deviation between the above-mentioned prior map and spatial layout map, and use the global deviation , correct the regional local map. Then, in each layout sub-map, a layout sub-map matching the target sub-map is determined, and the corrected regional local map is matched with the determined layout sub-map. Furthermore, if the local map of the area matches the determined layout submap, then the local map of the area matches the target submap; if the local map of the area does not match the determined layout submap, then it can be output for representation. Notification message of failed local map matching in the above area. In this way, by matching the regional local map with the layout submap that matches the target submap respectively, the matching relationship between the regional local map and the target submap can be verified, thereby improving the accuracy of relocation.

S304：对图像采集设备所采集的当前图像与目标子地图所关联的各个基准图像进行相似度匹配，确定与当前图像所匹配的目标基准图像，并根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿。S304: Perform similarity matching between the current image collected by the image acquisition device and each reference image associated with the target sub-map, determine the target reference image that matches the current image, and determine based on the target pose associated with the target reference image. The relocation pose of the target robot relative to the prior map.

由于先验地图是基于目标机器人所搭载的图像采集设备所采集的多帧图像构建的，从而，每个先验子地图均关联有图像采集设备所采集的多帧图像，基于此，可以将每个先验子地图所关联的图像，作为基准图像。Since the prior map is constructed based on multiple frames of images collected by the image acquisition equipment mounted on the target robot, each prior sub-map is associated with multiple frames of images collected by the image acquisition equipment. Based on this, each prior sub-map can be The image associated with a priori submap is used as the base image.

基于此，在确定与上述区域局部地图模板匹配的目标子地图后，为了确定目标机器人相对于先验地图的重定位位姿，可以确定目标子图像所关联的各个基准图像，并获取目标机器人所搭载的图像采集设备所采集的当前图像。Based on this, after determining the target sub-map that matches the local map template of the above area, in order to determine the relocation pose of the target robot relative to the prior map, each reference image associated with the target sub-image can be determined, and the target robot's location can be obtained. The current image collected by the mounted image acquisition device.

之后，可以将上述当前图像和各个基准图像进行相似度匹配，确定与当前图像所匹配的目标基准图像。也就是说，电子设备可以计算当前图像与每个基准图像的相似度，并将最大相似度所对应的基准图像作为与当前图像所匹配的目标基准图像。Afterwards, similarity matching can be performed between the current image and each reference image to determine the target reference image that matches the current image. That is to say, the electronic device can calculate the similarity between the current image and each reference image, and use the reference image corresponding to the maximum similarity as the target reference image that matches the current image.

其中，可选的，一种具体实现方式中，上述步骤S304，可以包括如下步骤51：Among them, optionally, in a specific implementation manner, the above step S304 may include the following step 51:

步骤51：对图像采集设备所采集的最近的关键帧与目标子地图所关联的各个基准图像进行相似度匹配，确定与当前图像所匹配的目标基准图像。Step 51: Perform similarity matching between the latest key frame collected by the image acquisition device and each reference image associated with the target submap, and determine the target reference image that matches the current image.

在本具体实现方式中，由于可以将图像采集设备最近所采集的图像所关联的位姿，作为目标机器人的位姿，而关键帧可以是图像采集设备所采集的具有代表性的图像，基于此，为了使得重定位的位姿接近于目标机器人的当前位姿，且为了提高定位的准确性，可以获取图像采集设备所采集的最近的关键帧，并利用该最近的关键帧确定该目标机器人的重定位位姿。In this specific implementation, since the pose associated with the image recently collected by the image acquisition device can be used as the pose of the target robot, and the key frame can be a representative image collected by the image acquisition device, based on this , in order to make the repositioned pose close to the current pose of the target robot, and in order to improve the accuracy of positioning, the most recent key frame collected by the image acquisition device can be obtained, and the most recent key frame can be used to determine the position of the target robot. Reposition pose.

电子设备可以获取图像采集设备所采集的最近的关键帧，例如，电子设备可以在所接收的图像采集设备所采集的多帧图像中，确定采集时间最接近当前时刻的关键帧，作为图像采集设备所采集的最近的关键帧；再例如，电子设备可以将最新接收的图像采集设备所采集的图像，作为图像采集设备所采集的最近的关键帧。The electronic device can obtain the most recent key frame collected by the image acquisition device. For example, the electronic device can determine the key frame whose acquisition time is closest to the current moment among the multiple frames of images collected by the received image acquisition device, as the image acquisition device. The most recent key frame collected; for another example, the electronic device may use the latest received image collected by the image collection device as the latest key frame collected by the image collection device.

之后，电子设备可以将上述图像采集设备所采集的最近的关键帧与目标子地图所关联的各个基准图像进行相似度匹配，从而，基于相似度匹配结果，确定与当前图像所匹配的目标基准图像。Afterwards, the electronic device can perform similarity matching between the most recent key frame collected by the image acquisition device and each reference image associated with the target submap, thereby determining the target reference image that matches the current image based on the similarity matching result. .

其中，可选的，电子设备可以将计算图像采集设备所采集的最近的关键帧与各个基准图像的相似度，之后，将最大相似度所对应的基准图像，确定与当前图像所匹配的目标基准图像。Optionally, the electronic device may calculate the similarity between the most recent key frame collected by the image acquisition device and each reference image, and then determine the target reference that matches the current image based on the reference image corresponding to the maximum similarity. image.

通常，在图像采集设备采集每个基准图像时，该目标机器人均具有与该基准图像相对应的位姿，从而，每个基准图像均关联有一个位姿。Generally, when the image acquisition device collects each reference image, the target robot has a posture corresponding to the reference image, so that each reference image is associated with a posture.

基于此，在确定与当前图像所匹配的目标基准图像后，电子设备可以根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿。Based on this, after determining the target reference image that matches the current image, the electronic device can determine the relocation pose of the target robot relative to the a priori map according to the target pose associated with the target reference image.

可选的，在上述每个基准图像所关联的位姿为：图像采集设备采集该基准图像时，目标机器人相对于先验地图的位姿时，上述目标基准图像所关联的目标位姿，即是目标机器人相对于先验地图的重定位位姿。Optionally, the pose associated with each of the above-mentioned reference images is: when the image acquisition device collects the reference image, the target pose associated with the above-mentioned target reference image is the pose of the target robot relative to the a priori map, that is, is the relocation pose of the target robot relative to the prior map.

可选的，一种具体实现方式中，每个基准图像所关联的位姿为：图像采集设备采集该基准图像时，图像采集设备相对于先验地图的位姿；Optionally, in a specific implementation, the pose associated with each reference image is: the pose of the image acquisition device relative to the a priori map when the image acquisition device collects the reference image;

上述步骤S304，根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿，可以包括如下步骤61：The above-mentioned step S304, determining the relocation pose of the target robot relative to the a priori map based on the target pose associated with the target reference image, may include the following step 61:

步骤61：根据图像采集设备与目标机器人的相对位姿，对目标基准图像所关联的目标位姿进行位姿转换，得到目标机器人相对于先验地图的重定位位姿。Step 61: According to the relative pose of the image acquisition device and the target robot, perform pose conversion on the target pose associated with the target reference image to obtain the repositioned pose of the target robot relative to the prior map.

在本具体实现方式中，上述每个基准图像所关联的位姿为：图像采集设备采集该基准图像时，图像采集设备相对于先验地图的位姿。In this specific implementation manner, the pose associated with each of the above reference images is: the pose of the image acquisition device relative to the a priori map when the image acquisition device collects the reference image.

从而，上述目标基准图像所关联的目标位姿为：图像采集设备采集该目标基准图像时，图像采集设备相对于先验地图的目标位姿。Therefore, the target pose associated with the above-mentioned target reference image is: when the image capture device collects the target reference image, the target pose of the image acquisition device relative to the a priori map.

如前所述，根据图像采集设备和目标机器人的位置关系，可以确定图像采集设备与该目标机器人的相对位姿。也就是说，在确定目标图像采集设备的位姿时，可以利用上述目标图像采集设备和上述图像采集设备与该目标机器人的相对位姿，确定目标机器人的位姿。As mentioned above, according to the positional relationship between the image acquisition device and the target robot, the relative posture of the image acquisition device and the target robot can be determined. That is to say, when determining the pose of the target image capture device, the relative pose of the target image capture device and the target robot can be used to determine the pose of the target robot.

基于此，在基于上述目标基准图像，确定图像采集设备相对于先验地图的目标位姿时，可以根据上述图像采集设备与目标机器人的相对位姿，对上述目标位姿进行位姿转换，即将上述图像采集设备相对于先验地图的目标位姿转化为目标机器人相对于先验地图的位姿，从而，得到目标机器人相对于先验地图的重定位位姿。Based on this, when determining the target pose of the image acquisition device relative to the a priori map based on the above target reference image, the above target pose can be transformed according to the relative pose of the above image acquisition device and the target robot, that is, The target pose of the above-mentioned image acquisition device relative to the prior map is converted into the pose of the target robot relative to the prior map, thereby obtaining the repositioned pose of the target robot relative to the prior map.

其中，由于上述根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿的过程，可以在确定目标机器人当前所在区域在先验地图中所对应的子区域的基础上，进一步，确定该目标机器人相对于先验地图的位姿，从而，可以将确定根据目标基准图像所关联的目标位姿，确定目标机器人相对于先验地图的重定位位姿的过程，称为目标机器人的精定位过程。Among them, due to the above-mentioned process of determining the relocation pose of the target robot relative to the a priori map based on the target pose associated with the target reference image, the sub-region corresponding to the current area of the target robot in the a priori map can be determined. On this basis, further, determine the pose of the target robot relative to the a priori map, thereby determining the target pose associated with the target reference image, and the process of determining the repositioning pose of the target robot relative to the a priori map, It is called the precise positioning process of the target robot.

以上可见，应用本申请实施例提供的方案，可以实现对目标机器人的重定位。并且，通过区域局部地图与先验子地图的模板匹配可以提高区域定位的准确性，从而，可以降低因区域误定位而导致的重定位失败的可能性，提高重定位的成功率；进一步的，通过当前图像与目标子地图的各个基准图像的相似度匹配，可以提高重定位的准确性。此外，仅需将当前图像与目标子地图的各个基准图像进行相似度匹配，而无需遍历完整的先验地图所关联的各个基准图像进行相似度匹配，可以减少重定位的计算量，进而，提高重定位效率。It can be seen from the above that by applying the solution provided by the embodiment of the present application, the target robot can be repositioned. Moreover, the accuracy of regional positioning can be improved through template matching between the regional local map and the prior submap, thereby reducing the possibility of relocation failure caused by regional mislocation and improving the success rate of relocation; further, The accuracy of relocalization can be improved by similarity matching between the current image and each reference image of the target submap. In addition, it only needs to perform similarity matching between the current image and each reference image of the target submap, without traversing each reference image associated with the complete prior map for similarity matching, which can reduce the calculation amount of relocation, thereby improving Relocation efficiency.

为了便于理解本申请实施例提供的一种重定位方法，下面结合图4所示的一种机器人重定位方法的具体实例，进行说明。In order to facilitate understanding of the relocation method provided by the embodiment of the present application, description will be given below with reference to a specific example of a robot relocation method shown in FIG. 4 .

如图4所示，图中户型图即为本申请实施例中的空间布局图，全局先验地图即为本申请实施例中的先验地图。目标机器人在指定房屋内执行清洁任务时，其定位丢失，从而，需要对该目标机器人进行重定位。其中，对目标机器人进行重定位时，可以包括如下步骤S401-S413：As shown in Figure 4, the house type diagram in the figure is the spatial layout diagram in the embodiment of the present application, and the global a priori map is the a priori map in the embodiment of the present application. When the target robot performed a cleaning task in a designated house, its positioning was lost, so the target robot needed to be repositioned. Among them, when relocating the target robot, the following steps S401-S413 may be included:

S401：导入户型图；S401: Import house plan;

S402：导入全局先验地图；S402: Import the global prior map;

S403：户型图识别得2D地图构建金字塔第一层；S403: The first layer of the pyramid is constructed from the 2D map recognized by the house plan;

S404：分割地图区域形成子地图集D以及可通行通道与子地图集的连接关系DT；S404: Divide the map area to form a sub-atlas D and the connection relationship DT between the passable channel and the sub-atlas;

S405：估算2D栅格地图构建金字塔第二层；S405: Estimate the second layer of the 2D raster map to build the pyramid;

S406：两层地图进行外形匹配，结合DT分割第二层地图得到子地图集P；S406: Perform shape matching on the two-layer maps, and combine the DT to segment the second-layer map to obtain the sub-atlas P;

S407：判断左右目图像是否同步；若是，执行步骤S408；S407: Determine whether the left and right eye images are synchronized; if so, execute step S408;

S408：获取左右目图像，确定初始化位姿；S408: Obtain the left and right eye images and determine the initial pose;

S409：创建关键帧图像；S409: Create key frame image;

S410：局部优化，估计机器人位姿；S410: Local optimization, estimating robot pose;

S411：构建2D栅格地图，将局部地图形成金字塔第三层；S411: Construct a 2D raster map and form the local map into the third layer of the pyramid;

S412：第三层地图与第二层地图做模板匹配，并结合DT进行校验，确定目标子地图，从而，实现粗定位；S412: The third-layer map and the second-layer map are template matched, and combined with DT for verification to determine the target sub-map, thereby achieving rough positioning;

S413：在目标子地图内进行关键帧相似度匹配，从而，实现精定位。S413: Perform key frame similarity matching within the target submap to achieve precise positioning.

在对目标机器人进行重定位之前，可以预先导入该指定房屋的户型图和该目标机器人对该指定房屋进行区域探索所得到的全局先验地图。之后，对所导入的户型图进行识别，得关于该户型图的2D(2-Dimension，二维平面)地图，并利用该户型图作为地图金字塔的第一层地图。并利用该户型图中的各个墙壁所在位置，对该户型图的2D地图进行地图区域进行分割，得到子地图集D，以及各个子地图与该户型图中的可通行通道的连接关系DT。Before relocating the target robot, the floor plan of the specified house and the global prior map obtained by the target robot's regional exploration of the specified house can be imported in advance. After that, the imported house plan is identified to obtain a 2D (2-Dimension, two-dimensional plane) map of the house plan, and the house plan is used as the first layer map of the map pyramid. And use the location of each wall in the house plan to divide the map area of the 2D map of the house plan to obtain a sub-map set D, and the connection relationship DT between each sub-map and the accessible passages in the house plan.

对所导入的全局先验地图进行估算，得到该全局先验地图的2D栅格地图，并将该2D栅格地图作为地图金字塔的第二层地图。之后，将第一层地图与第二层地图进行外形匹配，并结合连接关系DT，将2D栅格地图进行分割，得到全局先验地图的子地图集P。The imported global prior map is estimated to obtain a 2D grid map of the global prior map, and the 2D grid map is used as the second layer map of the map pyramid. After that, the first layer map and the second layer map are matched in shape, and combined with the connection relationship DT, the 2D grid map is segmented to obtain the sub-atlas P of the global prior map.

在目标机器人重启后，首先获取该目标机器人所搭载的双目相机所采集的左目图像和右目图像，之后，判断左目图像和右目图像是否同步。在上述左目图像和右目图像同步时，获取获取第一帧左目图像和右目图像，并利用上述第一帧左目图像和右目图像，确定目标机器人的初始化位姿。After the target robot restarts, first obtain the left eye image and the right eye image collected by the binocular camera mounted on the target robot, and then determine whether the left eye image and the right eye image are synchronized. When the above-mentioned left-eye image and right-eye image are synchronized, the first frame of the left-eye image and the right-eye image are obtained, and the above-mentioned first frame of the left-eye image and the right-eye image are used to determine the initialization pose of the target robot.

之后，获取双目相机所采集的多帧图像，并利用多帧图像，确定关键帧图像。之后，利用上述关键帧图像，进行局部优化，确定目标机器人在关键帧图像中的位姿和地图点。之后，利用上述关键帧图像和目标机器人的位姿，构建2D栅格地图，得到目标机器人当前所在区域的局部地图，并将该局部地图作为地图金字塔的第三层地图。Afterwards, obtain the multi-frame images collected by the binocular camera, and use the multi-frame images to determine the key frame image. Afterwards, the above key frame images are used to perform local optimization to determine the pose and map points of the target robot in the key frame images. Afterwards, the above key frame image and the pose of the target robot are used to construct a 2D grid map, and a local map of the current area of the target robot is obtained, and the local map is used as the third layer map of the map pyramid.

第三层地图与第二层地图进行模板匹配，即将局部地图与全局先验地图的子地图集P中的各个子地图进行模板匹配，并结合上述连接关系DT，确定与局部地图相匹配的目标子地图，从而，实现目标机器人的粗定位。The third-level map is template matched with the second-level map, that is, the local map is template matched with each sub-map in the sub-map set P of the global prior map, and combined with the above connection relationship DT, the target matching the local map is determined. sub-map, thereby achieving rough positioning of the target robot.

之后，将目标机器人所采集的最近关键帧，与上述目标子地图所关联的各个关键帧进行相似度匹配，从而，将与最近关键帧的相似度最大的关键帧确定为目标关键帧，并将目标关键帧所关联的目标位姿，确定为目标机器人的重定位位姿，从而，实现目标机器人的精定位。Afterwards, similarity matching is performed between the recent key frames collected by the target robot and each key frame associated with the above-mentioned target submap, thereby determining the key frame with the greatest similarity to the recent key frame as the target key frame, and The target pose associated with the target key frame is determined as the relocation pose of the target robot, thereby achieving precise positioning of the target robot.

基于相同的发明构思，相应于上述本申请实施例提供的图3所示的一种重定位方法，本申请实施例还提供了一种机器人。Based on the same inventive concept, corresponding to the relocation method shown in Figure 3 provided by the embodiment of the present application, the embodiment of the present application also provides a robot.

图5为本申请实施例提供的一种机器人的结构示意图，如图5所示，该机器人可以包括图像采集设备100和处理器200。FIG. 5 is a schematic structural diagram of a robot provided by an embodiment of the present application. As shown in FIG. 5 , the robot may include an image acquisition device 100 and a processor 200 .

其中，图像采集设备100，用于采集关于机器人所在区域的图像；Among them, the image acquisition device 100 is used to collect images of the area where the robot is located;

处理器200，用于基于图像采集设备100所采集的图像，执行本申请实施例提供的一种重定位方法。The processor 200 is configured to execute a relocation method provided by the embodiment of the present application based on the image collected by the image acquisition device 100 .

在本具体实现方式中，为了机器人进行重定位，图像采集设备100可以采集关于机器人所在区域的图像，之后，将所采集的图像发送给处理器200。这样，处理器200便可以基于上述图像采集设备100所采集的图像，执行本申请实施例提供的一种重定位方法，确定机器人相对于先验地图的重定位位姿。In this specific implementation manner, in order for the robot to reposition, the image acquisition device 100 can collect images about the area where the robot is located, and then send the collected images to the processor 200 . In this way, the processor 200 can execute a relocation method provided by the embodiment of the present application based on the image collected by the above-mentioned image acquisition device 100, and determine the relocation pose of the robot relative to the a priori map.

上文已详细说明了本申请实施例提供的一种重定位方法的具体实现方式，该处理器200即按照如前所述的具体实现方式执行本申请实施例所提供的一种重定位方法的各个步骤，此处不再赘述。The specific implementation of a relocation method provided by the embodiment of the present application has been described in detail. The processor 200 executes the relocation method provided by the embodiment of the present application in accordance with the specific implementation as described above. Each step will not be repeated here.

基于相同的发明构思，相应于上述本申请实施例提供的图3所示的一种重定位方法，本申请实施例还提供了一种重定位装置。Based on the same inventive concept, corresponding to the relocation method shown in Figure 3 provided by the above embodiment of the present application, the embodiment of the present application also provides a relocation device.

图6为本申请实施例提供的一种重定位装置的结构示意图，如图6所示，该装置可以包括如下模块：Figure 6 is a schematic structural diagram of a relocation device provided by an embodiment of the present application. As shown in Figure 6, the device may include the following modules:

初始位姿确定模块610，用于基于目标机器人所搭载图像采集设备所采集的第一帧图像，确定所述目标机器人的初始参考位姿；The initial pose determination module 610 is used to determine the initial reference pose of the target robot based on the first frame of images collected by the image acquisition device mounted on the target robot;

参考位姿确定模块620，用于获取所述图像采集设备所采集的多帧参考图像，并基于所述初始参考位姿，确定所述目标机器人在采集每帧参考图像时的参考位姿，得到每帧参考图像对应的参考位姿；The reference pose determination module 620 is used to obtain multiple frames of reference images collected by the image acquisition device, and based on the initial reference pose, determine the reference pose of the target robot when collecting each frame of reference image, and obtain The reference pose corresponding to each frame of reference image;

局部地图构建模块630，用于利用所述多帧参考图像和每帧参考图像对应的参考位姿，构建所述目标机器人当前所在的区域局部地图，并在预设的先验子地图中，确定与所述区域局部地图模板匹配的目标子地图；其中，所述先验子地图是利用预设的布局子地图对所述目标机器人所属空间的先验地图进行分割得到的，所述布局子地图是按照区域分割线对所述目标机器人所属空间的空间布局图进行分割得到的；The local map construction module 630 is used to construct a local map of the area where the target robot is currently located using the multi-frame reference images and the reference pose corresponding to each frame reference image, and determine in the preset a priori sub-map A target sub-map that matches the regional local map template; wherein the prior sub-map is obtained by segmenting the prior map of the space to which the target robot belongs using a preset layout sub-map, and the layout sub-map It is obtained by dividing the spatial layout diagram of the space where the target robot belongs according to the area dividing line;

重定位模块640，用于对所述图像采集设备所采集的当前图像与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像，并根据所述目标基准图像所关联的目标位姿，确定所述目标机器人相对于所述先验地图的重定位位姿。The relocation module 640 is used to perform similarity matching between the current image collected by the image acquisition device and each reference image associated with the target sub-map, determine the target reference image that matches the current image, and determine the target reference image according to the The target pose associated with the target reference image determines the relocation pose of the target robot relative to the a priori map.

以上可见，应用本申请实施例提供的方案，可以实现对目标机器人的重定位。并且，通过区域局部地图与先验子地图的模板匹配可以提高区域定位的准确性，从而，可以降低因区域误定位而导致的重定位失败的可能性，提高重定位的成功率；进一步的，通过当前图像与目标子地图的各个基准图像的相似度匹配，可以提高重定位的准确性。此外，仅需将当前图像与目标子地图的各个基准图像进行相似度匹配，而无需遍历完整的先验地图所关联的各个基准图像进行相似度匹配，可以减少重定位的计算量，进而，提高重定位效率。It can be seen from the above that by applying the solution provided by the embodiment of the present application, the target robot can be repositioned. Moreover, the accuracy of regional positioning can be improved through template matching between the regional local map and the prior submap, thereby reducing the possibility of relocation failure caused by regional mislocation and improving the success rate of relocation; further, The accuracy of relocalization can be improved by similarity matching between the current image and each reference image of the target submap. In addition, it only needs to perform similarity matching between the current image and each reference image of the target submap, without traversing each reference image associated with the complete prior map for similarity matching, which can reduce the calculation amount of relocation, thereby improving Relocation efficiency.

可选的，一种具体实现方式中，所述图像采集设备为双目相机，所述初始位姿确定模块610，具体用于：Optionally, in a specific implementation, the image acquisition device is a binocular camera, and the initial pose determination module 610 is specifically used to:

基于目标机器人所搭载图像采集设备所采集到第一帧同步的左目图像和右目图像，确定所述目标机器人的初始参考位姿。Based on the synchronized left eye image and right eye image of the first frame collected by the image acquisition equipment mounted on the target robot, the initial reference pose of the target robot is determined.

可选的，一种具体实现方式中，所述参考位姿确定模块620，具体用于：Optionally, in a specific implementation, the reference pose determination module 620 is specifically used to:

获取所述图像采集设备所采集到的多个关键帧，作为多帧参考图像；Obtain multiple key frames collected by the image acquisition device as multi-frame reference images;

其中，所述多个关键帧包括：按照预设时间间隔所采集的多帧图像、按照所述目标机器人的预设移动距离所采集的多帧图像，或者，包括预设图像特征的多帧图像。Wherein, the plurality of key frames include: multiple frames of images collected according to a preset time interval, multiple frames of images collected according to a preset movement distance of the target robot, or multiple frames of images including preset image features. .

可选的，一种具体实现方式中，所述装置还包括先验子地图构建模块，所述先验子地图构建模块，包括：Optionally, in a specific implementation manner, the device further includes a priori sub-map building module, and the prior sub-map building module includes:

获取子模块，用于获取所述先验地图和所述布局子地图；Obtain sub-module, used to obtain the a priori map and the layout sub-map;

先验子地图构建子模块，用于在所述先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的先验子地图。The a priori submap construction submodule is used to perform template matching on each layout submap in the a priori map to obtain the a priori submap matched by the layout submap.

可选的，一种具体实现方式中，所述先验子地图构建子模块，具体用于：Optionally, in a specific implementation manner, the a priori submap constructs a submodule, specifically used for:

在所述先验地图中，对每个布局子地图进行模板匹配，得到该布局子地图所匹配的初始子地图；In the a priori map, template matching is performed on each layout submap to obtain an initial submap matched by the layout submap;

获取每个布局子地图对应的连接关系和每个初始子地图对应的可通行关系，其中，每个布局子地图对应的连接关系为：每个布局子地图所表征区域与该布局子地图所表征区域的可通行通道的连接关系，每个初始子地图对应的可通行关系为：基于构建所述先验地图时，所述目标机器人的移动轨迹确定的，每个初始子地图与相邻初始子地图的可通行关系；Obtain the connection relationship corresponding to each layout submap and the passability relationship corresponding to each initial submap, where the connection relationship corresponding to each layout submap is: the area represented by each layout submap and the area represented by the layout submap The connection relationship of the passable channels in the area. The passability relationship corresponding to each initial sub-map is: based on the movement trajectory of the target robot when constructing the a priori map, each initial sub-map is connected to the adjacent initial sub-map. The accessibility relationship of the map;

针对每个布局子地图，利用该布局子地图对应的连接关系以及该布局子地图对应的初始子地图对应的可通行关系，对该布局子地图对应的初始子地图进行校正，得到该布局子地图所匹配的先验子地图。For each layout submap, use the connection relationship corresponding to the layout submap and the passability relationship corresponding to the initial submap corresponding to the layout submap to correct the initial submap corresponding to the layout submap to obtain the layout submap The matched prior submap.

可选的，一种具体实现方式中，所述局部地图构建模块630，具体用于：Optionally, in a specific implementation, the local map building module 630 is specifically used to:

针对每个先验子地图，对该先验子地图和所述区域局部地图进行模板匹配，并对该先验子地图所匹配的布局子地图和所述区域局部地图进行模板匹配，得到匹配结果；For each prior sub-map, template matching is performed on the prior sub-map and the regional local map, and template matching is performed on the layout sub-map matched by the prior sub-map and the regional local map to obtain a matching result. ;

基于所述匹配结果，在预设的先验子地图中，确定目标子地图。Based on the matching result, the target submap is determined in the preset a priori submap.

可选的，一种具体实现方式中，所述重定位模块640，具体用于：Optionally, in a specific implementation, the relocation module 640 is specifically used to:

对所述图像采集设备所采集的最近的关键帧与所述目标子地图所关联的各个基准图像进行相似度匹配，确定与所述当前图像所匹配的目标基准图像。Similarity matching is performed between the most recent key frame collected by the image acquisition device and each reference image associated with the target sub-map to determine the target reference image that matches the current image.

可选的，一种具体实现方式中，每个基准图像所关联的位姿为：所述图像采集设备采集该基准图像时，所述图像采集设备相对于所述先验地图的位姿；Optionally, in a specific implementation manner, the pose associated with each reference image is: the pose of the image capture device relative to the a priori map when the image capture device captures the reference image;

所述重定位模块640，具体用于：The relocation module 640 is specifically used for:

根据所述图像采集设备与所述目标机器人的相对位姿，对所述目标基准图像所关联的目标位姿进行位姿转换，得到所述目标机器人相对于所述先验地图的重定位位姿。According to the relative pose of the image acquisition device and the target robot, pose conversion is performed on the target pose associated with the target reference image to obtain the repositioned pose of the target robot relative to the a priori map. .

本申请实施例还提供了一种电子设备，如图7所示，包括:An embodiment of the present application also provides an electronic device, as shown in Figure 7, including:

存储器701，用于存放计算机程序；Memory 701, used to store computer programs;

处理器702，用于执行存储器701上所存放的程序时，实现上述本申请实施例提供的任一重定位方法的步骤。The processor 702 is configured to implement the steps of any relocation method provided by the above embodiments of the present application when executing the program stored on the memory 701.

并且上述电子设备还可以包括通信总线和/或通信接口，处理器702、通信接口、存储器701通过通信总线完成相互间的通信。And the above-mentioned electronic device may also include a communication bus and/or a communication interface. The processor 702, the communication interface, and the memory 701 complete communication with each other through the communication bus.

上述电子设备提到的通信总线可以是外设部件互连标准(Peripheral ComponentInterconnect，PCI)总线或扩展工业标准结构(Extended Industry StandardArchitecture，EISA)总线等。该通信总线可以分为地址总线、数据总线、控制总线等。为便于表示，图中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。The communication bus mentioned in the above-mentioned electronic equipment may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into address bus, data bus, control bus, etc. For ease of presentation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

通信接口用于上述电子设备与其他设备之间的通信。The communication interface is used for communication between the above-mentioned electronic devices and other devices.

存储器可以包括随机存取存储器(Random Access Memory，RAM)，也可以包括非易失性存储器(Non-Volatile Memory，NVM)，例如至少一个磁盘存储器。可选的，存储器还可以是至少一个位于远离前述处理器的存储装置。The memory may include random access memory (Random Access Memory, RAM) or non-volatile memory (Non-Volatile Memory, NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located far away from the aforementioned processor.

上述的处理器可以是通用处理器，包括中央处理器(Central Processing Unit，CPU)、网络处理器(Network Processor，NP)等；还可以是数字信号处理器(Digital SignalProcessor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (Network Processor, NP), etc.; it can also be a digital signal processor (Digital SignalProcessor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

在本申请提供的又一实施例中，还提供了一种计算机可读存储介质，该计算机可读存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现上述任一重定位方法的步骤。In yet another embodiment provided by the present application, a computer-readable storage medium is also provided. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, any one of the above relocation methods is implemented. step.

在本申请提供的又一实施例中，还提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述实施例中任一重定位方法。In yet another embodiment provided by this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform any of the relocation methods in the above embodiments.

在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者固态硬盘(Solid StateDisk，SSD)等。In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, e.g., the computer instructions may be transferred from a website, computer, server, or data center Transmission to another website, computer, server or data center by wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more available media integrated. The available media may be magnetic media (eg, floppy disk, hard disk, magnetic tape), optical media (eg, DVD), or solid state drive (Solid State Disk, SSD), etc.

需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations are mutually exclusive. any such actual relationship or sequence exists between them. Furthermore, the terms "comprises," "comprises," or any other variations thereof are intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements, but also those not expressly listed other elements, or elements inherent to the process, method, article or equipment. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article, or apparatus that includes the stated element.

本说明书中的各个实施例均采用相关的方式描述，各个实施例之间相同相似的部分互相参见即可，每个实施例重点说明的都是与其他实施例的不同之处。尤其，对于装置实施例、电子设备实施例、计算机可读存储介质实施例以及计算机程序产品实施例而言，由于其基本相似于方法实施例，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。Each embodiment in this specification is described in a related manner. The same and similar parts between the various embodiments can be referred to each other. Each embodiment focuses on its differences from other embodiments. In particular, for the device embodiments, electronic equipment embodiments, computer-readable storage medium embodiments and computer program product embodiments, since they are basically similar to the method embodiments, the descriptions are relatively simple. For relevant details, please refer to the method embodiments. Partial description is enough.

以上所述仅为本申请的较佳实施例，并非用于限定本申请的保护范围。凡在本申请的精神和原则之内所作的任何修改、等同替换、改进等，均包含在本申请的保护范围内。The above descriptions are only preferred embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this application are included in the protection scope of this application.

Claims (13)

1. A method of repositioning, the method comprising:

determining an initial reference pose of a target robot based on a first frame image acquired by image acquisition equipment carried by the target robot;

acquiring multi-frame reference images acquired by the image acquisition equipment, and determining the reference pose of the target robot when acquiring each frame of reference image based on the initial reference pose to obtain the reference pose corresponding to each frame of reference image;

constructing a local map of the area where the target robot is currently located by utilizing the multi-frame reference image and the reference pose corresponding to each frame reference image, and determining a target sub-map matched with the local map template of the area in a preset priori sub-map; the prior sub-map is obtained by dividing the prior map of the space of the target robot by using a preset layout sub-map, and the layout sub-map is obtained by dividing a space layout diagram of the space of the target robot according to a region dividing line;

And performing similarity matching on the current image acquired by the image acquisition equipment and each reference image associated with the target sub-map, determining a target reference image matched with the current image, and determining the repositioning pose of the target robot relative to the prior map according to the target pose associated with the target reference image.

2. The method of claim 1, wherein the image capture device is a binocular camera, and wherein determining the initial reference pose of the target robot based on the first frame image captured by the image capture device on which the target robot is mounted comprises:

and determining the initial reference pose of the target robot based on the left-eye image and the right-eye image which are acquired by the image acquisition equipment carried by the target robot and are synchronous in a first frame.

3. The method of claim 1, wherein the acquiring the multi-frame reference image acquired by the image acquisition device comprises:

acquiring a plurality of key frames acquired by the image acquisition equipment as multi-frame reference images;

wherein the plurality of key frames comprises: the method comprises the steps of acquiring multiple frames of images according to a preset time interval, acquiring multiple frames of images according to a preset moving distance of the target robot, or acquiring multiple frames of images comprising preset image features.

4. The method of claim 1, wherein the constructing the prior sub-map comprises:

acquiring the prior map and the layout sub-map;

and in the prior map, carrying out template matching on each layout sub-map to obtain the prior sub-map matched with the layout sub-map.

5. The method of claim 4, wherein the performing template matching on each layout sub-map in the prior map to obtain the prior sub-map matched with the layout sub-map comprises:

in the prior map, performing template matching on each layout sub-map to obtain an initial sub-map matched with the layout sub-map;

obtaining a connection relation corresponding to each layout sub-map and a passable relation corresponding to each initial sub-map, wherein the connection relation corresponding to each layout sub-map is as follows: the connection relation between the represented area of each layout sub-map and the passable channel of the represented area of the layout sub-map is that the passable relation corresponding to each initial sub-map is: based on the prior map, determining the passable relation between each initial sub-map and the adjacent initial sub-map according to the moving track of the target robot;

And correcting the initial sub-map corresponding to the layout sub-map by utilizing the connection relation corresponding to the layout sub-map and the passable relation corresponding to the initial sub-map corresponding to the layout sub-map aiming at each layout sub-map to obtain the prior sub-map matched with the layout sub-map.

6. The method of claim 4, wherein determining a target sub-map that matches the regional partial map template in a preset prior sub-map comprises:

for each prior sub-map, carrying out template matching on the prior sub-map and the regional local map, and carrying out template matching on the layout sub-map matched with the prior sub-map and the regional local map to obtain a matching result;

and determining a target sub-map in a preset priori sub-map based on the matching result.

7. A method according to claim 3, wherein said similarity matching of the current image acquired by the image acquisition device with respective reference images associated with the target sub-map, determining a target reference image matched with the current image, comprises:

and performing similarity matching on the latest key frame acquired by the image acquisition equipment and each reference image associated with the target sub-map, and determining a target reference image matched with the current image.

8. The method of any one of claims 1-7, wherein the pose associated with each reference image is: when the image acquisition equipment acquires the reference image, the pose of the image acquisition equipment relative to the prior map;

the determining the repositioning pose of the target robot relative to the prior map according to the target pose associated with the target reference image comprises the following steps:

and according to the relative pose of the image acquisition equipment and the target robot, carrying out pose conversion on the target pose associated with the target reference image to obtain the repositioning pose of the target robot relative to the prior map.

9. A robot, the robot comprising:

an image acquisition device for acquiring an image on an area where the robot is located;

a processor for performing the method of any of claims 1-8 based on the image acquired by the image acquisition device.

10. A relocating device characterised in that the device includes:

the initial pose determining module is used for determining an initial reference pose of the target robot based on a first frame image acquired by image acquisition equipment carried by the target robot;

The reference pose determining module is used for acquiring multi-frame reference images acquired by the image acquisition equipment, determining the reference pose of the target robot when acquiring each frame of reference image based on the initial reference pose, and obtaining the reference pose corresponding to each frame of reference image;

the local map construction module is used for constructing a local map of the area where the target robot is currently located by utilizing the multi-frame reference image and the reference pose corresponding to each frame reference image, and determining a target sub-map matched with the local map template of the area in a preset priori sub-map; the prior sub-map is obtained by dividing the prior map of the space of the target robot by using a preset layout sub-map, and the layout sub-map is obtained by dividing a space layout diagram of the space of the target robot according to a region dividing line;

and the repositioning module is used for carrying out similarity matching on the current image acquired by the image acquisition equipment and each reference image associated with the target sub-map, determining a target reference image matched with the current image, and determining the repositioning pose of the target robot relative to the prior map according to the target pose associated with the target reference image.

11. The apparatus of claim 10, wherein the image acquisition device is a binocular camera, and the initial pose determination module is specifically configured to:

determining an initial reference pose of a target robot based on a left-eye image and a right-eye image which are acquired by image acquisition equipment carried by the target robot and are synchronous in a first frame;

and/or the number of the groups of groups,

the reference pose determining module is specifically configured to:

acquiring a plurality of key frames acquired by the image acquisition equipment as multi-frame reference images;

wherein the plurality of key frames comprises: a plurality of images acquired according to a preset time interval, a plurality of images acquired according to a preset moving distance of the target robot, or a plurality of images including preset image features;

and/or the number of the groups of groups,

the apparatus further includes an a priori sub-map building module, the a priori sub-map building module comprising:

the acquisition sub-module is used for acquiring the prior map and the layout sub-map;

the prior sub-map construction sub-module is used for carrying out template matching on each layout sub-map in the prior map to obtain a prior sub-map matched with the layout sub-map;

And/or the number of the groups of groups,

the prior sub-map construction sub-module is specifically configured to:

in the prior map, performing template matching on each layout sub-map to obtain an initial sub-map matched with the layout sub-map;

obtaining a connection relation corresponding to each layout sub-map and a passable relation corresponding to each initial sub-map, wherein the connection relation corresponding to each layout sub-map is as follows: the connection relation between the represented area of each layout sub-map and the passable channel of the represented area of the layout sub-map is that the passable relation corresponding to each initial sub-map is: based on the prior map, determining the passable relation between each initial sub-map and the adjacent initial sub-map according to the moving track of the target robot;

correcting the initial sub-map corresponding to the layout sub-map by utilizing the connection relation corresponding to the layout sub-map and the passable relation corresponding to the initial sub-map corresponding to the layout sub-map aiming at each layout sub-map to obtain a priori sub-map matched with the layout sub-map;

and/or the number of the groups of groups,

the local map construction module is specifically configured to:

for each prior sub-map, carrying out template matching on the prior sub-map and the regional local map, and carrying out template matching on the layout sub-map matched with the prior sub-map and the regional local map to obtain a matching result;

Determining a target sub-map in a preset priori sub-map based on the matching result;

and/or the number of the groups of groups,

the repositioning module is specifically configured to:

performing similarity matching on the latest key frame acquired by the image acquisition equipment and each reference image associated with the target sub-map, and determining a target reference image matched with the current image;

and/or the number of the groups of groups,

the pose associated with each reference image is: when the image acquisition equipment acquires the reference image, the pose of the image acquisition equipment relative to the prior map;

the repositioning module is specifically configured to:

and according to the relative pose of the image acquisition equipment and the target robot, carrying out pose conversion on the target pose associated with the target reference image to obtain the repositioning pose of the target robot relative to the prior map.

12. An electronic device, comprising:

a memory for storing a computer program;

a processor for implementing the method of any of claims 1-8 when executing a program stored on a memory.

13. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the method of any of claims 1-8.