技术领域Technical Field
本申请涉及自动驾驶领域，特别是涉及一种车辆轨迹的处理方法以及车辆轨迹的处理装置。The present application relates to the field of autonomous driving, and in particular to a method for processing vehicle trajectories and a device for processing vehicle trajectories.
背景技术Background Art
近年来，随着人工智能技术的不断发展，自动驾驶技术在车辆上得到了越来越多的功能性应用。在相关技术中，可以先通过实际驾驶，从起始地到目的地获得一条或者多条实际驾驶轨迹，之后，使得循迹车辆循着实际驾驶轨迹进行自动驾驶。然而，在进行实际驾驶的车辆、以及后续的循迹车辆不同的情况下，循迹车辆有可能不会完全沿实际驾驶轨迹行驶，并且，实际驾驶的车辆在采集实际驾驶轨迹的同时，也可能会产生一定偏差，使得循迹车辆沿实际驾驶轨迹行驶时，会产生一定的安全风险。In recent years, with the continuous development of artificial intelligence technology, autonomous driving technology has been applied to more and more vehicles. In related technologies, one or more actual driving trajectories can be obtained from the starting point to the destination through actual driving, and then the tracking vehicle can be driven automatically along the actual driving trajectory. However, when the vehicle that performs actual driving and the subsequent tracking vehicle are different, the tracking vehicle may not completely follow the actual driving trajectory, and the actual driving vehicle may also produce certain deviations while collecting the actual driving trajectory, which may cause certain safety risks when the tracking vehicle drives along the actual driving trajectory.
发明内容Summary of the invention
本申请实施例提供了一种车辆轨迹的处理方法以及车辆轨迹的处理装置，能够根据实际驾驶轨迹得到适用于不同循迹车辆的自动驾驶轨迹。The embodiments of the present application provide a vehicle trajectory processing method and a vehicle trajectory processing device, which can obtain an automatic driving trajectory suitable for different tracking vehicles according to the actual driving trajectory.
本申请实施例一方面提供了一种车辆轨迹的处理方法，包括：获取多条实际驾驶轨迹，各实际驾驶轨迹包括一个实际采样点集合，每个实际采样点集合包括沿实际驾驶轨迹顺序排列的多个实际采样点；对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，每个控制点集合包括多个控制点；对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹；对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合，每个虚拟采样点集合中包括沿对应的虚拟驾驶轨迹顺序排列的多个虚拟采样点；对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合，中值点集合包括多个中值点；根据中值点集合，得到自动驾驶轨迹。On the one hand, an embodiment of the present application provides a method for processing a vehicle trajectory, including: obtaining multiple actual driving trajectories, each actual driving trajectory includes an actual sampling point set, and each actual sampling point set includes multiple actual sampling points arranged in sequence along the actual driving trajectory; selecting multiple actual sampling points in each actual sampling point set to obtain multiple control point sets, each control point set includes multiple control points; fitting each control point set to obtain multiple virtual driving trajectories; sampling each virtual driving trajectory to obtain multiple virtual sampling point sets, each virtual sampling point set includes multiple virtual sampling points arranged in sequence along the corresponding virtual driving trajectory; medianing multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain a median point set, the median point set includes multiple median points; and obtaining an automatic driving trajectory according to the median point set.
在一些实施例中，获取多条实际驾驶轨迹，包括获取每条实际驾驶轨迹；获取每条实际驾驶轨迹，包括：在驾驶过程中定时采集车辆位置信息，以得到包括时间标记的多个车辆位置信息；根据多个车辆位置信息得到沿实际驾驶轨迹顺序排列的多个实际采集点；根据多个实际采样点得到实际驾驶轨迹。In some embodiments, obtaining multiple actual driving trajectories includes obtaining each actual driving trajectory; obtaining each actual driving trajectory includes: regularly collecting vehicle position information during driving to obtain multiple vehicle position information including time stamps; obtaining multiple actual collection points arranged in sequence along the actual driving trajectory based on the multiple vehicle position information; obtaining the actual driving trajectory based on the multiple actual sampling points.
在一些实施例中，每个实际采样点集合包括N个沿实际驾驶轨迹顺序排列的实际采样点；对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，包括获取每个控制点集合；获取每个控制点集合，包括：选择多个实际采样点中的第i个实际采样点为控制点；计算第i+1个实际采样点与控制点之间的距离；比较距离与第一阈值；在距离大于等于第一阈值的情况下，选择第i+1个实际采样点为控制点；返回计算第i+1个实际采样点与控制点之间的距离，直至i等于N-1。In some embodiments, each actual sampling point set includes N actual sampling points arranged in sequence along the actual driving trajectory; multiple actual sampling points in each actual sampling point set are selected to obtain multiple control point sets, including obtaining each control point set; obtaining each control point set includes: selecting the i-th actual sampling point in the multiple actual sampling points as the control point; calculating the distance between the i+1-th actual sampling point and the control point; comparing the distance with a first threshold; when the distance is greater than or equal to the first threshold, selecting the i+1-th actual sampling point as the control point; returning to calculate the distance between the i+1-th actual sampling point and the control point until i is equal to N-1.
在一些实施例中，在比较距离与第一阈值之后，对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，还包括：在距离小于第一阈值的情况下，删去第i+1个实际采样点，得到更新的实际采样点集合；返回步骤计算第i+1个实际采样点与控制点之间的距离，直至距离大于等于第一阈值。In some embodiments, after comparing the distance with the first threshold, multiple actual sampling points in each actual sampling point set are selected to obtain multiple control point sets, further comprising: when the distance is less than the first threshold, deleting the i+1th actual sampling point to obtain an updated actual sampling point set; and returning to the step of calculating the distance between the i+1th actual sampling point and the control point until the distance is greater than or equal to the first threshold.
在一些实施例中，对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹，包括获取每条虚拟驾驶轨迹；获取每条虚拟驾驶轨迹，包括：对多个控制点进行B样条插值，得到虚拟驾驶轨迹。In some embodiments, fitting each set of control points to obtain multiple virtual driving trajectories includes obtaining each virtual driving trajectory; obtaining each virtual driving trajectory includes: performing B-spline interpolation on multiple control points to obtain the virtual driving trajectory.
在一些实施例中，对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合，包括得到每个虚拟采样点集合；得到每个虚拟采样点集合，包括：获取预设速度以及第一采样时间；根据预设速度以及第一采样时间得到第一采样距离；根据第一采样距离，沿虚拟驾驶轨迹等距离选取多个虚拟采样点。In some embodiments, each virtual driving trajectory is sampled to obtain multiple virtual sampling point sets, including obtaining each virtual sampling point set; obtaining each virtual sampling point set includes: obtaining a preset speed and a first sampling time; obtaining a first sampling distance according to the preset speed and the first sampling time; and selecting multiple virtual sampling points at equal distances along the virtual driving trajectory according to the first sampling distance.
在一些实施例中，得到每个虚拟采样点集合，还包括：计算虚拟采样点处虚拟驾驶轨迹的曲率参数；在曲率参数大于等于第二阈值的情况下，根据第二采样距离，从虚拟采样点向相邻的虚拟采样点进行取样，得到两个补充虚拟采样点；根据虚拟采样点以及两个补充虚拟采样点，得到更新的虚拟采样点集合；其中，第一采样距离大于第二采样距离。In some embodiments, obtaining each virtual sampling point set also includes: calculating the curvature parameter of the virtual driving trajectory at the virtual sampling point; when the curvature parameter is greater than or equal to a second threshold, sampling from the virtual sampling point to an adjacent virtual sampling point according to a second sampling distance to obtain two supplementary virtual sampling points; obtaining an updated virtual sampling point set according to the virtual sampling point and the two supplementary virtual sampling points; wherein the first sampling distance is greater than the second sampling distance.
在一些实施例中，对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合，包括：对多个虚拟采样点集合中顺序相同的多个虚拟采样点的坐标取平均值，得到中值点集合；或对多个虚拟采样点集合中顺序相同的多个虚拟采样点的坐标取中间值，得到中值点集合。In some embodiments, taking the median of multiple virtual sampling points of the same order in multiple virtual sampling point sets to obtain a median point set includes: taking the average value of the coordinates of multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain the median point set; or taking the middle value of the coordinates of multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain the median point set.
本申请实施例另一方面提供了一种车辆轨迹的处理装置，包括：获取模块，用于获取多条实际驾驶轨迹，各实际驾驶轨迹包括一个实际采样点集合，每个实际采样点集合包括沿实际驾驶轨迹顺序排列的多个实际采样点；选择模块，用于对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，每个控制点集合包括多个控制点；拟合模块，用于对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹；采样模块，用于对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合，每个虚拟采样点集合中包括沿对应的虚拟驾驶轨迹顺序排列的多个虚拟采样点；取中模块，用于对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合，中值点集合包括多个中值点；以及处理模块，用于根据中值点集合，得到自动驾驶轨迹。On the other hand, an embodiment of the present application provides a vehicle trajectory processing device, including: an acquisition module, used to acquire multiple actual driving trajectories, each actual driving trajectory includes an actual sampling point set, and each actual sampling point set includes multiple actual sampling points arranged in sequence along the actual driving trajectory; a selection module, used to select multiple actual sampling points in each actual sampling point set to obtain multiple control point sets, each control point set includes multiple control points; a fitting module, used to fit each control point set to obtain multiple virtual driving trajectories; a sampling module, used to sample each virtual driving trajectory to obtain multiple virtual sampling point sets, each virtual sampling point set includes multiple virtual sampling points arranged in sequence along the corresponding virtual driving trajectory; a median module, used to median multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain a median point set, the median point set includes multiple median points; and a processing module, used to obtain an automatic driving trajectory according to the median point set.
本申请实施例再一方面提供了一种电子设备，包括：处理器以及存储有计算机程序指令的存储器；处理器执行计算机程序指令时实现如以上的车辆轨迹的处理方法。On the other hand, an embodiment of the present application provides an electronic device, including: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the vehicle trajectory processing method as described above is implemented.
本申请实施例还一方面提供了一种计算机可读存储介质，计算机可读存储介质上存储有计算机程序指令，计算机程序指令被处理器执行时实现如以上的车辆轨迹的处理方法。On the other hand, an embodiment of the present application provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the vehicle trajectory processing method as described above is implemented.
本申请实施例提供的车辆轨迹的处理方法，通过获取多条实际驾驶轨迹，各实际驾驶轨迹包括一个实际采样点集合，对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹，对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合，对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合，中值点集合包括多个中值点，以及，根据中值点集合，得到自动驾驶轨迹。根据本申请实施例，对多个实际采样点进行选择，得到能够反应实际驾驶轨迹走向的多个控制点，减少了冗余的实际采样点对所得到的虚拟驾驶轨迹的影响，再通过对多条虚拟驾驶轨迹进行采样、以及对多个虚拟采样点集合中顺序相同的多个虚拟采样点取中，得到自动驾驶轨迹，能够提升对车辆轨迹处理的准确性，使得自动驾驶轨迹能够适用于不同循迹车辆，并且提升循迹车辆在自动驾驶过程中的安全性。The vehicle trajectory processing method provided in the embodiment of the present application obtains multiple actual driving trajectories, each of which includes an actual sampling point set, selects multiple actual sampling points in each actual sampling point set to obtain multiple control point sets, fits each control point set to obtain multiple virtual driving trajectories, samples each virtual driving trajectory to obtain multiple virtual sampling point sets, medians multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain a median point set, the median point set includes multiple median points, and obtains an automatic driving trajectory according to the median point set. According to the embodiment of the present application, multiple actual sampling points are selected to obtain multiple control points that can reflect the direction of the actual driving trajectory, reducing the influence of redundant actual sampling points on the obtained virtual driving trajectory, and then by sampling multiple virtual driving trajectories and median multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain an automatic driving trajectory, the accuracy of vehicle trajectory processing can be improved, so that the automatic driving trajectory can be applied to different tracking vehicles, and the safety of the tracking vehicle during the automatic driving process is improved.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
为了更清楚地说明本申请实施例的技术方案，下面将对本申请实施例中所需要使用的附图作简单地介绍，显而易见地，下面所描述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，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
图1是根据本申请一种实施例提供的车辆轨迹的处理装置的结构示意图；FIG1 is a schematic diagram of the structure of a vehicle trajectory processing device provided according to an embodiment of the present application;
图2是根据本申请一种实施例提供的车辆轨迹的处理方法的流程示意图；FIG2 is a flow chart of a method for processing vehicle trajectories according to an embodiment of the present application;
图3示出了图2所示流程示意图中一些步骤的一种示例性的示意图；FIG3 shows an exemplary schematic diagram of some steps in the flowchart shown in FIG2 ;
图4示出了图2所示流程示意图中一些步骤的一种示例性的示意图；FIG4 shows an exemplary schematic diagram of some steps in the flowchart shown in FIG2 ;
图5示出了图2所示流程示意图中一些步骤的一种示例性的示意图；FIG5 shows an exemplary schematic diagram of some steps in the flowchart shown in FIG2 ;
图6示出了图2所示流程示意图中一些步骤的一种示例性的示意图；FIG6 shows an exemplary schematic diagram of some steps in the flowchart shown in FIG2 ;
图7示出了图6所示流程示意图的一种示例性的示意图；FIG. 7 shows an exemplary schematic diagram of the flow diagram shown in FIG. 6 ;
图8示出了图2所示流程示意图中一些步骤的一种示例性的示意图。FIG. 8 shows an exemplary schematic diagram of some steps in the flowchart shown in FIG. 2 .
附图中：In the attached figure:
100-车辆轨迹的处理装置；1-获取模块；2-选择模块；3-拟合模块；4-采样模块；5-取中模块；6-处理模块。100 - vehicle trajectory processing device; 1 - acquisition module; 2 - selection module; 3 - fitting module; 4 - sampling module; 5 - centering module; 6 - processing module.
具体实施方式DETAILED DESCRIPTION
为了能够更清楚地理解本申请的上述目的、特征和优点，下面将对本申请的方案进行进一步描述。需要说明的是，在不冲突的情况下，本申请的实施例及实施例中的特征可以相互组合。In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the scheme of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.
在下面的描述中阐述了很多具体细节以便于充分理解本申请，但本申请还可以采用其他不同于在此描述的方式来实施；显然，说明书中的实施例只是本申请的一部分实施例，而不是全部的实施例。In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only part of the embodiments of the present application, rather than all of the embodiments.
需要注意的是，除非另有说明，本申请实施例使用的技术术语或者科学术语应当为本申请实施例所属领域技术人员所理解的通常意义。It should be noted that, unless otherwise specified, the technical terms or scientific terms used in the embodiments of the present application should have the common meanings understood by technicians in the field to which the embodiments of the present application belong.
此外，技术术语“第一”、“第二”等仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。在本申请实施例的描述中，“多个”的含义是两个以上，除非另有明确具体的限定。In addition, the technical terms "first", "second", etc. are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.
在本申请实施例的描述中，除非另有明确的规定和限定，技术术语“安装”、“相连”、“连接”、“固定”等术语应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或成一体；也可以是机械连接，也可以是电连接；可以是直接相连，也可以通过中间媒介间接相连，可以是两个元件内部的连通或两个元件的相互作用关系。对于本领域的普通技术人员而言，可以根据具体情况理解上述术语在本申请实施例中的具体含义。In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.
通常来说，车辆的自动驾驶通过获取高精度地图数据以及定位数据实现。然而，对于在矿区进行自动驾驶的情况，由于矿区道路环境复杂、且道路环境在不停变化，无法获得行驶区域的高精度地图数据，或者，所获得的高精度地图数据的时效性较差，传统的自动驾驶规划的方式不能应用于矿区环境。因此，在矿区进行自动驾驶的情况下，可以先通过实际驾驶，从起始地到目的地获得一条或者多条实际驾驶轨迹，再通过对实际驾驶轨迹进行处理，得到自动驾驶轨迹。Generally speaking, the autonomous driving of vehicles is achieved by obtaining high-precision map data and positioning data. However, in the case of autonomous driving in mining areas, due to the complex road environment in mining areas and the constant changes in the road environment, it is impossible to obtain high-precision map data of the driving area, or the timeliness of the obtained high-precision map data is poor, and the traditional autonomous driving planning method cannot be applied to the mining environment. Therefore, in the case of autonomous driving in mining areas, one or more actual driving trajectories can be obtained from the starting point to the destination through actual driving, and then the actual driving trajectories can be processed to obtain the autonomous driving trajectory.
鉴于此，本申请实施例提供一种车辆轨迹的处理方法，通过获取多条实际驾驶轨迹，各实际驾驶轨迹包括一个实际采样点集合，对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹，对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合，对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合，中值点集合包括多个中值点，以及，根据中值点集合，得到自动驾驶轨迹。根据本申请实施例，对多个实际采样点进行选择，得到能够反应实际驾驶轨迹走向的多个控制点，减少了冗余的实际采样点对所得到的虚拟驾驶轨迹的影响，再通过对多条虚拟驾驶轨迹进行采样、以及对多个虚拟采样点集合中顺序相同的多个虚拟采样点取中，得到自动驾驶轨迹，能够提升对车辆轨迹处理的准确性，使得自动驾驶轨迹能够适用于不同循迹车辆，并且提升循迹车辆在自动驾驶过程中的安全性。In view of this, the embodiment of the present application provides a method for processing vehicle trajectories, by obtaining multiple actual driving trajectories, each actual driving trajectory includes an actual sampling point set, selecting multiple actual sampling points in each actual sampling point set to obtain multiple control point sets, fitting each control point set to obtain multiple virtual driving trajectories, sampling each virtual driving trajectory to obtain multiple virtual sampling point sets, taking the middle of multiple virtual sampling points in the same order in the multiple virtual sampling point sets to obtain a median point set, the median point set includes multiple median points, and, according to the median point set, obtaining an automatic driving trajectory. According to the embodiment of the present application, multiple actual sampling points are selected to obtain multiple control points that can reflect the direction of the actual driving trajectory, reducing the influence of redundant actual sampling points on the obtained virtual driving trajectory, and then by sampling multiple virtual driving trajectories and taking the middle of multiple virtual sampling points in the same order in the multiple virtual sampling point sets to obtain an automatic driving trajectory, the accuracy of vehicle trajectory processing can be improved, so that the automatic driving trajectory can be applied to different tracking vehicles, and the safety of the tracking vehicle during the automatic driving process is improved.
图1是根据本申请一种实施例提供的车辆轨迹的处理装置100的结构示意图。如图1所示，本申请实施例还提供一种车辆轨迹的处理装置100。车辆轨迹的处理装置100包括获取模块1、选择模块2、拟合模块3、采样模块4、取中模块5以及处理模块6。FIG1 is a schematic diagram of the structure of a vehicle trajectory processing device 100 provided according to an embodiment of the present application. As shown in FIG1 , the embodiment of the present application further provides a vehicle trajectory processing device 100. The vehicle trajectory processing device 100 includes an acquisition module 1, a selection module 2, a fitting module 3, a sampling module 4, a centering module 5, and a processing module 6.
获取模块1用于获取多条实际驾驶轨迹，各实际驾驶轨迹包括一个实际采样点集合，每个实际采样点集合包括沿实际驾驶轨迹顺序排列的多个实际采样点。选择模块2用于对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，每个控制点集合包括多个控制点。拟合模块3用于对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹。采样模块4用于对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合，每个虚拟采样点集合中包括沿对应的虚拟驾驶轨迹顺序排列的多个虚拟采样点。取中模块5用于对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合，中值点集合包括多个中值点。处理模块6用于根据中值点集合，得到自动驾驶轨迹。The acquisition module 1 is used to acquire multiple actual driving trajectories, each of which includes an actual sampling point set, and each actual sampling point set includes multiple actual sampling points arranged in sequence along the actual driving trajectory. The selection module 2 is used to select multiple actual sampling points in each actual sampling point set to obtain multiple control point sets, and each control point set includes multiple control points. The fitting module 3 is used to fit each control point set to obtain multiple virtual driving trajectories. The sampling module 4 is used to sample each virtual driving trajectory to obtain multiple virtual sampling point sets, and each virtual sampling point set includes multiple virtual sampling points arranged in sequence along the corresponding virtual driving trajectory. The median module 5 is used to median multiple virtual sampling points of the same order in multiple virtual sampling point sets to obtain a median point set, and the median point set includes multiple median points. The processing module 6 is used to obtain an automatic driving trajectory based on the median point set.
可选地，车辆轨迹的处理装置100可以安装于循迹车辆，也可以设置于服务器内，本申请实施例对此不做限制。Optionally, the vehicle trajectory processing device 100 can be installed on the tracking vehicle or set in a server, which is not limited in the embodiment of the present application.
图2是根据本申请一种实施例提供的车辆轨迹的处理方法的流程示意图。本申请实施例的车辆轨迹的处理方法，由图1中示出的车辆轨迹的处理装置100实施。如图2所示，车辆轨迹的处理方法包括以下步骤。Fig. 2 is a flow chart of a method for processing vehicle trajectories according to an embodiment of the present application. The method for processing vehicle trajectories of the embodiment of the present application is implemented by the vehicle trajectories processing device 100 shown in Fig. 1. As shown in Fig. 2, the method for processing vehicle trajectories includes the following steps.
S100、获取多条实际驾驶轨迹，各实际驾驶轨迹包括一个实际采样点集合。S100: Acquire multiple actual driving trajectories, each actual driving trajectory including an actual sampling point set.
S200、对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合。S200: Select multiple actual sampling points in each actual sampling point set to obtain multiple control point sets.
S300、对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹。S300: Fit each control point set to obtain multiple virtual driving trajectories.
S400、对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合。S400: Sampling each virtual driving trajectory to obtain multiple virtual sampling point sets.
S500、对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合。S500 , taking the median of multiple virtual sampling points in the same order in multiple virtual sampling point sets to obtain a median point set.
S600、根据中值点集合，得到自动驾驶轨迹。S600: Obtain an automatic driving trajectory according to the median point set.
步骤S100中，每个实际采样点集合包括沿实际驾驶轨迹顺序排列的多个实际采样点。可以视作由多个实际采样点顺序连线而得到实际驾驶轨迹。In step S100, each actual sampling point set includes a plurality of actual sampling points arranged in sequence along the actual driving trajectory, which can be regarded as the actual driving trajectory obtained by sequentially connecting the plurality of actual sampling points.
图3示出了图2所示流程示意图中一些步骤的一种示例性的示意图。如图2所示，本申请一些实施例中，步骤S100中，通过获取每条实际驾驶轨迹获取多条实际驾驶轨迹，获取每条实际驾驶轨迹包括以下步骤。Fig. 3 shows an exemplary schematic diagram of some steps in the flowchart shown in Fig. 2. As shown in Fig. 2, in some embodiments of the present application, in step S100, multiple actual driving trajectories are obtained by obtaining each actual driving trajectory, and obtaining each actual driving trajectory includes the following steps.
步骤S110、在驾驶过程中定时采集车辆位置信息，以得到包括时间标记的多个车辆位置信息。Step S110: Collect vehicle position information at regular intervals during driving to obtain a plurality of vehicle position information including time stamps.
步骤S120、根据多个车辆位置信息得到沿实际驾驶轨迹顺序排列的多个实际采集点。Step S120: obtaining a plurality of actual collection points sequentially arranged along the actual driving trajectory according to the plurality of vehicle position information.
步骤S130、根据多个实际采样点得到实际驾驶轨迹。Step S130: obtaining an actual driving trajectory according to a plurality of actual sampling points.
步骤S110中，在获取模块1设置于实际驾驶的车辆的情况下，可以由获取模块1对实际驾驶车辆位置信息进行采集。车辆位置信息可以是在全球定位系统(GPS，GlobalPositioning System)层面而言的车辆的位置坐标，也可以是在相对起点以及终点而言的车辆的位置坐标。可以定时在车辆沿实际驾驶轨迹行驶的过程中，以一定频率记录车辆的位置坐标，将当前时间生成时间标记，也即在实际驾驶车辆的行驶过程中，以时间标记表示实际驾驶车辆在不同时刻处于实际驾驶轨迹上的位置。可选地，可以以10Hz的频率定时采集车辆位置信息。应理解，获取模块1在采集过程中可能存在随机定位误差，因此，获取模块1进行采集的频率应尽量高，以减少所获得的车辆位置信息不准确的概率。In step S110, when the acquisition module 1 is set to the actual driven vehicle, the acquisition module 1 can collect the actual driving vehicle position information. The vehicle position information can be the position coordinates of the vehicle at the level of the global positioning system (GPS, Global Positioning System), or it can be the position coordinates of the vehicle relative to the starting point and the end point. The position coordinates of the vehicle can be recorded at a certain frequency during the vehicle's driving along the actual driving trajectory, and the current time can be generated as a time mark, that is, during the actual driving of the vehicle, the time mark is used to indicate the position of the actual driving vehicle on the actual driving trajectory at different times. Optionally, the vehicle position information can be collected regularly at a frequency of 10Hz. It should be understood that the acquisition module 1 may have random positioning errors during the acquisition process. Therefore, the frequency of acquisition by the acquisition module 1 should be as high as possible to reduce the probability of inaccurate vehicle position information obtained.
步骤S120中，由于步骤S110中获取了多个车辆位置信息，车辆位置信息可以包括车辆的位置坐标以及记录位置坐标的顺序或时间，可以将车辆位置信息表示在全球定位系统层面的坐标系，或相对起点以及终点而言的坐标系中。例如，使得位置坐标包括在第一方向上的第一坐标以及在第二方向上的第二坐标，第一方向以及第二方向彼此相交，根据第一坐标以及第二坐标，将多个车辆位置信息作为坐标系中的点进行标注，即能够得到多个实际采集点。In step S120, since multiple vehicle position information is obtained in step S110, the vehicle position information may include the position coordinates of the vehicle and the order or time of recording the position coordinates, and the vehicle position information may be represented in a coordinate system at the global positioning system level, or in a coordinate system relative to the starting point and the end point. For example, the position coordinates include a first coordinate in a first direction and a second coordinate in a second direction, and the first direction and the second direction intersect each other. According to the first coordinate and the second coordinate, the multiple vehicle position information are marked as points in the coordinate system, that is, multiple actual collection points can be obtained.
步骤S130中，将在步骤S120中获得的多个实际采集点，以车辆位置信息记录的顺序或时间进行连线。由于在步骤S110中，获取多个车辆位置信息的频率较高，多个实际采集点的排布较为密集，在这种情况下对其进行连线，即能够较为准确地记录实际驾驶轨迹。In step S130, the multiple actual collection points obtained in step S120 are connected in the order or time of vehicle position information recording. Since the frequency of obtaining multiple vehicle position information in step S110 is high, the arrangement of multiple actual collection points is relatively dense. In this case, connecting them can more accurately record the actual driving trajectory.
可选地，步骤S110、步骤S120以及步骤S130可以由车辆轨迹的处理装置100的获取模块1实施，车辆轨迹的处理装置100也可以通过获取模块1从其它装置或设备获取实际驾驶轨迹。例如，实际驾驶的车辆包括采集装置，在由起点至终点行驶的过程中定时采集车辆位置信息，在到达终点后，将所采集的数据，也即包括实际采样点集合的实际驾驶轨迹传输至车辆轨迹的处理装置100的获取模块1。本申请实施例对此不做限制。Optionally, step S110, step S120, and step S130 may be implemented by the acquisition module 1 of the vehicle trajectory processing device 100, and the vehicle trajectory processing device 100 may also acquire the actual driving trajectory from other devices or equipment through the acquisition module 1. For example, the actually driven vehicle includes a collection device, which collects vehicle position information at regular intervals during the process of driving from the starting point to the end point, and after reaching the end point, transmits the collected data, that is, the actual driving trajectory including the actual sampling point set, to the acquisition module 1 of the vehicle trajectory processing device 100. This embodiment of the application is not limited to this.
步骤S200中，每个控制点集合包括多个控制点。多个控制点是对步骤S100中得到的实际采样点集合中的多个实际采样点进行选择得到的，是多个实际采样点中能够反应实际驾驶轨迹走向的坐标点。也即是说，由于获取模块1定时采集车辆位置信息的频率较高，导致多个实际采样点排布较为密集，并非每个实际采样点都对实际驾驶轨迹的走向起到决定作用，因此，由选择模块2对多个实际采样点进行选择，也即对多个实际采样点进行稀疏化处理，以去除冗余的实际采样点，获得多个控制点。In step S200, each control point set includes multiple control points. The multiple control points are obtained by selecting multiple actual sampling points in the actual sampling point set obtained in step S100, and are coordinate points that can reflect the direction of the actual driving trajectory among the multiple actual sampling points. In other words, since the acquisition module 1 regularly collects vehicle position information at a high frequency, the multiple actual sampling points are arranged more densely, and not every actual sampling point plays a decisive role in the direction of the actual driving trajectory. Therefore, the selection module 2 selects multiple actual sampling points, that is, performs sparse processing on the multiple actual sampling points to remove redundant actual sampling points and obtain multiple control points.
图4示出了图2所示流程示意图中一些步骤的一种示例性的示意图。如图4所示，本申请一些实施例中，步骤S200中，通过获取每个控制点集合获取多个控制点集合，获取每个控制点集合包括以下步骤。Fig. 4 shows an exemplary schematic diagram of some steps in the flowchart shown in Fig. 2. As shown in Fig. 4, in some embodiments of the present application, in step S200, multiple control point sets are obtained by obtaining each control point set, and obtaining each control point set includes the following steps.
S210、选择多个实际采样点中的第i个实际采样点为控制点。S210: Select the i-th actual sampling point among multiple actual sampling points as a control point.
S220、计算第i+1个实际采样点与控制点之间的距离。S220, calculating the distance between the (i+1)th actual sampling point and the control point.
S230、比较距离与第一阈值。S230: Compare the distance with a first threshold.
S240、在距离大于等于第一阈值的情况下，选择第i+1个实际采样点为控制点。S240: When the distance is greater than or equal to the first threshold, select the (i+1)th actual sampling point as the control point.
步骤S210中，选择多个实际采样点中的第i个实际采样点为控制点。例如，在第一次循环中，可以选择多个实际采样点中的第一个实际采样点为控制点。In step S210, the ith actual sampling point among the multiple actual sampling points is selected as the control point. For example, in the first cycle, the first actual sampling point among the multiple actual sampling points may be selected as the control point.
步骤S220中，计算第i+1个实际采样点与控制点之间的距离。例如，在第一次循环中，可以计算第二个实际采样点与在步骤S210中选择的控制点的距离，也即根据实际采样点在在全球定位系统层面的坐标系或相对起点以及终点而言的坐标系中的位置坐标，计算第二个实际采样点与第一个实际采样点之间的距离。如此，能够通过多个实际采样点之间的距离进行判断，与所选择的控制点相邻的实际采样点是否能够反应实际驾驶轨迹的走向。In step S220, the distance between the i+1th actual sampling point and the control point is calculated. For example, in the first cycle, the distance between the second actual sampling point and the control point selected in step S210 can be calculated, that is, the distance between the second actual sampling point and the first actual sampling point is calculated based on the position coordinates of the actual sampling point in the coordinate system at the global positioning system level or in the coordinate system relative to the starting point and the end point. In this way, it is possible to judge whether the actual sampling points adjacent to the selected control point can reflect the direction of the actual driving trajectory through the distances between multiple actual sampling points.
步骤S230中，将在步骤S220中获得的控制点与第i+1个实际采样点之间的距离同第一阈值作比较。可选地，第一阈值可以与实际驾驶车辆的车身长度相关。In step S230, the distance between the control point obtained in step S220 and the (i+1)th actual sampling point is compared with a first threshold. Optionally, the first threshold may be related to the body length of the actual driving vehicle.
步骤S240中，在控制点与第i+1个实际采样点之间的距离大于等于第一阈值的情况下，以第i+1个实际采样点为下一个控制点。例如，在第一次循环中，第一个实际采样点为控制点，在第一阈值为车身长度的情况下，第二个实际采样点不处于一个车身长度能够到达的范围，因此，第二个实际采样点能够对驾驶轨迹的走向起到决定作用，以第二个实际采样点为控制点。In step S240, when the distance between the control point and the i+1th actual sampling point is greater than or equal to the first threshold, the i+1th actual sampling point is used as the next control point. For example, in the first cycle, the first actual sampling point is the control point, and when the first threshold is the vehicle body length, the second actual sampling point is not within the range that can be reached by a vehicle body length. Therefore, the second actual sampling point can play a decisive role in the direction of the driving trajectory, and the second actual sampling point is used as the control point.
每个实际采样点集合可被视作包括N个沿实际驾驶轨迹顺序排列的实际采样点。在步骤S240后，返回步骤S220，直至i等于N-1。也即是说，在步骤S210获得第一个控制点之后，在一种情形下，使得步骤S220、步骤S230以及步骤S240为一循环，将第一个实际采样点与第二个实际采样点之间的距离、第二个实际采样点与第三个实际采样点之间的距离……第N-1个实际采样点与第N个实际采样点之间的距离与第一阈值相比较，以得到多个控制点。Each set of actual sampling points can be regarded as including N actual sampling points arranged in sequence along the actual driving trajectory. After step S240, return to step S220 until i equals N-1. That is to say, after the first control point is obtained in step S210, in one case, steps S220, S230 and S240 are a loop, and the distance between the first actual sampling point and the second actual sampling point, the distance between the second actual sampling point and the third actual sampling point...the distance between the N-1th actual sampling point and the Nth actual sampling point are compared with the first threshold to obtain multiple control points.
本申请一些实施例中，在步骤S230之后，步骤S200中，还对每个实际采样点集合执行以下步骤。In some embodiments of the present application, after step S230, in step S200, the following steps are further performed on each actual sampling point set.
S240’、在距离小于第一阈值的情况下，删去第i+1个实际采样点，得到更新的实际采样点集合。S240': When the distance is less than the first threshold, delete the (i+1)th actual sampling point to obtain an updated set of actual sampling points.
步骤S240’中，在控制点与第i+1个实际采样点之间的距离小于第一阈值的情况下，以第i+1个实际采样点为冗余点，从而删去第i+1个实际采样点，使得更新的实际采样点集合包括N-1个实际采样点。例如，在第一次循环中，第一个实际采样点为控制点，在第一阈值为车身长度的情况下，若第二个实际采样点处于一个车身长度能够到达的范围内，则第二个实际采样点未能对实际驾驶轨迹的走向起到决定作用，即可将第二个实际采样点视作冗余点，可以删去第二个实际采样点。In step S240', when the distance between the control point and the i+1th actual sampling point is less than the first threshold, the i+1th actual sampling point is taken as a redundant point, thereby deleting the i+1th actual sampling point, so that the updated actual sampling point set includes N-1 actual sampling points. For example, in the first cycle, the first actual sampling point is the control point, and when the first threshold is the vehicle body length, if the second actual sampling point is within the range that can be reached by a vehicle body length, the second actual sampling point fails to play a decisive role in the direction of the actual driving trajectory, and the second actual sampling point can be regarded as a redundant point and can be deleted.
在步骤S240’后，返回步骤S220，直至控制点与第i+1个实际采样点之间的距离大于等于第一阈值。例如，在删去第二个实际采样点之后，得到更新的实际采样点集合，此时，更新的实际采样点集合中的第二个实际采样点也即为原实际采样点集合中第三个实际采样点，之后，返回步骤S220，再次计算第一个实际采样点与第二个实际采样点之间的距离，以此为循环得到下一个能够对驾驶轨迹的走向起到决定作用的控制点。也即是说，在步骤S210获得第一个控制点之后，在另一种情形下，使得步骤S220、步骤S230以及步骤S240’为一循环。After step S240', return to step S220 until the distance between the control point and the i+1th actual sampling point is greater than or equal to the first threshold. For example, after deleting the second actual sampling point, an updated set of actual sampling points is obtained. At this time, the second actual sampling point in the updated set of actual sampling points is also the third actual sampling point in the original set of actual sampling points. After that, return to step S220, calculate the distance between the first actual sampling point and the second actual sampling point again, and use this as a loop to obtain the next control point that can play a decisive role in the direction of the driving trajectory. That is to say, after obtaining the first control point in step S210, in another case, steps S220, S230 and S240' are made into a loop.
步骤S300中，拟合是一种把平面上一系列的点，用一条光滑的曲线连接起来的方法。由拟合模块3对控制点集合进行拟合，能够得到虚拟驾驶轨迹。In step S300, fitting is a method of connecting a series of points on a plane with a smooth curve. The fitting module 3 fits the control point set to obtain a virtual driving trajectory.
图5示出了图2所示流程示意图中一些步骤的一种示例性的示意图。如图5所示，本申请一些实施例中，步骤S300中，通过获取每条虚拟驾驶轨迹获取多条虚拟驾驶轨迹，获取每条虚拟驾驶轨迹包括以下步骤。Fig. 5 shows an exemplary schematic diagram of some steps in the flowchart shown in Fig. 2. As shown in Fig. 5, in some embodiments of the present application, in step S300, multiple virtual driving trajectories are obtained by obtaining each virtual driving trajectory, and obtaining each virtual driving trajectory includes the following steps.
S310、对多个控制点进行B样条插值，得到虚拟驾驶轨迹。S310 , performing B-spline interpolation on multiple control points to obtain a virtual driving trajectory.
步骤S310中，在本实施例中，通过B样条插值的方法，将所得到的多个控制点以光滑曲线连接，从而得到虚拟驾驶轨迹。B样条插值是通过多个点的权重递归得到光滑曲线的方法，本申请实施例在此不做赘述。In step S310, in this embodiment, the obtained multiple control points are connected with a smooth curve by a B-spline interpolation method to obtain a virtual driving trajectory. B-spline interpolation is a method of recursively obtaining a smooth curve by weighting multiple points, and the embodiment of the present application is not described in detail here.
步骤S400中，通过采样模块4对虚拟驾驶轨迹进行采样，使得每个虚拟采样点集合中包括沿对应的虚拟驾驶轨迹顺序排列的多个虚拟采样点。In step S400 , the virtual driving trajectory is sampled by the sampling module 4 , so that each virtual sampling point set includes a plurality of virtual sampling points sequentially arranged along the corresponding virtual driving trajectory.
图6示出了图2所示流程示意图中一些步骤的一种示例性的示意图。如图6所示，本申请一些实施例中，步骤S400中，通过得到每个虚拟采样点集合得到多个虚拟采样点集合，得到每个虚拟采样点集合包括以下步骤。Fig. 6 shows an exemplary schematic diagram of some steps in the flowchart shown in Fig. 2. As shown in Fig. 6, in some embodiments of the present application, in step S400, multiple virtual sampling point sets are obtained by obtaining each virtual sampling point set, and obtaining each virtual sampling point set includes the following steps.
S410、获取预设速度以及第一采样时间。S410: Obtain a preset speed and a first sampling time.
S420、根据预设速度以及第一采样时间得到第一采样距离。S420: Obtain a first sampling distance according to a preset speed and a first sampling time.
S430、根据第一采样距离，沿虚拟驾驶轨迹等距离选取多个虚拟采样点。S430: Select multiple virtual sampling points at equal distances along the virtual driving trajectory according to the first sampling distance.
步骤S410中，可以预先设定预设速度以及第一采样时间。预设速度可以是在后续自动驾驶过程中对自动驾驶车辆的规定速度。并且，可以对自动驾驶车辆沿虚拟驾驶轨迹行驶的过程进行模拟，与获取装置对实际驾驶轨迹进行采样相类似地，以第一采样时间为间隔对虚拟驾驶轨迹进行采样。可选地，第一采样时间可以大于获取装置对实际驾驶轨迹进行采样的时间。In step S410, a preset speed and a first sampling time may be preset. The preset speed may be a prescribed speed for the autonomous driving vehicle in the subsequent autonomous driving process. In addition, the process of the autonomous driving vehicle driving along the virtual driving trajectory may be simulated, and the virtual driving trajectory may be sampled at intervals of the first sampling time, similar to the sampling of the actual driving trajectory by the acquisition device. Optionally, the first sampling time may be greater than the time for the acquisition device to sample the actual driving trajectory.
步骤S420中，根据预设速度以及第一采样时间得到第一采样距离。通过预设速度以及第一采样时间相乘得到第一采样距离。In step S420, a first sampling distance is obtained according to the preset speed and the first sampling time. The first sampling distance is obtained by multiplying the preset speed and the first sampling time.
步骤S430中，根据第一采样距离，在虚拟驾驶轨迹上等距离选取多个虚拟采样点。可选地，虚拟采样点也可以包括在模拟车辆在全球定位系统层面的坐标系或相对起点以及终点而言的坐标系中的虚拟位置坐标，以及沿虚拟驾驶轨迹的取样顺序。In step S430, multiple virtual sampling points are selected at equal distances on the virtual driving trajectory according to the first sampling distance. Optionally, the virtual sampling points may also include virtual position coordinates of the simulated vehicle in a coordinate system at the global positioning system level or in a coordinate system relative to the starting point and the end point, and a sampling order along the virtual driving trajectory.
在一些可选的实施例中，由于等距离对虚拟驾驶轨迹进行采样，可能不能完全覆盖所有的对虚拟驾驶轨迹具有决定作用的控制点，为了使得虚拟驾驶轨迹能够尽量贴合实际驾驶轨迹，可以对所选取的多个虚拟采样点进行进一步采样。In some optional embodiments, since the virtual driving trajectory is sampled at equal distances, it may not be possible to completely cover all control points that are decisive for the virtual driving trajectory. In order to make the virtual driving trajectory fit the actual driving trajectory as closely as possible, the selected multiple virtual sampling points may be further sampled.
图7示出了图6所示流程示意图的一种示例性的示意图。如图7所示，步骤S400中，得到每个虚拟采样点集合还可以包括以下步骤。Fig. 7 shows an exemplary schematic diagram of the flow diagram shown in Fig. 6. As shown in Fig. 7, in step S400, obtaining each virtual sampling point set may further include the following steps.
S440、计算虚拟采样点处虚拟驾驶轨迹的曲率参数。S440: Calculate the curvature parameter of the virtual driving trajectory at the virtual sampling point.
S450、在曲率参数大于等于第二阈值的情况下，根据第二采样距离，从虚拟采样点向相邻的虚拟采样点进行取样，得到两个补充虚拟采样点。S450: When the curvature parameter is greater than or equal to the second threshold, sampling is performed from the virtual sampling point to an adjacent virtual sampling point according to the second sampling distance to obtain two supplementary virtual sampling points.
S460、根据虚拟采样点以及两个补充虚拟采样点，得到更新的虚拟采样点集合。S460: Obtain an updated virtual sampling point set according to the virtual sampling point and the two supplementary virtual sampling points.
其中，第一采样距离大于第二采样距离。The first sampling distance is greater than the second sampling distance.
步骤S440中，根据在全球定位系统层面的坐标系或相对起点以及终点而言的坐标系中，虚拟采样点的位置坐标以及虚拟驾驶轨迹的形状，计算得到虚拟驾驶轨迹在该点的曲率参数。In step S440 , the curvature parameter of the virtual driving track at the point is calculated based on the position coordinates of the virtual sampling point and the shape of the virtual driving track in the coordinate system at the global positioning system level or the coordinate system relative to the starting point and the end point.
步骤S450中，在曲率参数大于等于第二阈值的情况下，说明当前虚拟采样点所在的一段虚拟驾驶轨迹较为曲折，导致单独的虚拟采样点不能对虚拟驾驶轨迹起到决定作用。因此，可以根据第二采样距离从虚拟采样点向两侧进行取样，得到补充虚拟采样点。可选地，使得第二采样距离小于第一采样距离，能够保证补充虚拟采样点位于当前虚拟采样点与相邻的虚拟采样点之间，以决定当前虚拟采样点两侧的虚拟驾驶轨迹的走向。In step S450, when the curvature parameter is greater than or equal to the second threshold, it means that the virtual driving track where the current virtual sampling point is located is relatively tortuous, resulting in that the single virtual sampling point cannot play a decisive role in the virtual driving track. Therefore, sampling can be performed from both sides of the virtual sampling point according to the second sampling distance to obtain a supplementary virtual sampling point. Optionally, making the second sampling distance smaller than the first sampling distance can ensure that the supplementary virtual sampling point is located between the current virtual sampling point and the adjacent virtual sampling point to determine the direction of the virtual driving track on both sides of the current virtual sampling point.
步骤S460中，在得到补充虚拟采样点后，使得多个虚拟采样点以及两个补充虚拟采样点共同作为虚拟采样点集合中的多个虚拟采样点，以对虚拟采样点进行更新。In step S460, after the supplementary virtual sampling points are obtained, the multiple virtual sampling points and the two supplementary virtual sampling points are used together as multiple virtual sampling points in the virtual sampling point set to update the virtual sampling points.
可选地，在步骤S460之后，可以进一步对多个补充虚拟采样点处虚拟驾驶轨迹的曲率参数进行计算，并在曲率参数仍大于等于第二阈值的情况下，再次从补充虚拟采样点向两侧进行取样，直至所得到的虚拟采样点以及补充虚拟采样点能够决定虚拟驾驶轨迹的走向。Optionally, after step S460, the curvature parameters of the virtual driving trajectory at multiple supplementary virtual sampling points may be further calculated, and when the curvature parameters are still greater than or equal to the second threshold, sampling may be performed again from the supplementary virtual sampling points to both sides until the obtained virtual sampling points and the supplementary virtual sampling points are able to determine the direction of the virtual driving trajectory.
步骤S500中，由于每个虚拟采样点集合中包括沿对应的虚拟驾驶轨迹顺序排列的多个虚拟采样点，在包括多个虚拟采样点集合的情况下，通过取中模块5对每个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到中值点集合，中值点集合包括多个中值点。In step S500, since each virtual sampling point set includes multiple virtual sampling points arranged in sequence along the corresponding virtual driving trajectory, in the case of including multiple virtual sampling point sets, the multiple virtual sampling points in the same order in each virtual sampling point set are centered by the centering module 5 to obtain a median point set, and the median point set includes multiple median points.
图8示出了图2所示流程示意图中一些步骤的一种示例性的示意图。如图8所示，在一些可选的实施例中，步骤S500包括以下步骤。Fig. 8 shows an exemplary schematic diagram of some steps in the flowchart shown in Fig. 2. As shown in Fig. 8, in some optional embodiments, step S500 includes the following steps.
S510、对多个虚拟采样点集合中顺序相同的多个虚拟采样点的坐标取平均值，得到中值点集合。S510 , taking the average value of the coordinates of multiple virtual sampling points in the same order in multiple virtual sampling point sets to obtain a median point set.
步骤S510中，由于多条虚拟驾驶路线皆从起点至终点延伸，且多个虚拟采样点的采样规则都是相同的，因此，多个虚拟采样点集合中顺序相同的多个虚拟采样点应彼此接近。对顺序相同的多个虚拟采样点的虚拟位置坐标进行取平均值，能够得到一个中值点，在虚拟采样点集合中包括沿对应的虚拟驾驶轨迹顺序排列的多个虚拟采样点的情况下，能够得到多个中值点，也即中值点集合。In step S510, since the multiple virtual driving routes all extend from the starting point to the end point, and the sampling rules of the multiple virtual sampling points are the same, the multiple virtual sampling points of the same order in the multiple virtual sampling point sets should be close to each other. By taking the average of the virtual position coordinates of the multiple virtual sampling points of the same order, a median point can be obtained. In the case where the virtual sampling point set includes multiple virtual sampling points arranged in sequence along the corresponding virtual driving trajectory, multiple median points, that is, a median point set, can be obtained.
在另一些可选的实施例中，步骤S500中包括以下步骤。In some other optional embodiments, step S500 includes the following steps.
S510’、对多个虚拟采样点集合中顺序相同的多个虚拟采样点的坐标取中间值，得到中值点集合。S510', taking the middle value of the coordinates of multiple virtual sampling points of the same order in multiple virtual sampling point sets to obtain a median point set.
步骤S510’中，与步骤S510相类似地，对顺序相同的多个虚拟采样点的虚拟位置坐标进行取中间值，也能够获取多个中值点。In step S510', similar to step S510, the virtual position coordinates of multiple virtual sampling points in the same order are taken as the middle value, and multiple median points can also be obtained.
步骤S600中，由处理模块6对多个中值点进行连线或拟合处理，得到自动驾驶轨迹。经过以上步骤，能够根据实际驾驶轨迹得到适用于不同循迹车辆的自动驾驶轨迹，提升自动驾驶轨迹的准确性，并且提升自动驾驶过程的安全性。In step S600, the processing module 6 connects or fits the multiple median points to obtain the automatic driving trajectory. After the above steps, the automatic driving trajectory suitable for different tracking vehicles can be obtained according to the actual driving trajectory, thereby improving the accuracy of the automatic driving trajectory and the safety of the automatic driving process.
基于以上描述，本申请实施例还提供一种电子设备，电子设备包括处理器以及存储有计算机程序指令的存储器。处理器执行计算机程序指令时实现如上的车辆轨迹的处理方法。Based on the above description, an embodiment of the present application further provides an electronic device, the electronic device comprising a processor and a memory storing computer program instructions. When the processor executes the computer program instructions, the above vehicle trajectory processing method is implemented.
处理器可以包括中央处理器(CPU)，或者特定集成电路(Application SpecificIntegrated Circuit，ASIC)，或者可以被配置成实施本申请实施例的一个或多个集成电路。The processor may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
存储器可以包括用于数据或指令的大容量存储器。例如，存储器可包括硬盘驱动器(Hard Disk Drive，HDD)、软盘驱动器、闪存、光盘、磁光盘、磁带或通用串行总线(Universal Serial Bus，USB)驱动器或者以上两个或更多个的组合。在适宜的情况下，存储器可包括可移除或不可移除(或固定)的介质。在适宜的情况下，存储器可在综合网关容灾设备的内部或外部。在特定实施例中，存储器是非易失性固态存储器。The memory may include a large capacity memory for data or instructions. For example, the memory may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of the above. Where appropriate, the memory may include a removable or non-removable (or fixed) medium. Where appropriate, the memory may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory is a non-volatile solid-state memory.
存储器可以包括只读存储器(ROM)，随机存取存储器(RAM)，磁盘存储介质设备，光存储介质设备，闪存设备，电气、光学或其他物理/有形的存储器存储设备。因此，通常，存储器包括一个或多个编码有包括计算机可执行指令的软件的有形(非暂态)计算机可读存储介质(例如，存储器设备)，并且当该软件被执行(例如，由一个或多个处理器)时，其可操作来执行参考根据本申请实施例的一方面的方法所描述的操作。The memory may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical/tangible memory storage device. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of an embodiment of the present application.
另外，结合以上描述，本申请实施例还提供一种计算机存储介质。计算机可读存储介质上存储有计算机程序指令，计算机程序指令被处理器执行时，能够实现如上的车辆轨迹的处理方法。In addition, in combination with the above description, the embodiment of the present application further provides a computer storage medium. The computer readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the above vehicle trajectory processing method can be implemented.
综上所述，本申请实施例提供的车辆轨迹的处理方法，通过获取多条实际驾驶轨迹，各实际驾驶轨迹包括一个实际采样点集合，对每个实际采样点集合中的多个实际采样点进行选择，得到多个控制点集合，对每个控制点集合进行拟合，得到多条虚拟驾驶轨迹，对每条虚拟驾驶轨迹进行采样，得到多个虚拟采样点集合，对多个虚拟采样点集合中顺序相同的多个虚拟采样点进行取中，得到一个中值点集合，中值点集合包括多个中值点，以及，根据中值点集合，得到自动驾驶轨迹。根据本申请实施例，对多个实际采样点进行选择，得到能够反应实际驾驶轨迹走向的多个控制点，减少了冗余的实际采样点对所得到的虚拟驾驶轨迹的影响，再通过对多条虚拟驾驶轨迹进行采样、以及对多个虚拟采样点集合中顺序相同的多个虚拟采样点取中，得到自动驾驶轨迹，能够提升对车辆轨迹处理的准确性，使得自动驾驶轨迹能够适用于不同循迹车辆，并且提升循迹车辆在自动驾驶过程中的安全性。In summary, the vehicle trajectory processing method provided by the embodiment of the present application obtains multiple actual driving trajectories, each of which includes an actual sampling point set, selects multiple actual sampling points in each actual sampling point set to obtain multiple control point sets, fits each control point set to obtain multiple virtual driving trajectories, samples each virtual driving trajectory to obtain multiple virtual sampling point sets, medians multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain a median point set, the median point set includes multiple median points, and, according to the median point set, obtains an automatic driving trajectory. According to the embodiment of the present application, multiple actual sampling points are selected to obtain multiple control points that can reflect the direction of the actual driving trajectory, reducing the influence of redundant actual sampling points on the obtained virtual driving trajectory, and then by sampling multiple virtual driving trajectories and median multiple virtual sampling points of the same order in the multiple virtual sampling point sets to obtain an automatic driving trajectory, the accuracy of vehicle trajectory processing can be improved, so that the automatic driving trajectory can be applied to different tracking vehicles, and the safety of the tracking vehicle during the automatic driving process is improved.
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”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 any such actual relationship or order between these entities or operations. Moreover, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211620639.0ACN116012972B (en) | 2022-12-15 | 2022-12-15 | Vehicle track processing method and vehicle track processing device |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211620639.0ACN116012972B (en) | 2022-12-15 | 2022-12-15 | Vehicle track processing method and vehicle track processing device |
| Publication Number | Publication Date |
|---|---|
| CN116012972A CN116012972A (en) | 2023-04-25 |
| CN116012972Btrue CN116012972B (en) | 2024-10-29 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211620639.0AActiveCN116012972B (en) | 2022-12-15 | 2022-12-15 | Vehicle track processing method and vehicle track processing device |
| Country | Link |
|---|---|
| CN (1) | CN116012972B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119884267A (en)* | 2024-12-27 | 2025-04-25 | 柳州五菱新能源汽车有限公司 | High-precision map data processing method, device, storage medium and processor |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107346134A (en)* | 2017-08-29 | 2017-11-14 | 北京矿冶研究总院 | Unmanned control method and device for underground mining articulated vehicle |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8126642B2 (en)* | 2008-10-24 | 2012-02-28 | Gray & Company, Inc. | Control and systems for autonomously driven vehicles |
| CN112572416A (en)* | 2020-11-20 | 2021-03-30 | 芜湖格陆博智能科技有限公司 | Parking method and system for unmanned vehicle in mining area |
| US12014126B2 (en)* | 2020-12-23 | 2024-06-18 | Woven By Toyota, U.S., Inc. | Generating accurate and diverse simulations for evaluation of autonomous-driving systems |
| CN115346288A (en)* | 2021-05-12 | 2022-11-15 | 前海七剑科技(深圳)有限公司 | Simulation driving record acquisition method and system, electronic equipment and storage medium |
| CN114664094B (en)* | 2022-05-26 | 2022-08-26 | 青岛慧拓智能机器有限公司 | Vehicle track prediction method, device, chip, terminal and computer equipment |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107346134A (en)* | 2017-08-29 | 2017-11-14 | 北京矿冶研究总院 | Unmanned control method and device for underground mining articulated vehicle |
| Publication number | Publication date |
|---|---|
| CN116012972A (en) | 2023-04-25 |
| Publication | Publication Date | Title |
|---|---|---|
| CN113494917A (en) | Map construction method and system, method for making navigation strategy and storage medium | |
| CN112013862B (en) | A Crowdsourcing Trajectory-Based Pedestrian Network Extraction and Update Method | |
| CN108052101B (en) | Robot repositioning method and device | |
| CN116012972B (en) | Vehicle track processing method and vehicle track processing device | |
| AU2019433004B2 (en) | Method and apparatus for seeking a delivery locker, device, and storage medium | |
| CN111024079B (en) | Method and system for matching with route according to multiple position points | |
| CN109031372B (en) | Method for automatically extracting key points of driving route from satellite positioning data | |
| CN104977016B (en) | Navigation processing method and mobile intelligent terminal | |
| CN116363856A (en) | A method and device for generating road intersection topology | |
| CN116883965A (en) | Lane marking method and device, electronic equipment and storage medium | |
| CN118277749A (en) | Method, device, equipment, medium and vehicle for calculating elevation confidence of road network data | |
| CN110503663B (en) | A random multi-target automatic detection and tracking method based on frame sampling detection | |
| CN104076375A (en) | Path search method and path search device | |
| CN118585832A (en) | Vehicle trajectory identification method, device, equipment, storage medium and program product | |
| CN113484843A (en) | Method and device for determining external parameters between laser radar and integrated navigation | |
| CN118549939A (en) | Method for global positioning of vehicle based on laser radar | |
| CN118082515A (en) | Vehicle battery abnormality identification method, device, vehicle and storage medium | |
| CN114993331B (en) | Route planning method, electronic device and storage medium | |
| CN118795505A (en) | Vehicle trajectory generation method, device and electronic device | |
| CN115082629B (en) | Environmental map construction method and robot using the environmental map construction method | |
| CN115167402A (en) | Method and system for pathfinding of a weeding robot | |
| CN114935773A (en) | Online road matching method, device, medium and equipment based on Beidou positioning | |
| CN117419732B (en) | Road network-based perception target positioning deviation rectifying method, equipment and storage medium | |
| CN114494724B (en) | River crack extraction method and device | |
| CN119383564B (en) | Method, device, equipment and computer storage medium for determining road trajectory |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |