技术领域Technical field
本申请实施例涉及机器人控制技术领域，特别涉及一种四足机器人的控制方法、装置、机器人及存储介质。Embodiments of the present application relate to the field of robot control technology, and in particular to a control method, device, robot and storage medium for a quadruped robot.
背景技术Background technique
随着人工智能技术的发展，一些组织和科研机构相继推出了诸如双足机器人、四足机器人等多款能够实现自动行走的机器人。With the development of artificial intelligence technology, some organizations and scientific research institutions have successively launched a variety of robots that can achieve autonomous walking, such as bipedal robots and quadrupedal robots.
以四足机器人为例，目前对于四足机器人的研究大多侧重于本体结构设计，对于其运动控制的研究相对较少，从而导致目前的四足机器人基本只具备在简单的环境中四足爬行的运动能力。Take quadruped robots as an example. Most of the current research on quadruped robots focuses on the design of the body structure, and there is relatively little research on its motion control. As a result, the current quadruped robots basically only have the ability to crawl on four legs in a simple environment. Athletic ability.
在较为复杂的环境中，四足机器人如何稳定运动，是当前亟待解决的重要技术问题。How a quadruped robot can move stably in a more complex environment is an important technical issue that needs to be solved urgently.
发明内容Contents of the invention
本申请提供了一种四足机器人的控制方法、装置、机器人及存储介质，能够实现四足机器人在复杂环境中的进行随机落脚运动。所述技术方案如下：This application provides a control method, device, robot and storage medium for a quadruped robot, which can realize random landing motion of the quadruped robot in a complex environment. The technical solutions are as follows:
根据本申请的一方面，提供了一种四足机器人的控制方法，所述方法包括：According to one aspect of the present application, a control method for a quadruped robot is provided. The method includes:
获取所述四足机器人所处环境的地形信息，所述地形信息包括若干个候选落脚点；Obtain terrain information of the environment where the quadruped robot is located, where the terrain information includes several candidate footholds;
在所述若干个候选落脚点中确定若干个期望落脚点，所述期望落脚点是支撑所述四足机器人在所述环境中运动的落脚点；Determine several desired footholds among the several candidate footholds, where the desired footholds are footholds that support the movement of the quadruped robot in the environment;
控制所述若干个关节中的至少一个关节的力矩，以使得所述四足机器人在所述若干个期望落脚点上进行随机落脚运动，所述随机落脚运动中的所述期望落脚点具有随机性或无规律性。Control the moment of at least one joint among the several joints so that the quadruped robot performs random landing motion on the several desired landing points, and the desired landing point in the random landing motion has randomness Or irregular.
根据本申请的一方面，提供了一种四足机器人的控制装置，所述装置包括：According to one aspect of the present application, a control device for a quadruped robot is provided, and the device includes:
获取模块，用于获取所述四足机器人所处环境的地形信息，所述地形信息包括若干个候选落脚点；An acquisition module, used to acquire terrain information of the environment where the quadruped robot is located, where the terrain information includes several candidate footholds;
确定模块，用于在所述若干个候选落脚点中确定若干个期望落脚点，所述期望落脚点是支撑所述四足机器人在所述环境中运动的落脚点；A determination module, configured to determine several desired footholds among the several candidate footholds, where the desired footholds are footholds that support the movement of the quadruped robot in the environment;
控制模块，用于控制所述若干个关节中的至少一个关节的力矩，以使得所述四足机器人在所述若干个期望落脚点上进行随机落脚运动，所述随机落脚运动中的所述期望落脚点具有随机性或无规律性。A control module, configured to control the torque of at least one joint among the several joints, so that the quadruped robot performs a random landing motion on the several desired landing points, and the desired landing point in the random landing motion is The footholds are random or irregular.
根据本申请的另一方面，提供了一种四足机器人，所述四足机器人包括处理器和存储器，所述存储器中存储有至少一条指令，所述至少一条指令由所述处理器加载并执行以实现如上所述的四足机器人的控制方法。According to another aspect of the present application, a quadruped robot is provided. The quadruped robot includes a processor and a memory. At least one instruction is stored in the memory. The at least one instruction is loaded and executed by the processor. To realize the control method of the quadruped robot as mentioned above.
根据本申请的另一方面，提供了一种计算机存储介质，计算机可读存储介质中存储有至少一条计算机程序，至少一条计算机程序由处理器加载并执行以实现如上方面所述的四足机器人的控制方法。According to another aspect of the present application, a computer storage medium is provided. The computer-readable storage medium stores at least one computer program. The at least one computer program is loaded and executed by a processor to implement the quadruped robot as described above. Control Method.
根据本申请的另一方面，提供了一种计算机程序产品，上述计算机程序产品包括计算机程序，所述计算机程序存储在计算机可读存储介质中；所述计算机程序由计算机设备的处理器从所述计算机可读存储介质读取并执行，使得所述计算机设备执行如上方面所述的四足机器人的控制方法。According to another aspect of the present application, a computer program product is provided. The computer program product includes a computer program, and the computer program is stored in a computer-readable storage medium; the computer program is obtained by a processor of a computer device from the computer program. The computer-readable storage medium is read and executed, so that the computer device executes the control method of the quadruped robot as described above.
根据本申请的另一方面，提供了一种芯片，所述芯片包括可编程逻辑电路或程序，安装有所述芯片的设备用于实现如上所述的四足机器人的控制方法。According to another aspect of the present application, a chip is provided. The chip includes a programmable logic circuit or program, and a device equipped with the chip is used to implement the control method of a quadruped robot as described above.
本申请提供的技术方案带来的有益效果至少包括：The beneficial effects brought by the technical solution provided by this application at least include:
通过获取四足机器人所处环境的地形信息，在若干个候选落脚点中确定若干个支撑四足机器人在环境中运动的期望落脚点；控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上进行随机落脚运动。本申请基于四足机器人所处环境的地形信息，使得四足机器人能够在复杂环境下稳定的实现随机落脚运动，提升了四足机器人在复杂环境下运动的稳定性以及展示效果。By obtaining the terrain information of the environment where the quadruped robot is located, several desired footholds are determined among several candidate footholds to support the movement of the quadruped robot in the environment; the moment of at least one joint among the several joints is controlled so that the four-legged robot can move in the environment. The foot robot performs random landing motion on several desired landing points. This application is based on the terrain information of the environment where the quadruped robot is located, so that the quadruped robot can stably realize random landing motion in a complex environment, and improves the stability and display effect of the quadruped robot's movement in complex environments.
附图说明Description of drawings
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例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，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 in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without exerting creative efforts.
图1是本申请一个示例性实施例提供的四足机器人的控制方法的示意图；Figure 1 is a schematic diagram of a control method for a quadruped robot provided by an exemplary embodiment of the present application;
图2是本申请一个示例性实施例提供的机器人控制框架图；Figure 2 is a robot control framework diagram provided by an exemplary embodiment of the present application;
图3是本申请一个示例性实施例提供的四足机器人的结构示意图；Figure 3 is a schematic structural diagram of a quadruped robot provided by an exemplary embodiment of the present application;
图4是本申请一个示例性实施例提供的四足机器人的控制方法的流程图；Figure 4 is a flow chart of a control method for a quadruped robot provided by an exemplary embodiment of the present application;
图5是本申请一个示例性实施例提供的四足机器人的控制方法的流程图；Figure 5 is a flow chart of a control method for a quadruped robot provided by an exemplary embodiment of the present application;
图6是本申请一个示例性实施例提供的梅花桩环境的示意图；Figure 6 is a schematic diagram of the plum blossom pile environment provided by an exemplary embodiment of the present application;
图7是本申请一个示例性实施例提供的四足机器人在图6所示的梅花桩环境中四足偏移跳舞的俯视示意图；Figure 7 is a schematic top view of a quadruped robot provided by an exemplary embodiment of the present application dancing with its four legs offset in the plum blossom pile environment shown in Figure 6;
图8是本申请一个示例性实施例提供的四足机器人在图6所示的梅花桩环境中双足偏移跳舞的俯视示意图；Figure 8 is a schematic top view of a quadruped robot provided by an exemplary embodiment of the present application dancing with its two feet offset in the plum blossom pile environment shown in Figure 6;
图9是本申请一个示例性实施例提供的四足机器人在图6所示的梅花桩环境中单足偏移跳舞的俯视示意图；Figure 9 is a schematic top view of a quadruped robot provided by an exemplary embodiment of the present application dancing with one foot offset in the plum blossom pile environment shown in Figure 6;
图10是本申请一个示例性实施例提供的四足机器人在图6所示的梅花桩环境中运动的俯视示意图；Figure 10 is a schematic top view of a quadruped robot provided by an exemplary embodiment of the present application moving in the plum blossom pile environment shown in Figure 6;
图11是本申请一个示例性实施例提供的足部具有3个关节电机的腿机械结构示意图；Figure 11 is a schematic diagram of the mechanical structure of a leg with three joint motors provided by an exemplary embodiment of the present application;
图12是本申请一个示例性实施例提供的四足机器人的控制装置的框图；Figure 12 is a block diagram of a control device of a quadruped robot provided by an exemplary embodiment of the present application;
图13是本申请一个示例性实施例提供的计算机设备的结构示意图。Figure 13 is a schematic structural diagram of a computer device provided by an exemplary embodiment of the present application.
具体实施方式Detailed ways
为使本申请的目的、技术方案和优点更加清楚，下面将结合附图对本申请实施方式作进一步地详细描述。In order to make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.
首先对本申请实施例涉及的若干个名词进行简介：First, a brief introduction to several terms involved in the embodiments of this application:
人工智能(Artificial Intelligence，AI)是利用数字计算机或者数字计算机控制的机器模拟、延伸和扩展人的智能，感知环境、获取知识并使用知识获得最佳结果的理论、方法、技术及应用系统。换句话说，人工智能是计算机科学的一个综合技术，它企图了解智能的实质，并生产出一种新的能以人类智能相似的方式做出反应的智能机器。人工智能也就是研究各种智能机器的设计原理与实现方法，使机器具有感知、推理与决策的功能。Artificial Intelligence (AI) is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is the study of the design principles and implementation methods of various intelligent machines, so that the machines have the functions of perception, reasoning and decision-making.
人工智能技术是一门综合学科，涉及领域广泛，既有硬件层面的技术也有软件层面的技术。人工智能基础技术一般包括如传感器、专用人工智能芯片、云计算、分布式存储、大数据处理技术、操作/交互系统、机电一体化等技术。人工智能软件技术主要包括计算机视觉技术、语音处理技术、自然语言处理技术以及机器学习/深度学习等几大方向。Artificial intelligence technology is a comprehensive subject that covers a wide range of fields, including both hardware-level technology and software-level technology. Basic artificial intelligence technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, and machine learning/deep learning.
随着人工智能技术研究和进步，人工智能技术在多个领域展开研究和应用，例如常见的智能家居、智能穿戴设备、虚拟助理、智能音箱、智能营销、无人驾驶、自动驾驶、无人机、机器人、智能医疗、智能客服等，相信随着技术的发展，人工智能技术将在更多的领域得到应用，并发挥越来越重要的价值。With the research and progress of artificial intelligence technology, artificial intelligence technology has been researched and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, driverless driving, autonomous driving, and drones. , robots, smart medical care, smart customer service, etc. I believe that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.
本申请技术方案主要涉及人工智能技术中的机器人技术，主要涉及机器人智能控制。The technical solution of this application mainly involves robot technology in artificial intelligence technology, mainly involving intelligent control of robots.
其中，机器人是利用机械传动、现代微电子技术组合而成的一种能模仿人或动物的某种技能的机械电子设备，机器人是在电子、机械及信息技术的基础上发展而来的。机器人的样子不一定必须像人，只要能自主完成人类所赋予他的任务与命令，就属于机器人大家族的成员。机器人是一种自动化的机器，这种机器具备一些与人或生物相似的智能能力，如感知能力、规划能力、动作能力和协同能力，是一种具有高度灵活性的自动化机器。随着计算机技术和人工智能技术的发展，使机器人在功能和技术层次上有了很大的提高，移动机器人和机器人的视觉和触觉等技术就是典型的代表。Among them, a robot is a mechanical electronic device that can imitate certain skills of humans or animals, which is a combination of mechanical transmission and modern microelectronic technology. Robots are developed on the basis of electronic, mechanical and information technology. A robot does not necessarily have to look like a human being. As long as it can independently complete the tasks and orders given to it by humans, it is a member of the robot family. A robot is an automated machine that has some intelligent capabilities similar to those of humans or living things, such as perception, planning, action, and collaboration. It is an automated machine with high flexibility. With the development of computer technology and artificial intelligence technology, the functions and technical levels of robots have been greatly improved. Mobile robots and robot vision and touch technologies are typical representatives.
本申请实施例提供了一种四足机器人的控制方法的技术方案，如图1所示出的四足机器人的控制方法的示意图，该方法可以由机器人或者控制设备执行。该方法包括：Embodiments of the present application provide a technical solution for a control method for a quadruped robot. Figure 1 is a schematic diagram of a control method for a quadruped robot. This method can be executed by a robot or a control device. The method includes:
示例性地，四足机器人包括基座部1、设置在基座部1上的四个足部2以及四个足部对应的若干个关节。以四足机器人所处环境为梅花桩为例，四足机器人获取梅花桩的平面坐标信息及高度信息；四足机器人基于运动方式在若干个候选落脚点3中确定若干个期望落脚点4；四足机器人控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点4上进行随机落脚运动。By way of example, a quadruped robot includes a base part 1, four feet 2 provided on the base part 1, and several joints corresponding to the four feet. Taking the environment of the quadruped robot as a plum blossom pile as an example, the quadruped robot obtains the plane coordinate information and height information of the plum blossom pile; the quadruped robot determines several expected footholds 4 among several candidate footholds 3 based on the movement method; 4. The foot robot controls the torque of at least one joint among several joints, so that the quadruped robot performs random landing motion on several desired landing points 4 .
地形信息包括凸起的若干个候选落脚点3，候选落脚点3包括至少一个梅花桩。The terrain information includes several raised candidate footholds 3, and the candidate footholds 3 include at least one plum blossom pile.
候选落脚点3是指在环境中能够提供支撑的落脚点，例如，每一个梅花桩均可作为候选落脚点3。Candidate footing points 3 refer to footing points that can provide support in the environment. For example, every plum blossom pile can be used as a candidate footing point 3.
期望落脚点4是指支撑四足机器人在环境中运动的落脚点，例如，基于四足机器人的运动方式在梅花桩均确定的期望落脚点4。The expected foothold 4 refers to the foothold that supports the movement of the quadruped robot in the environment. For example, the expected foothold 4 is determined based on the movement mode of the quadruped robot at the plum blossom pile.
可选地，期望落脚点4包括全部候选落脚点3，或，期望落脚点4包括全部候选落脚点3的子集(也即一部分)。Optionally, the desired foothold 4 includes all candidate footpoints 3 , or the desired foothold 4 includes a subset (that is, a part) of all candidate footpoints 3 .
示例性地，四足机器人控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点4上基于基本运动方式进行随机落脚运动。Illustratively, the quadruped robot controls the torque of at least one joint among several joints, so that the quadruped robot performs random landing motions based on basic movement patterns on several desired landing points 4 .
其中，基本运动方式是四足机器人在非随机落脚运动模式下的运动方式。Among them, the basic movement method is the movement method of the quadruped robot in the non-random landing movement mode.
可选地，基本运动方式包括如下运动方式中的至少一种：Optionally, the basic movement pattern includes at least one of the following movement patterns:
向第一方向移动，例如，四足机器人通过控制四足摆动，从而带动四足机器人向第一方向移动；Move in the first direction, for example, the quadruped robot controls the swing of the four legs to drive the quadruped robot to move in the first direction;
向第二方向转圈，例如，四足机器人通过控制四足朝同一方向旋转，从而带动四足机器人转圈；Turn in a circle in the second direction. For example, the quadruped robot controls the four legs to rotate in the same direction, thereby driving the quadruped robot to turn in a circle;
四个足部内部收缩，例如，四足机器人通过控制四足均收缩至基座部的底部；The four feet contract internally. For example, a quadruped robot controls all four feet to contract to the bottom of the base;
四个足部外部伸展，例如，四足机器人通过控制四足向外延伸；External extension of the four legs, for example, a quadruped robot extends outward by controlling the four legs;
四个足部中的两个足部内部收缩，例如，四足机器人通过控制四足中的任意两个足部收缩至基座部的底部；Two of the four feet are contracted internally. For example, a quadruped robot controls any two of the four feet to contract to the bottom of the base;
四个足部中的两个足部外部伸展，例如，四足机器人通过控制四足中的任意两个足部外伸远离基座部；Two of the four legs are externally extended. For example, a quadruped robot controls any two of the four legs to extend away from the base;
四个足部中的单个足部内部收缩，例如，四足机器人通过控制四足中的任意一个足部收缩至基座部的底部；The internal contraction of a single foot among the four feet, for example, a quadruped robot controls any one of the four feet to contract to the bottom of the base;
四个足部中的单个足部外部伸展，例如，四足机器人通过控制四足中的任意一个足部外伸远离基座部；The external extension of a single foot among the four legs, for example, a quadruped robot controls the external extension of any one of the four legs away from the base;
基座部往复晃动，例如，四足机器人在控制四足移动的过程中，基座部在往复晃动，产生摇头晃脑的效果；The base is rocking back and forth. For example, when a quadruped robot is controlling the movement of its four legs, the base is rocking back and forth, producing a head and head shaking effect;
四个足部的以变化步频方式运动，例如，四足机器人在控制四足移动的过程中，以变化步频进行移动；The four legs move with varying cadences. For example, a quadruped robot moves with varying cadences during the process of controlling the movement of its four legs;
四个足部的以变化步态方式运动，例如，四足机器人在控制四足移动的过程中，以变化步态方式进行移动，比如，直线型步态、S型步态。The four legs move in a changing gait. For example, a quadruped robot moves in a changing gait during the process of controlling the movement of its four legs, such as linear gait and S-shaped gait.
综上所述，本实施例提供的方法，通过获取四足机器人所处梅花桩环境的地形信息，在若干个候选落脚点中确定若干个支撑四足机器人在环境中运动的期望落脚点；控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上进行随机落脚运动。本申请基于四足机器人所处环境的地形信息，使得四足机器人能够在复杂环境下稳定的实现随机落脚运动，提升了四足机器人在复杂环境下运动时的稳定性以及展示效果。To sum up, the method provided in this embodiment determines several expected footholds among several candidate footholds that support the movement of the quadruped robot in the environment by obtaining the terrain information of the plum blossom pile environment where the quadruped robot is located; control The torque of at least one joint among several joints enables the quadruped robot to perform random landing motion on several desired landing points. This application is based on the terrain information of the environment in which the quadruped robot is located, so that the quadruped robot can stably realize random foot movement in a complex environment, and improves the stability and display effect of the quadruped robot when moving in a complex environment.
请参考图2，其示出了本申请一个示例性的实施例提供的四足机器人控制框架图。在一个示例中，如图2所示，以四足机器人控制场景为例。该方案实施环境可以包括四足机器人10和控制设备20(可选)。Please refer to FIG. 2 , which shows a control framework diagram of a quadruped robot provided by an exemplary embodiment of the present application. In one example, as shown in Figure 2, a quadruped robot control scenario is taken as an example. The implementation environment of this solution may include a quadruped robot 10 and a control device 20 (optional).
示例性地，如图2所示，四足机器人10包括：机身101(也可以称之为基底座或底盘)和腿机械结构102。机身101内设置有四足机器人10的控制器。机身101向腿机械结构102发布指令，以控制腿机械结构102的活动。腿机械结构102上设置有多个关节，每个关节处可以设置一个关节电机，也可以设置多个关节电机。以腿机械结构102中的一个腿机械结构为例，腿机械结构上设置有关节103和关节104，其中，关节103处设置有一台关节电机，关节104处设置有两台关节电机。For example, as shown in FIG. 2 , the quadruped robot 10 includes: a body 101 (which can also be called a base or chassis) and a leg mechanical structure 102 . The controller of the quadruped robot 10 is provided in the body 101 . The body 101 issues instructions to the leg mechanical structure 102 to control the activities of the leg mechanical structure 102 . The leg mechanical structure 102 is provided with multiple joints, and each joint may be provided with one joint motor or multiple joint motors. Taking one of the leg mechanical structures 102 as an example, the leg mechanical structure is provided with a joint 103 and a joint 104. Among them, a joint motor is provided at the joint 103 and two joint motors are provided at the joint 104.
控制设备20可以包括但不限于手机、电脑、智能语音交互设备、智能家电、车载终端、飞行器等；或者，控制设备20也可以是服务器。控制设备20可用于控制四足机器人10。The control device 20 may include but is not limited to a mobile phone, a computer, an intelligent voice interaction device, a smart home appliance, a vehicle-mounted terminal, an aircraft, etc.; or the control device 20 may also be a server. The control device 20 may be used to control the quadruped robot 10 .
四足机器人10和控制设备20之间可以通过网络进行通信，如无线网络、有线网络等。The quadruped robot 10 and the control device 20 can communicate through a network, such as a wireless network, a wired network, etc.
示例性地，控制设备20在获取到四足机器人10在当前时刻的运动状态和地形后，可以根据运动状态和地形，预测四足机器人10在下一时刻的运动状态，并基于预测到的运动状态来控制四足机器人10运动，以使得四足机器人10能够准确且高效的执行动作。For example, after acquiring the motion state and terrain of the quadruped robot 10 at the current moment, the control device 20 can predict the motion state of the quadruped robot 10 at the next moment based on the motion state and terrain, and based on the predicted motion state To control the movement of the quadruped robot 10 so that the quadruped robot 10 can perform actions accurately and efficiently.
可选地，该过程也可以由四足机器人10完成。例如，四足机器人10根据当前时刻的运动状态和地形，预测下一时刻的运动状态，并基于预测到的运动状态来执行运动。本申请实施例对于机器人的控制的执行主体不作限定。Optionally, this process can also be completed by the quadruped robot 10. For example, the quadruped robot 10 predicts the motion state at the next moment based on the motion state and terrain at the current moment, and performs motion based on the predicted motion state. The embodiment of the present application does not limit the execution subject of robot control.
图3示出了本申请一个示例性实施例提供的四足机器人的结构示意图。四足机器人10包括基座部110，前腿部120，后腿部130，基座部110与前腿部120和后腿部130连接。可选的，前腿部120与后腿部130是相同或不相同的。前腿部120分为左前腿和右前腿，可选的，左前腿和右前腿是相同或不相同的；后腿部130分为左后腿和右后腿，可选的，左后腿和右后腿是相同或不相同的。Figure 3 shows a schematic structural diagram of a quadruped robot provided by an exemplary embodiment of the present application. The quadruped robot 10 includes a base part 110, front legs 120, and rear legs 130. The base part 110 is connected to the front legs 120 and the rear legs 130. Optionally, the front legs 120 and the rear legs 130 are the same or different. The front legs 120 are divided into left front legs and right front legs, optionally, the left front legs and the right front legs are the same or different; the rear legs 130 are divided into left rear legs and right rear legs, optionally, the left rear legs and The right hind legs may or may not be identical.
本申请实施例以前腿部120与后腿部130相同，前腿部120的左前腿和右前腿相同，后腿部130的左后腿和右后腿相同为例进行示意性说明，但不意味着对四足机器人的腿部结构做出限定。In the embodiment of the present application, the front leg 120 and the rear leg 130 are the same, the left front leg and the right front leg of the front leg 120 are the same, and the left rear leg and the right rear leg of the rear leg 130 are the same for schematic explanation, but this does not mean that To limit the leg structure of the quadruped robot.
前腿部120包括第一左腿连杆1201、第二左腿连杆1202、第一右腿连杆1203和第二右腿连杆1204；后腿部130包括第三左腿连杆1301、第四左腿连杆1302、第三右腿连杆1303和第四右腿连杆1304。第一左腿连杆1201的第一端部与基座部110连接，第一左腿连杆1201的第二端部与第二左腿连杆1202的第一端部连接以形成第一左腿转动副；第一右腿连杆1203的第一端部与基座部110连接，第一右腿连杆1203的第二端部与第二右腿连杆1204的第一端部连接以形成第一右腿转动副。第三左腿连杆1301的第一端部与基座部110连接，第三左腿连杆1301的第二端部与第四左腿连杆1302的第一端部连接以形成第二左腿转动副；第三右腿连杆1303的第一端部与基座部110连接，第三右腿连杆1303的第二端部与第四右腿连杆1304的第一端部连接以形成第二右腿转动副。The front leg 120 includes a first left leg link 1201, a second left leg link 1202, a first right leg link 1203 and a second right leg link 1204; the rear leg 130 includes a third left leg link 1301, The fourth left leg link 1302, the third right leg link 1303 and the fourth right leg link 1304. The first end of the first left leg link 1201 is connected to the base portion 110 , and the second end of the first left leg link 1201 is connected to the first end of the second left leg link 1202 to form a first left leg link 1201 . Leg rotation pair; the first end of the first right leg link 1203 is connected to the base portion 110, and the second end of the first right leg link 1203 is connected to the first end of the second right leg link 1204. Form the first right leg rotation pair. The first end of the third left leg link 1301 is connected to the base portion 110 , and the second end of the third left leg link 1301 is connected to the first end of the fourth left leg link 1302 to form a second left leg link 1301 . Leg rotation pair; the first end of the third right leg link 1303 is connected to the base portion 110, and the second end of the third right leg link 1303 is connected to the first end of the fourth right leg link 1304. Form the second right leg rotation pair.
前腿部120和后腿部130具有若干个关节，前腿部120具有至少一个关节，后腿部130具有至少一个关节。在一些实施例中，前腿部120具有第一髋关节1205和第一膝关节1206，后腿部130具有第二髋关节1305和第二膝关节1306。The front leg 120 and the rear leg 130 have several joints, the front leg 120 has at least one joint, and the rear leg 130 has at least one joint. In some embodiments, the front leg 120 has a first hip joint 1205 and a first knee joint 1206 and the rear leg 130 has a second hip joint 1305 and a second knee joint 1306 .
图4是本申请一个示例性实施例提供的四足机器人的控制方法的流程图。该方法可以由上述图2所示实施例中的四足机器人10或者控制设备20执行。Figure 4 is a flow chart of a control method for a quadruped robot provided by an exemplary embodiment of the present application. This method can be performed by the quadruped robot 10 or the control device 20 in the embodiment shown in FIG. 2 .
该方法包括：The method includes:
步骤402：获取四足机器人所处环境的地形信息。Step 402: Obtain terrain information of the environment where the quadruped robot is located.
四足机器人包括基座部、设置在基座部上的四个足部以及四个足部对应的若干个关节。The quadruped robot includes a base, four feet arranged on the base, and several joints corresponding to the four feet.
地形信息包括凸起的若干个候选落脚点。The terrain information includes several candidate landing points for the bulge.
候选落脚点是指环境中能够提供支撑的落脚点。以梅花桩环境为例，梅花桩为梅花桩环境中能够提供支撑的落脚点，梅花桩之间的间隙则不能提供支撑的落脚点。Candidate footholds are footholds in the environment that can provide support. Taking the plum blossom pile environment as an example, the plum blossom piles are the footholds that can provide support in the plum blossom pile environment, while the gaps between the plum blossom piles cannot provide the footholds for support.
地形信息是指复杂环境对应的信息，有别于简单的地形信息，例如，简单的地形信息可以为平坦的地面、具有一定坡度的坡面或具有一定规律或规则的地形环境。Terrain information refers to information corresponding to complex environments, which is different from simple terrain information. For example, simple terrain information can be flat ground, a slope with a certain slope, or a terrain environment with certain rules or rules.
复杂环境为不规则或无规律的环境。在复杂环境中，四足机器人仅能依靠复杂环境中的特定落脚点位置来完成在复杂环境中的运动。Complex environments are irregular or irregular environments. In complex environments, quadruped robots can only rely on specific foothold positions in the complex environment to complete their movements in the complex environment.
例如，以梅花桩环境为例，梅花桩环境中凸起的梅花桩为候选落脚点，其中，每个梅花桩的高度、粗细、间隔距离、排布方式各不相同，四足机器人只能通过踩在梅花桩上才能在梅花桩环境中运动。For example, take the plum blossom pile environment as an example. The raised plum blossom piles in the plum blossom pile environment are candidate landing points. The height, thickness, spacing distance, and arrangement of each plum blossom pile are different. The four-legged robot can only pass through Only by stepping on the plum blossom piles can you move in the plum blossom pile environment.
步骤404：在若干个候选落脚点中确定若干个期望落脚点。Step 404: Determine several expected footholds among several candidate footholds.
期望落脚点是支撑四足机器人在环境中运动的落脚点。The desired foothold is the foothold that supports the movement of the quadruped robot in the environment.
可选地，期望落脚点为候选落脚点的全部，或，期望落脚点为候选落脚点的部分，但不限于此，本申请实施例对此不作具体限定。Optionally, the desired footing point is all of the candidate footing points, or the desired footing point is a part of the candidate footing point, but is not limited to this, and the embodiment of the present application does not specifically limit this.
步骤406：控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上进行随机落脚运动。Step 406: Control the torque of at least one joint among the several joints so that the quadruped robot performs random landing motion on several desired landing points.
基本运动方式是指通过预设程序或预设路径实现指向性运动的方式。The basic movement method refers to the method of achieving directional movement through a preset program or a preset path.
例如，控制四足机器人的四足向前行走、跑动，或，控制四足机器人的四足向同一方向画圆实现四足机器人的转圈，但不限于此。For example, the four legs of the quadruped robot can be controlled to walk or run forward, or the four legs of the quadruped robot can be controlled to draw a circle in the same direction to realize the quadruped robot turning in circles, but it is not limited to this.
随机落脚运动(也可称为跳舞)是指在指向性运动的基础上添加随机性动作，得到随机的，或无规律的，或无规则的运动方式。随机落脚运动中的期望落脚点具有随机性或无规律性。Random foot movement (also called dancing) refers to adding random movements on the basis of directional movement to obtain random, irregular, or irregular movement. The expected foothold in random foot movement is random or irregular.
例如，在控制四足机器人执行规律性运动的基础上添加无规则的运动，从而实现四足机器人在不规则或无规律的复杂环境下依然可以平稳运行。比如，在控制四足机器人转圈时，即控制四足机器人的四个足部分别以设定的轨迹执行画圈动作，但在复杂的环境中，转圈过程中可能会因为复杂的环境，使得四足机器人的转圈运动被迫中止；因此，在此基础上为每个足部添加随机的偏移量，从而使得四足机器人的随机落脚运动能够在不规则或无规律的复杂环境中适应性运动。For example, by controlling a quadruped robot to perform regular movements and adding irregular movements, the quadruped robot can still operate smoothly in irregular or irregular complex environments. For example, when controlling a quadruped robot to turn in a circle, the four legs of the quadruped robot are controlled to perform circle movements on a set trajectory. However, in a complex environment, the four legs of the quadruped robot may not move due to the complex environment during the circle turning process. The circular motion of the legged robot is forced to stop; therefore, a random offset is added to each foot on this basis, so that the random landing motion of the quadruped robot can adaptively move in irregular or irregular complex environments. .
机器人可以包括多个关节，每个关节用于连接机器人的两个不同部分，从而通过对该关节施加力矩，使得该关节所连接的这两个部分之间的相对位置关系能够发生变化。The robot may include a plurality of joints, each joint being used to connect two different parts of the robot, so that by applying a moment to the joint, the relative positional relationship between the two parts connected by the joint can change.
例如，在四足机器人中，四足机器人的基座部与四个足部通过若干个关节进行连接，四足机器人的每个足部可以是一个整体，也可以由至少两部分构成。给各个关节分别施加相应的力矩，从而使得各个关节产生运动，最终实现控制四足机器人执行至少两种运动。For example, in a quadruped robot, the base part and the four legs of the quadruped robot are connected through several joints. Each foot of the quadruped robot can be a whole body, or it can be composed of at least two parts. Corresponding torques are applied to each joint, thereby causing each joint to move, and ultimately the quadruped robot is controlled to perform at least two movements.
可选地，四足机器人通过关节控制信息实现对该四足机器人的关节运动进行控制，关节控制信息可以对该四足机器人的每一个关节进行单独控制，也可以对该四足机器人的每一个关节进行集体控制，本申请实施例对此不作具体限定。Optionally, the quadruped robot controls the joint motion of the quadruped robot through joint control information. The joint control information can control each joint of the quadruped robot individually, or can also control each joint of the quadruped robot. The joints are collectively controlled, which is not specifically limited in the embodiments of this application.
综上所述，本实施例提供的方法，通过获取四足机器人所处环境的地形信息，在若干个候选落脚点中确定若干个支撑四足机器人在环境中运动的期望落脚点；控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上进行随机落脚运动。本申请基于四足机器人所处环境的地形信息，使得四足机器人能够在复杂环境下稳定的实现随机落脚运动，提升了四足机器人在复杂环境下运动时的稳定性。To sum up, the method provided in this embodiment determines several desired footholds among several candidate footholds that support the movement of the quadruped robot in the environment by obtaining the terrain information of the environment where the quadruped robot is located; and controls several The torque of at least one of the joints is used to make the quadruped robot perform random landing motion on several desired landing points. This application is based on the terrain information of the environment in which the quadruped robot is located, so that the quadruped robot can stably realize random landing motion in a complex environment, and improves the stability of the quadruped robot when moving in a complex environment.
图5是本申请一个示例性实施例提供的四足机器人的控制方法的流程图。该方法可以由上述图2所示实施例中的四足机器人10或者控制设备20执行。Figure 5 is a flow chart of a control method for a quadruped robot provided by an exemplary embodiment of the present application. This method can be performed by the quadruped robot 10 or the control device 20 in the embodiment shown in FIG. 2 .
该方法包括：The method includes:
步骤502：获取四足机器人所处环境的地形信息。Step 502: Obtain terrain information of the environment where the quadruped robot is located.
四足机器人包括基座部、设置在基座部上的四个足部以及四个足部对应的若干个关节。The quadruped robot includes a base, four feet arranged on the base, and several joints corresponding to the four feet.
地形信息包括若干个候选落脚点。The terrain information includes several candidate footholds.
候选落脚点是指环境中能够提供支撑的落脚点。Candidate footholds are footholds in the environment that can provide support.
可选地，以梅花桩环境为例，候选落脚点是梅花桩环境中的至少一个凸起的梅花桩，四足机器人的足部禁止踩踏相邻两个梅花桩之间的间隙。Optionally, taking the plum blossom pile environment as an example, the candidate landing point is at least one raised plum blossom pile in the plum blossom pile environment, and the feet of the quadruped robot are prohibited from stepping on the gap between two adjacent plum blossom piles.
示例性地，四足机器人通过摄像头采集和/或动作捕捉设备和/或定位设备获取四足机器人所处环境的地形信息。其中，地形信息包括候选落脚点的平面坐标信息及高度信息。For example, the quadruped robot obtains terrain information of the environment where the quadruped robot is located through camera acquisition and/or motion capture equipment and/or positioning equipment. Among them, the terrain information includes plane coordinate information and height information of the candidate foothold.
例如，通过摄像头采集四足机器人所处环境的图片，并将四足机器人所处环境的图片转化为四足机器人可以理解的地形信息。For example, a camera collects pictures of the environment where a quadruped robot lives, and converts the pictures of the environment where the quadruped robot lives into terrain information that the quadruped robot can understand.
或，利用通过动作捕捉设备探查四足机器人所处环境，并通过动作捕捉设备将探查信息转化为四足机器人可以理解的地形信息。Or, use motion capture equipment to detect the environment where the quadruped robot is located, and use the motion capture equipment to convert the exploration information into terrain information that the quadruped robot can understand.
或，直接向四足机器人输入所处环境的地形信息。Or, directly input the terrain information of the environment to the quadruped robot.
四足机器人能够理解的地形信息可表示为：The terrain information that a quadruped robot can understand can be expressed as:
式中，h为候选落脚点的高度信息，n为候选落脚点的外法向信息(即候选落脚点的方向信息)，为候选落脚点的平面坐标信息。In the formula, h is the height information of the candidate footing point, n is the outer normal information of the candidate footing point (that is, the direction information of the candidate footing point), is the plane coordinate information of the candidate foothold.
例如，以图6所示出的梅花桩环境为例，梅花桩环境中梅花桩为候选落脚点601，四足机器人通过梅花桩环境中梅花桩支撑四足机器人的足部，从而实现四足机器人通过梅花桩环境。For example, take the plum blossom pile environment shown in Figure 6 as an example. The plum blossom pile in the plum blossom pile environment is the candidate landing point 601. The quadruped robot supports the feet of the quadruped robot through the plum blossom piles in the plum blossom pile environment, thereby realizing a quadruped robot. Through plum pile environment.
可选地，梅花桩环境中的每个梅花桩的高度、粗细、间隔距离、排布方式可各不相同，本申请实施例对此不作具体限定。Optionally, the height, thickness, spacing distance, and arrangement of each plum-blossom pile in the plum-blossom pile environment may be different, which is not specifically limited in the embodiments of the present application.
步骤504：在若干个候选落脚点中确定若干个期望落脚点。Step 504: Determine several desired footholds among several candidate footholds.
期望落脚点是支撑四足机器人在环境中运动的落脚点。The desired foothold is the foothold that supports the movement of the quadruped robot in the environment.
可选地，期望落脚点为候选落脚点的全部，或，期望落脚点为候选落脚点的部分，但不限于此，本申请实施例对此不作具体限定。Optionally, the desired footing point is all of the candidate footing points, or the desired footing point is a part of the candidate footing point, but is not limited to this, and the embodiment of the present application does not specifically limit this.
在一种可能的实现方式中，四足机器人基于基本运动方式下的运动轨迹，确定四足机器人的轨迹落脚点；四足机器人将轨迹落脚点按照偏移量进行偏移，得到偏移落脚点；四足机器人在若干个候选落脚点中选择与偏移落脚点最近的候选落脚点，确定为基于基本运动方式进行随机落脚运动时的期望落脚点。In one possible implementation, the quadruped robot determines the trajectory footing point of the quadruped robot based on the movement trajectory in the basic movement mode; the quadruped robot offsets the trajectory footing point according to the offset amount to obtain the offset footing point. ; The quadruped robot selects the candidate footing point closest to the offset footing point among several candidate footing points, and determines it as the expected footing point when performing random footing movements based on the basic movement method.
其中，偏移量是在以轨迹落脚点为基准位置，以关节的活动范围为选择范围内确定的偏移量。Among them, the offset is determined based on the trajectory footing point as the reference position and the range of motion of the joint as the selection range.
可选地，在偏移落脚点的平面坐标与候选落脚点的平面坐标之间的距离最近的情况下，将候选落脚点确定为基于基本运动方式进行随机落脚运动时的期望落脚点。Optionally, when the distance between the plane coordinates of the offset footing point and the plane coordinates of the candidate footing point is the shortest, the candidate footing point is determined as the expected footing point when random footing motion is performed based on the basic movement pattern.
例如，偏移落脚点的平面坐标为P偏移＝[x；y；0]，则期望落脚点P期待的确定公式可表示为：For example, if the plane coordinate of the offset footing point is Poffset = [x; y; 0], then the formula for determining theexpected footing point P can be expressed as:
p期待＝{p|min(||p偏移-([1；1；0]p)T||),p∈p候选}pexpectation = {p|min(||poffset -([1;1;0]p)T ||),p∈pcandidate }
式中，P偏移为偏移落脚点；P候选为地形信息中的若干个候选落脚点的集合；P为一个候选落脚点；P期待为期待落脚点。In the formula, Poffset is the offset footing point; Pcandidate is a set of several candidate footing points in the terrain information; P is a candidate footing point; Pexpectation is the expected footing point.
基本运动方式是四足机器人在非随机落脚运动模式下的运动方式。The basic movement method is the movement method of a quadruped robot in a non-random landing movement mode.
可选地，基本运动方式包括如下运动方式中的至少一种：Optionally, the basic movement pattern includes at least one of the following movement patterns:
向第一方向移动，例如，四足机器人通过控制四足摆动，从而带动四足机器人向第一方向移动；Move in the first direction, for example, the quadruped robot controls the swing of the four legs to drive the quadruped robot to move in the first direction;
向第二方向转圈，例如，四足机器人通过控制四足朝同一方向旋转，从而带动四足机器人转圈；Turn in a circle in the second direction. For example, the quadruped robot controls the four legs to rotate in the same direction, thereby driving the quadruped robot to turn in a circle;
四个足部内部收缩，例如，四足机器人通过控制四足均收缩至基座部的底部；The four feet contract internally. For example, a quadruped robot controls all four feet to contract to the bottom of the base;
四个足部外部伸展，例如，四足机器人通过控制四足向外延伸；External extension of the four legs, for example, a quadruped robot extends outward by controlling the four legs;
四个足部中的两个足部内部收缩，例如，四足机器人通过控制四足中的任意两个足部收缩至基座部的底部；Two of the four feet are contracted internally. For example, a quadruped robot controls any two of the four feet to contract to the bottom of the base;
四个足部中的两个足部外部伸展，例如，四足机器人通过控制四足中的任意两个足部外伸远离基座部；Two of the four legs are externally extended. For example, a quadruped robot controls any two of the four legs to extend away from the base;
四个足部中的单个足部内部收缩，例如，四足机器人通过控制四足中的任意一个足部收缩至基座部的底部；The internal contraction of a single foot among the four feet, for example, a quadruped robot controls any one of the four feet to contract to the bottom of the base;
四个足部中的单个足部外部伸展，例如，四足机器人通过控制四足中的任意一个足部外伸远离基座部；The external extension of a single foot among the four legs, for example, a quadruped robot controls the external extension of any one of the four legs away from the base;
基座部往复晃动，例如，四足机器人在控制四足移动的过程中，基座部在往复晃动，产生摇头晃脑的效果；The base is rocking back and forth. For example, when a quadruped robot is controlling the movement of its four legs, the base is rocking back and forth, producing a head and head shaking effect;
四个足部的以变化步频方式运动，例如，四足机器人在控制四足移动的过程中，以变化步频进行移动；The four legs move with varying cadences. For example, a quadruped robot moves with varying cadences during the process of controlling the movement of its four legs;
四个足部的以变化步态方式运动，例如，四足机器人在控制四足移动的过程中，以变化步态方式进行移动，比如，直线型步态、S型步态。The four legs move in a changing gait. For example, a quadruped robot moves in a changing gait during the process of controlling the movement of its four legs, such as linear gait and S-shaped gait.
在一种可能的实现方式中，四足机器人控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上基于基本运动方式进行运动。In a possible implementation, the quadruped robot controls the torque of at least one joint among several joints, so that the quadruped robot moves based on basic movement patterns at several desired footholds.
示例性地，基于四足机器人在基本运动方式下的运动轨迹，四足机器人确定四足机器人的轨迹落脚点；四足机器人在若干个候选落脚点中选择与轨迹落脚点最近的候选落脚点，确定为基于基本运动方式进行运动时的期望落脚点。For example, based on the movement trajectory of the quadruped robot in the basic movement mode, the quadruped robot determines the trajectory footing point of the quadruped robot; the quadruped robot selects the candidate footing point closest to the trajectory footing point among several candidate footing points, Identified as the desired landing point when exercising based on basic movement patterns.
可选地，四足机器人在轨迹落脚点的平面坐标与候选落脚点的平面坐标之间的距离小于第二距离阈值的情况下，将候选落脚点确定为基于基本运动方式进行运动时的期望落脚点。Optionally, when the distance between the plane coordinates of the trajectory footing point and the plane coordinates of the candidate footing point is less than the second distance threshold, the quadruped robot determines the candidate footing point as the expected footing when moving based on the basic movement mode. point.
例如，轨迹落脚点的平面坐标为P轨迹＝[x；y；0]，则期望落脚点P期待的确定公式可表示为：For example, the plane coordinates of the trajectory footing point are Ptrajectory = [x; y; 0], then the formula for determining theexpected footing point P can be expressed as:
p期待＝{p|min(||p轨迹-([1；1；0]p)T||),p∈p候选}pexpectation = {p|min(||ptrajectory -([1;1;0]p)T ||),p∈pcandidate }
式中，P轨迹为轨迹落脚点；P候选为地形信息中的凸起的若干个候选落脚点的集合；P为一个候选落脚点；P期待为期待落脚点。In the formula, Ptrajectory is the trajectory footing point; Pcandidate is a set of several candidate footing points that are raised in the terrain information; P is a candidate footing point; Pexpectation is the expected footing point.
步骤506：控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上基于基本运动方式进行进行随机落脚运动。Step 506: Control the torque of at least one joint among the several joints, so that the quadruped robot performs random landing motion based on the basic movement pattern at several desired landing points.
机器人可以包括多个关节，每个关节用于连接机器人的两个不同部分，从而通过对该关节施加力矩，使得该关节所连接的这两个部分之间的相对位置关系能够发生变化。The robot may include a plurality of joints, each joint being used to connect two different parts of the robot, so that by applying a moment to the joint, the relative positional relationship between the two parts connected by the joint can change.
随机落脚运动是指四足机器人通过改变四个足部的位置，或，基座部的质心位置，或，同时改变四个足部的位置及基座部的质心位置来实现四足机器人形态、动作改变的方式。随机落脚运动中的期望落脚点具有随机性或无规律性。Random landing motion means that the quadruped robot realizes the shape of the quadruped robot by changing the positions of the four feet, or the position of the center of mass of the base, or simultaneously changing the positions of the four feet and the position of the center of mass of the base. The way actions change. The expected foothold in random foot movement is random or irregular.
可选地，轨迹落脚点按照偏移量进行偏移得到偏移落脚点的情况包括如下几种类型，但不限于此，本申请实施例对此不作具体限定。Optionally, the situation in which the trajectory footing point is offset according to the offset amount to obtain the offset footing point includes the following types, but is not limited to these, and the embodiment of the present application does not specifically limit this.
类型一：将与四个足部中的每个足部对应的轨迹落脚点按照与每个足部对应的偏移量进行偏移，得到与每个足部对应的偏移落脚点。Type 1: Offset the trajectory footing point corresponding to each of the four feet according to the offset amount corresponding to each foot to obtain the offset footing point corresponding to each foot.
可选地，将四个足部中的第一足部对应的轨迹落脚点按照与第一足部对应的第一偏移量进行偏移，得到与第一足部对应的第一偏移落脚点；将四个足部中的第二足部对应的轨迹落脚点按照与第二足部对应的第二偏移量进行偏移，得到与第二足部对应的第二偏移落脚点；将四个足部中的第三足部对应的轨迹落脚点按照与第三足部对应的第三偏移量进行偏移，得到与第三足部对应的第三偏移落脚点；将四个足部中的第四足部对应的轨迹落脚点按照与第四足部对应的第四偏移量进行偏移，得到与第四足部对应的第四偏移落脚点。Optionally, the trajectory footing point corresponding to the first foot among the four feet is offset according to the first offset corresponding to the first foot, to obtain the first offset footing corresponding to the first foot. point; offset the trajectory footing point corresponding to the second foot among the four feet according to the second offset corresponding to the second foot, to obtain the second offset footing point corresponding to the second foot; The trajectory footing point corresponding to the third foot among the four feet is offset according to the third offset corresponding to the third foot to obtain the third offset footing point corresponding to the third foot; The trajectory footing point corresponding to the fourth foot among the four feet is offset according to the fourth offset amount corresponding to the fourth foot, and a fourth offset footing point corresponding to the fourth foot is obtained.
可选地，第一偏移量、第二偏移量、第三偏移量、第四偏移量可以为相等、不等或不完全相等，本申请实施例对此不作具体限定。Optionally, the first offset, the second offset, the third offset, and the fourth offset may be equal, unequal, or not completely equal, which is not specifically limited in the embodiment of the present application.
可选地，第一偏移量、第二偏移量、第三偏移量、第四偏移量自行设定，或，在各自足部对应的关节的活动范围内随机生成偏移量，本申请实施例对此不作具体限定。Optionally, the first offset, the second offset, the third offset, and the fourth offset are set by themselves, or the offsets are randomly generated within the range of motion of the joints corresponding to the respective feet, The embodiments of the present application do not specifically limit this.
类型二：将与两个对角足部中的每个足部对应的轨迹落脚点按照与每个足部对应的偏移量进行偏移，得到与每个足部对应的偏移落脚点。其中，两个对角足部是四个足部中位于同一对角线上的两个足部。Type 2: Offset the trajectory footing point corresponding to each of the two diagonal feet according to the offset amount corresponding to each foot to obtain the offset footing point corresponding to each foot. Among them, the two diagonal feet are the two feet located on the same diagonal line among the four feet.
可选地，将四个足部中的第一足部和第二足部的轨迹落脚点按照与第一足部对应的第一偏移量进行偏移，得到与第一足部对应的第一偏移落脚点；将第二足部对应的轨迹落脚点按照与第二足部对应的第二偏移量进行偏移，得到与第二足部对应的第二偏移落脚点；将四个足部中的第三足部和第四足部固定在对应的第三轨迹落脚点和第四轨迹落脚点。Optionally, the trajectory landing points of the first foot and the second foot among the four feet are offset according to the first offset corresponding to the first foot, to obtain the third offset corresponding to the first foot. One offset footing point; offset the trajectory footing point corresponding to the second foot according to the second offset corresponding to the second foot to obtain the second offset footing point corresponding to the second foot; The third foot part and the fourth foot part among the three feet parts are fixed at the corresponding third trajectory footing point and the fourth trajectory footing point.
类型三：将与单个足部对应的轨迹落脚点按照与单个足部对应的偏移量进行偏移，得到与单个足部对应的偏移落脚点；其中，单个足部是四个足部中的一个足部。Type 3: Offset the trajectory footing point corresponding to a single foot according to the offset amount corresponding to the single foot, and obtain the offset footing point corresponding to the single foot; where the single foot is the center of the four feet. of a foot.
可选地，将四个足部中的第一足部的轨迹落脚点按照与第一足部对应的第一偏移量进行偏移，得到与第一足部对应的第一偏移落脚点；将四个足部中的第二足部、第三足部和第四足部固定在对应的第二轨迹落脚点、第三轨迹落脚点和第四轨迹落脚点。Optionally, the trajectory footing point of the first foot among the four feet is offset according to the first offset corresponding to the first foot, to obtain the first offset footing point corresponding to the first foot. ; Fix the second foot, the third foot and the fourth foot among the four feet at the corresponding second trajectory footing point, the third trajectory footing point and the fourth trajectory footing point.
可选地，四足机器人在以轨迹落脚点为基准位置，以关节的活动范围为选择范围内随机生成偏移量。Optionally, the quadruped robot randomly generates offsets within the selected range using the trajectory footing point as the reference position and the range of motion of the joints.
图7示出了本申请实施例涉及的四足机器人在图6所示的梅花桩环境中四足偏移跳舞的俯视示意图，如图7所示，图中的白色圆圈代表候选落脚点703，方框为四足机器人的基座部701，与方框相连接的圆圈代表基座部701上的四个足部，足部所在的圆圈代表期望落脚点704，黑色箭头代表足部偏移方向。Figure 7 shows a top view of the quadruped robot involved in the embodiment of the present application dancing with its four legs offset in the plum blossom pile environment shown in Figure 6. As shown in Figure 7, the white circle in the figure represents the candidate foothold 703. The box represents the base part 701 of the quadruped robot. The circles connected to the box represent the four feet on the base part 701. The circle where the feet are located represents the desired foothold 704. The black arrow represents the foot offset direction. .
如图7中的(a)图所示，四足机器人的基本运动方式为转圈，在转圈运动的基础上，四足机器人将四个足部中的每个足部对应的轨迹落脚点按照与每个足部对应的偏移量向四足机器人的基座部701底部偏移，图7中(a)图中的四个足部所在位置为每个足部对应的偏移落脚点702；如图7中的(b)图所示，在偏移落脚点702的平面坐标与候选落脚点703的平面坐标之间的距离最近的情况下，将候选落脚点703确定为基于转圈运动式进行跳舞时的期望落脚点704，从而使得四足机器人在若干个期望落脚点704上基于转圈运动方式进行跳舞。As shown in (a) of Figure 7, the basic movement mode of the quadruped robot is to turn in circles. On the basis of the circle movement, the quadruped robot moves the trajectory corresponding to each of the four feet according to the The offset corresponding to each foot is offset toward the bottom of the base portion 701 of the quadruped robot. The positions of the four feet in (a) of Figure 7 are the offset footing points 702 corresponding to each foot; As shown in (b) of FIG. 7 , when the distance between the plane coordinates of the offset footing point 702 and the plane coordinates of the candidate footing point 703 is the shortest, the candidate footing point 703 is determined to be performed based on the circular motion. The desired footholds 704 during dancing allow the quadruped robot to dance based on the circular motion on several desired footholds 704 .
如图7中的(c)图所示，四足机器人的基本运动方式为转圈，在转圈运动的基础上，四足机器人将四个足部中的每个足部对应的轨迹落脚点按照与每个足部对应的偏移量向四足机器人的基座部701外部偏移，图7中(c)图中的四个足部所在位置为每个足部对应的偏移落脚点702；如图7中的(d)图所示，在偏移落脚点702的平面坐标与候选落脚点703的平面坐标之间的距离最近的情况下，将候选落脚点703确定为基于转圈运动式进行跳舞时的期望落脚点704，从而使得四足机器人在若干个期望落脚点704上基于转圈运动方式进行跳舞。As shown in (c) of Figure 7, the basic movement mode of the quadruped robot is to turn in circles. On the basis of the circle movement, the quadruped robot sets the trajectory corresponding to each of the four feet according to the The offset corresponding to each foot is offset to the outside of the base portion 701 of the quadruped robot. The positions of the four feet in (c) of Figure 7 are the offset footing points 702 corresponding to each foot; As shown in (d) of FIG. 7 , when the distance between the plane coordinates of the offset footing point 702 and the plane coordinates of the candidate footing point 703 is the shortest, the candidate footing point 703 is determined to be performed based on the circular motion. The desired footholds 704 during dancing allow the quadruped robot to dance based on the circular motion on several desired footholds 704 .
图8示出了本申请实施例涉及的四足机器人在图6所示的梅花桩环境中双足偏移跳舞的俯视示意图，如图8所示，图中的白色圆圈代表候选落脚点803，方框为四足机器人的基座部801，与方框相连接的圆圈代表基座部801上的四个足部，足部所在的圆圈代表期望落脚点804，黑色箭头代表足部偏移方向。Figure 8 shows a top view of the quadruped robot involved in the embodiment of the present application dancing with its feet offset in the plum blossom pile environment shown in Figure 6. As shown in Figure 8, the white circle in the figure represents the candidate foothold 803. The box represents the base part 801 of the quadruped robot. The circles connected to the box represent the four feet on the base part 801. The circle where the feet are located represents the desired footing point 804. The black arrow represents the offset direction of the feet. .
如图8中的(a)图所示，四足机器人的基本运动方式为转圈，在转圈运动的基础上，四足机器人将四个足部中对称的两个足部对应的轨迹落脚点按照与每个足部对应的偏移量向四足机器人的基座部801底部偏移，图8中(a)图中的两个对称偏移足部所在的位置为对应的偏移落脚点802，未偏移的两个足部所在的位置为对应的轨迹落脚点805；如图8中的(b)图所示，在偏移落脚点802的平面坐标与候选落脚点803的平面坐标之间的距离最近的情况下，将候选落脚点803确定为基于转圈运动式进行跳舞时的期望落脚点804，在未偏移的轨迹落脚点805的平面坐标与候选落脚点803的平面坐标之间的距离最近的情况下，将候选落脚点803确定为基于转圈运动式进行跳舞时的未偏移的轨迹落脚点805对应的期望落脚点804，从而使得四足机器人在若干个期望落脚点804上基于转圈运动方式进行跳舞。As shown in (a) of Figure 8, the basic movement mode of the quadruped robot is to turn in circles. On the basis of the circle movement, the quadruped robot sets the trajectory corresponding to the two symmetrical foot points among the four feet according to The offset corresponding to each foot is offset toward the bottom of the base portion 801 of the quadruped robot. The position of the two symmetrically offset feet in (a) of Figure 8 is the corresponding offset footing point 802 , the positions of the two non-offset feet are the corresponding trajectory foot points 805; as shown in (b) of Figure 8, between the plane coordinates of the offset foot point 802 and the plane coordinates of the candidate foot point 803 In the case where the distance between the two is the shortest, the candidate foothold 803 is determined as the expected foothold 804 when dancing based on the circle motion. In the case of the closest distance, the candidate foothold 803 is determined as the expected foothold 804 corresponding to the undeviated trajectory footpoint 805 when dancing based on the circle motion, so that the quadruped robot can be on several expected footholds 804 Dance based on circular motion.
如图8中的(c)图所示，四足机器人的基本运动方式为转圈，在转圈运动的基础上，四足机器人将四个足部中对称的两个足部对应的轨迹落脚点802按照与每个足部对应的偏移量向四足机器人的基座部801外部偏移，图8中(c)图中的对称的两个足部所在位置为每个足部对应的偏移落脚点802，未偏移的两个足部所在的位置为对应的轨迹落脚点805；如图8中的(d)图所示，在偏移落脚点802的平面坐标与候选落脚点803的平面坐标之间的距离最近的情况下，将候选落脚点803确定为基于转圈运动式进行跳舞时的期望落脚点804，在未偏移的轨迹落脚点805的平面坐标与候选落脚点803的平面坐标之间的距离最近的情况下，将候选落脚点803确定为基于转圈运动式进行跳舞时的未偏移的轨迹落脚点805对应的期望落脚点804，从而使得四足机器人在若干个期望落脚点804上基于转圈运动方式进行跳舞。As shown in (c) of Figure 8, the basic movement mode of the quadruped robot is to turn in circles. On the basis of the circle movement, the quadruped robot sets the trajectory corresponding to the two symmetrical foot points 802 of the four feet. Offset to the outside of the base portion 801 of the quadruped robot according to the offset amount corresponding to each foot. The position of the two symmetrical feet in (c) of Figure 8 is the offset corresponding to each foot. The foothold point 802, the position of the two non-offset feet is the corresponding trajectory foothold point 805; as shown in (d) of Figure 8, the plane coordinates of the offset foothold point 802 and the candidate foothold point 803 In the case where the distance between the plane coordinates is the shortest, the candidate foothold 803 is determined as the expected foothold 804 when dancing based on the circle motion. In the case where the distance between the coordinates is the shortest, the candidate foothold 803 is determined as the expected foothold 804 corresponding to the undeviated trajectory footpoint 805 when dancing based on the circle motion, thereby allowing the quadruped robot to land at several desired footholds. Click 804 to dance based on the circle movement.
图9示出了本申请实施例涉及的四足机器人在图6所示的梅花桩环境中单足偏移跳舞的俯视示意图，如图9所示，图中的白色圆圈代表候选落脚点903，方框为四足机器人的基座部901，与方框相连接的圆圈代表基座部901上的四个足部，足部所在的圆圈代表期望落脚点904，黑色箭头代表足部偏移方向。Figure 9 shows a top view of the quadruped robot involved in the embodiment of the present application dancing with one foot offset in the plum blossom pile environment shown in Figure 6. As shown in Figure 9, the white circle in the figure represents the candidate foothold 903. The box represents the base part 901 of the quadruped robot. The circles connected to the box represent the four feet on the base part 901. The circle where the feet are located represents the desired footing point 904. The black arrow represents the offset direction of the feet. .
如图9中的(a)图所示，四足机器人的基本运动方式为转圈，在转圈运动的基础上，四足机器人将四个足部中的一个足部对应的轨迹落脚点按照与该足部对应的偏移量向四足机器人的基座部901底部偏移，图9中(a)图中的单个偏移足部所在的位置为对应的偏移落脚点902，未偏移的三个足部所在的位置为对应的轨迹落脚点905；如图9中的(b)图所示，在偏移落脚点902的平面坐标与候选落脚点903的平面坐标之间的距离最近的情况下，将候选落脚点903确定为基于转圈运动式进行跳舞时的期望落脚点904，在未偏移的轨迹落脚点905的平面坐标与候选落脚点903的平面坐标之间的距离最近的情况下，将候选落脚点903确定为基于转圈运动式进行跳舞时的未偏移的轨迹落脚点905对应的期望落脚点904，从而使得四足机器人在若干个期望落脚点904上基于转圈运动方式进行跳舞。As shown in (a) of Figure 9, the basic movement mode of the quadruped robot is to turn in circles. On the basis of the circle movement, the quadruped robot sets the trajectory corresponding to one of the four feet according to the footing point. The corresponding offset amount of the foot is offset towards the bottom of the base part 901 of the quadruped robot. The position of the single offset foot in (a) of Figure 9 is the corresponding offset footing point 902. The unoffset The positions of the three feet are the corresponding trajectory foot points 905; as shown in (b) of Figure 9, the distance between the plane coordinates of the offset foot point 902 and the plane coordinates of the candidate foot point 903 is the closest. In this case, the candidate foothold 903 is determined as the desired foothold 904 when dancing based on the circle movement pattern, and the distance between the plane coordinates of the undeviated trajectory footpoint 905 and the plane coordinates of the candidate footpoint 903 is the shortest. Next, the candidate foothold 903 is determined as the expected foothold 904 corresponding to the undeviated trajectory footpoint 905 when dancing based on the circle motion, thereby allowing the quadruped robot to dance based on the circle motion on several desired footholds 904 Dance.
如图9中的(c)图所示，四足机器人的基本运动方式为转圈，在转圈运动的基础上，四足机器人将四个足部中单个足部对应的轨迹落脚点902按照与该足部对应的偏移量向四足机器人的基座部901外部偏移，图9中(c)图中的单个足部所在位置为足部对应的偏移落脚点902，未偏移的足部所在的位置为对应的轨迹落脚点905；如图9中的(d)图所示，在偏移落脚点902的平面坐标与候选落脚点903的平面坐标之间的距离最近的情况下，将候选落脚点903确定为基于转圈运动式进行跳舞时的期望落脚点904，在未偏移的轨迹落脚点905的平面坐标与候选落脚点903的平面坐标之间的距离最近的情况下，将候选落脚点903确定为基于转圈运动式进行跳舞时的未偏移的轨迹落脚点905对应的期望落脚点904，从而使得四足机器人在若干个期望落脚点904上基于转圈运动方式进行跳舞。As shown in (c) of Figure 9, the basic movement mode of the quadruped robot is turning in circles. On the basis of the circle movement, the quadruped robot sets the trajectory landing point 902 corresponding to a single foot among the four feet according to the The corresponding offset amount of the foot is offset to the outside of the base part 901 of the quadruped robot. The position of a single foot in (c) of Figure 9 is the corresponding offset footing point 902 of the foot. The undeviated foot The position where the bottom is located is the corresponding trajectory footing point 905; as shown in (d) of Figure 9, when the distance between the plane coordinates of the offset footing point 902 and the plane coordinates of the candidate footing point 903 is the shortest, The candidate foothold point 903 is determined as the expected foothold point 904 when dancing based on the circle motion. When the distance between the plane coordinates of the undeviated trajectory footpoint 905 and the plane coordinates of the candidate foothold point 903 is the shortest, The candidate foothold 903 is determined to be the expected foothold 904 corresponding to the undeviated trajectory footpoint 905 when dancing based on the circling motion, so that the quadruped robot dances based on the circling motion on several desired footholds 904 .
需要说明的是，四足机器人对应的足部偏移量可以统一设定，或，单独设定，或，以关节的活动范围为选择范围内随机生成，但不限于此，本申请实施例对此不作具体限定。It should be noted that the foot offsets corresponding to the quadruped robot can be set uniformly, or individually, or randomly generated within the selected range based on the range of motion of the joints, but are not limited to this. The embodiments of the present application are This is not specifically limited.
可以理解的是，四足机器人对应的运动方式可以在基于基本运动方式进行跳舞的基础上添加前后左右的运动指令，从而增加展示效果。It is understandable that the corresponding movement method of the quadruped robot can add front, back, left, and right movement instructions on top of dancing based on the basic movement method, thereby increasing the display effect.
四足机器人对应的运动方式可以在基于基本运动方式进行跳舞的基础上添加基座部姿态改变的运动指令，从而产生摇头晃脑的展示效果。The corresponding movement method of the quadruped robot can add movement instructions for changing the posture of the base part on the basis of dancing based on the basic movement method, thereby producing a head-and-shaking display effect.
四足机器人对应的运动方式可以在基于基本运动方式进行跳舞动的基础上添加改变步频和/或改变运动步态的运动指令，从而增加展示效果。The movement method corresponding to the quadruped robot can add movement instructions to change the step frequency and/or change the movement gait on the basis of dancing based on the basic movement method, thereby increasing the display effect.
步骤508：控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上基于基本运动方式进行运动。Step 508: Control the torque of at least one joint among the several joints, so that the quadruped robot moves based on the basic movement pattern on several desired footholds.
机器人可以包括多个关节，每个关节用于连接机器人的两个不同部分，从而通过对该关节施加力矩，使得该关节所连接的这两个部分之间的相对位置关系能够发生变化。The robot may include a plurality of joints, each joint being used to connect two different parts of the robot, so that by applying a moment to the joint, the relative positional relationship between the two parts connected by the joint can change.
示例性地，四足机器人控制控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上基于基本运动方式进行运动。Exemplarily, the quadruped robot controls the torque of at least one joint among several joints, so that the quadruped robot moves based on basic movement patterns on several desired footholds.
图10示出了本申请实施例涉及的四足机器人在图6所示的梅花桩环境中运动的俯视示意图，如图10所示，图中的白色圆圈代表候选落脚点1003，方框为四足机器人的基座部1001，与方框相连接的圆圈代表基座部1001上的四个足部，足部所在的圆圈代表期望落脚点1004。Figure 10 shows a schematic top view of the quadruped robot involved in the embodiment of the present application moving in the plum blossom pile environment shown in Figure 6. As shown in Figure 10, the white circle in the figure represents the candidate foothold 1003, and the square box is four The base part 1001 of the foot robot, the circles connected to the square box represent the four feet on the base part 1001, and the circle where the feet are located represents the desired landing point 1004.
如图10中的(a)图所示，四足机器人的基本运动方式为转圈，基于转圈运动，四足机器人的四个足部按照每个足部对应的轨迹落脚点1002进行运动；如图10中的(b)图所示，在轨迹落脚点1002的平面坐标与候选落脚点1003的平面坐标之间的距离小于第二距离阈值的情况下，将候选落脚点1003确定为基于转圈运动式进行运动的期望落脚点1004，从而使得四足机器人在若干个期望落脚点1004上基于转圈运动方式进行运动。As shown in (a) of Figure 10, the basic movement mode of the quadruped robot is turning in circles. Based on the turning movement, the four feet of the quadruped robot move according to the trajectory landing point 1002 corresponding to each foot; as shown in the figure As shown in Figure 10(b), when the distance between the plane coordinates of the trajectory footing point 1002 and the plane coordinates of the candidate footing point 1003 is less than the second distance threshold, the candidate footing point 1003 is determined to be based on the circle motion formula. The desired foothold 1004 for movement is determined, so that the quadruped robot moves based on the circular motion on several desired footholds 1004 .
综上所述，本实施例提供的方法，通过获取四足机器人所处环境的地形信息，在若干个候选落脚点中确定若干个支撑四足机器人在环境中运动的期望落脚点；控制若干个关节中的至少一个关节的力矩，以使得四足机器人在若干个期望落脚点上基于基本运动方式进行随机落脚运动，或，以使得四足机器人在若干个期望落脚点上基于基本运动方式进行运动。本申请基于四足机器人所处环境的地形信息，使得四足机器人能够在复杂环境下稳定的实现随机落脚运动，提升了四足机器人在复杂环境下运动时的稳定性以及展示效果。To sum up, the method provided in this embodiment determines several desired footholds among several candidate footholds that support the movement of the quadruped robot in the environment by obtaining the terrain information of the environment where the quadruped robot is located; and controls several The torque of at least one of the joints is used to enable the quadruped robot to perform random landing motion based on the basic movement pattern on several desired foothold points, or to cause the quadruped robot to move based on the basic movement pattern on several desired foothold points. . This application is based on the terrain information of the environment in which the quadruped robot is located, so that the quadruped robot can stably realize random foot movement in a complex environment, and improves the stability and display effect of the quadruped robot when moving in a complex environment.
示例性地，以四足机器人的一个足部具有3个关节电机的腿机械结构为例，如图11所示，第1个关节1102在四足机器人的底盘1101的本地坐标系下的本地位置坐标可以记为(L1，L2，0)，第2个关节1103在第1个关节1102的本地坐标系下的本地位置坐标可以记为(0，L3，0)，第3个关节1104在第2个关节1103的本地坐标系下的本地位置坐标可以记为(0，0，-L4)，落脚点1105在在第3个关节1104的本地坐标系下的本地位置坐标可以记为(0，0，-L5)。基于四足机器人的电机转角确定四足机器人的机身姿态。For example, taking a leg mechanical structure with three joint motors in one foot of a quadruped robot as an example, as shown in Figure 11, the local position of the first joint 1102 in the local coordinate system of the quadruped robot's chassis 1101 The coordinates can be recorded as (L1 , L2 , 0), the local position coordinates of the second joint 1103 in the local coordinate system of the first joint 1102 can be recorded as (0, L3 , 0), and the third joint 1103 can be recorded as (0, L 3 , 0). The local position coordinates of 1104 in the local coordinate system of the second joint 1103 can be recorded as (0, 0, -L4 ), and the local position coordinates of the foothold 1105 in the local coordinate system of the third joint 1104 can be recorded as is (0, 0, -L5 ). The body posture of the quadruped robot is determined based on the motor rotation angle of the quadruped robot.
调用以下方程组可以得到四足机器人的电机转角：The motor rotation angle of the quadruped robot can be obtained by calling the following set of equations:
其中，θ1、θ2和θ3分别为图中肢体上的电机转角。x、y、z为落脚点在底盘的本地坐标系下的本地位置坐标。Among them, θ1 , θ2 and θ3 are the motor rotation angles on the limbs in the figure respectively. x, y, z are the local position coordinates of the foothold in the local coordinate system of the chassis.
当前机身姿态是指四足机器人在当前周期的机身姿态。可选地，当前机身姿态可以用欧拉角表示，也可以用四元数表示。当前关节电机转角是指四足机器人的关节电机在当前周期的转角。The current body posture refers to the body posture of the quadruped robot in the current cycle. Optionally, the current body attitude can be expressed by Euler angles or quaternions. The current joint motor rotation angle refers to the rotation angle of the quadruped robot's joint motor in the current cycle.
可选地，m个关节电机在当前周期的目标电机转角，可以基于机器人控制模型中的策略网络计算得到，其具体内容可以如下：通过策略网络对当前状态、前j个周期的历史状态、前i个周期的历史动作和后k个周期的参考状态进行数据处理，得到m个关节电机在当前周期的偏置目标转角；从参考状态序列中确定机器人在下一个周期对应的m个关节电机的参考电机转角；加权计算m个关节电机在当前周期的偏置目标转角和下一个周期对应的m个关节电机的参考电机转角，得到m个关节电机在当前周期的目标电机转角。Optionally, the target motor rotation angles of the m joint motors in the current cycle can be calculated based on the policy network in the robot control model. The specific content can be as follows: through the policy network, the current state, the historical state of the previous j cycles, and the previous The historical actions of i cycles and the reference states of the next k cycles are processed to obtain the offset target rotation angles of the m joint motors in the current cycle; the reference of the m joint motors corresponding to the robot in the next cycle is determined from the reference state sequence. Motor rotation angle; weighted calculation of the offset target rotation angle of the m joint motors in the current cycle and the reference motor rotation angle of the m joint motors in the next cycle, to obtain the target motor rotation angle of the m joint motors in the current cycle.
需要说明的是，本申请所涉及的信息(包括但不限于地形信息)、数据(包括但不限于用于分析的数据、存储的数据、展示的数据等)以及信号，均为经对象授权或者经过各方充分授权的，且相关数据的收集、使用和处理需要遵守相关国家和地区的相关法律法规和标准。It should be noted that the information (including but not limited to terrain information), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the subject or Fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.
图12示出了本申请一个示例性实施例提供的四足机器人的控制装置的结构框图。该装置可以通过软件、硬件或者两者的结合实现成为四足机器人的全部或一部分，该装置包括：Figure 12 shows a structural block diagram of a control device for a quadruped robot provided by an exemplary embodiment of the present application. The device can be implemented as all or part of a quadruped robot through software, hardware, or a combination of both. The device includes:
获取模块1201，用于获取所述四足机器人所处环境的地形信息，所述地形信息包括若干个候选落脚点；The acquisition module 1201 is used to acquire terrain information of the environment where the quadruped robot is located, where the terrain information includes several candidate footholds;
确定模块1202，用于在所述若干个候选落脚点中确定若干个期望落脚点，所述期望落脚点是支撑所述四足机器人在所述环境中运动的落脚点；Determining module 1202, configured to determine several expected footholds among the several candidate footholds, where the expected footholds are footholds that support the movement of the quadruped robot in the environment;
控制模块1203，用于控制所述若干个关节中的至少一个关节的力矩，以使得所述四足机器人在所述若干个期望落脚点上进行随机落脚运动，所述随机落脚运动中的所述期望落脚点具有随机性或无规律性。The control module 1203 is used to control the torque of at least one joint among the several joints, so that the quadruped robot performs a random landing motion on the several desired landing points, and the said random landing motion is Expect landing points to be random or irregular.
在一种可能的实现方式中，控制模块1203，用于控制所述若干个关节中的至少一个关节的力矩，以使得所述四足机器人在所述若干个期望落脚点上基于基本运动方式进行随机落脚运动。In a possible implementation, the control module 1203 is used to control the torque of at least one joint among the several joints, so that the quadruped robot moves based on the basic movement pattern on the several desired footholds. Random foot movement.
其中，所述基本运动方式包括如下运动方式中的至少一种：Wherein, the basic movement patterns include at least one of the following movement patterns:
向第一方向移动，例如，四足机器人通过控制四足摆动，从而带动四足机器人向第一方向移动；Move in the first direction, for example, the quadruped robot controls the swing of the four legs to drive the quadruped robot to move in the first direction;
向第二方向转圈，例如，四足机器人通过控制四足朝同一方向旋转，从而带动四足机器人转圈；Turn in a circle in the second direction. For example, the quadruped robot controls the four legs to rotate in the same direction, thereby driving the quadruped robot to turn in a circle;
四个足部内部收缩，例如，四足机器人通过控制四足均收缩至基座部的底部；The four feet contract internally. For example, a quadruped robot controls all four feet to contract to the bottom of the base;
四个足部外部伸展，例如，四足机器人通过控制四足向外延伸；External extension of the four legs, for example, a quadruped robot extends outward by controlling the four legs;
四个足部中的两个足部内部收缩，例如，四足机器人通过控制四足中的任意两个足部收缩至基座部的底部；Two of the four feet are contracted internally. For example, a quadruped robot controls any two of the four feet to contract to the bottom of the base;
四个足部中的两个足部外部伸展，例如，四足机器人通过控制四足中的任意两个足部外伸远离基座部；Two of the four legs are externally extended. For example, a quadruped robot controls any two of the four legs to extend away from the base;
四个足部中的单个足部内部收缩，例如，四足机器人通过控制四足中的任意一个足部收缩至基座部的底部；The internal contraction of a single foot among the four feet, for example, a quadruped robot controls any one of the four feet to contract to the bottom of the base;
四个足部中的单个足部外部伸展，例如，四足机器人通过控制四足中的任意一个足部外伸远离基座部；The external extension of a single foot among the four legs, for example, a quadruped robot controls the external extension of any one of the four legs away from the base;
基座部往复晃动，例如，四足机器人在控制四足移动的过程中，基座部在往复晃动，产生摇头晃脑的效果；The base is rocking back and forth. For example, when a quadruped robot is controlling the movement of its four legs, the base is rocking back and forth, producing a head and head shaking effect;
四个足部的以变化步频方式运动，例如，四足机器人在控制四足移动的过程中，以变化步频进行移动；The four legs move with varying cadences. For example, a quadruped robot moves with varying cadences during the process of controlling the movement of its four legs;
四个足部的以变化步态方式运动，例如，四足机器人在控制四足移动的过程中，以变化步态方式进行移动，比如，直线型步态、S型步态。The four legs move in a changing gait. For example, a quadruped robot moves in a changing gait during the process of controlling the movement of its four legs, such as linear gait and S-shaped gait.
在一种可能的实现方式中，确定模块1202，用于基于所述四足机器人在所述基本运动方式下的运动轨迹，确定所述四足机器人的轨迹落脚点；In a possible implementation, the determination module 1202 is configured to determine the trajectory footing point of the quadruped robot based on the movement trajectory of the quadruped robot in the basic movement mode;
将所述轨迹落脚点按照偏移量进行偏移，得到偏移落脚点；Offset the trajectory footing point according to the offset amount to obtain the offset footing point;
在所述若干个候选落脚点中选择与所述偏移落脚点最近的候选落脚点，确定为基于所述基本运动方式进行随机落脚运动时的所述期望落脚点；Select the candidate foothold point closest to the offset foothold point among the several candidate foothold points, and determine it as the expected foothold point when performing random foothold movements based on the basic movement pattern;
其中，所述偏移量是在以所述轨迹落脚点为基准位置，以所述关节的活动范围为选择范围内确定的偏移量。Wherein, the offset amount is determined within the range of the starting point of the trajectory as the reference position and the range of motion of the joint as the selection range.
在一种可能的实现方式中，确定模块1202，用于在所述偏移落脚点的平面坐标与所述候选落脚点的平面坐标之间的距离最近的情况下，将所述候选落脚点确定为基于所述基本运动方式进行随机落脚运动时的所述期望落脚点。In a possible implementation, the determination module 1202 is configured to determine the candidate foothold point when the distance between the plane coordinates of the offset foothold point and the plane coordinates of the candidate foothold point is the shortest. It is the expected foothold when performing random foot movement based on the basic movement pattern.
在一种可能的实现方式中，确定模块1202，用于将与四个足部中的每个足部对应的所述轨迹落脚点按照与所述每个足部对应的偏移量进行偏移，得到与所述每个足部对应的偏移落脚点。In a possible implementation, the determination module 1202 is configured to offset the trajectory footing point corresponding to each of the four feet according to an offset corresponding to each foot. , to obtain the offset footing point corresponding to each foot.
在一种可能的实现方式中，确定模块1202，用于将与两个对角足部中的每个足部对应的所述轨迹落脚点按照与所述每个足部对应的偏移量进行偏移，得到与所述每个足部对应的偏移落脚点；In a possible implementation, the determination module 1202 is configured to determine the trajectory footing point corresponding to each of the two diagonal feet according to the offset corresponding to each foot. Offset to obtain the offset footing point corresponding to each foot;
其中，所述两个对角足部是所述四个足部中位于同一对角线上的两个足部。Wherein, the two diagonal feet are two feet located on the same diagonal line among the four feet.
在一种可能的实现方式中，确定模块1202，用于将与单个足部对应的所述轨迹落脚点按照与所述单个足部对应的偏移量进行偏移，得到与所述单个足部对应的偏移落脚点；In a possible implementation, the determination module 1202 is configured to offset the trajectory footing point corresponding to a single foot according to an offset amount corresponding to the single foot, to obtain an offset corresponding to the single foot. The corresponding offset footing point;
其中，所述单个足部是所述四个足部中的一个足部。Wherein, the single foot is one of the four feet.
在一种可能的实现方式中，确定模块1202，用于在以所述轨迹落脚点为基准位置，以所述关节的活动范围为选择范围内随机生成所述偏移量。In a possible implementation, the determination module 1202 is configured to randomly generate the offset within a selection range using the trajectory footing point as the reference position and the range of motion of the joint as the selection range.
在一种可能的实现方式中，控制模块1203，用于控制所述若干个关节中的至少一个关节的力矩，以使得所述四足机器人在所述若干个期望落脚点上基于基本运动方式进行运动。In a possible implementation, the control module 1203 is used to control the torque of at least one joint among the several joints, so that the quadruped robot moves based on the basic movement pattern on the several desired footholds. sports.
在一种可能的实现方式中，确定模块1202，用于基于所述四足机器人在所述基本运动方式下的运动轨迹，确定所述四足机器人的轨迹落脚点。In a possible implementation, the determination module 1202 is configured to determine the trajectory footing point of the quadruped robot based on the movement trajectory of the quadruped robot in the basic movement mode.
在一种可能的实现方式中，确定模块1202，用于在所述若干个候选落脚点中选择与所述轨迹落脚点最近的候选落脚点，确定为基于所述基本运动方式进行运动时的所述期望落脚点。In a possible implementation, the determination module 1202 is configured to select the candidate foothold closest to the trajectory footpoint among the several candidate footholds, and determine it as the foothold point when exercising based on the basic movement mode. Describe the starting point of expectations.
在一种可能的实现方式中，确定模块1202，用于在所述轨迹落脚点的平面坐标与所述候选落脚点的平面坐标之间的距离小于第二距离阈值的情况下，将所述候选落脚点确定为基于所述基本运动方式进行运动时的所述期望落脚点。In a possible implementation, the determination module 1202 is configured to determine the candidate destination point when the distance between the plane coordinates of the trajectory footing point and the plane coordinates of the candidate footing point is less than a second distance threshold. The foothold point is determined as the expected foothold point when exercising based on the basic movement pattern.
在一种可能的实现方式中，获取模块1201，用于通过摄像头采集和/或动作捕捉设备和/或定位设备获取所述四足机器人所处环境的所述地形信息。In a possible implementation, the acquisition module 1201 is configured to acquire the terrain information of the environment where the quadruped robot is located through camera acquisition and/or motion capture equipment and/or positioning equipment.
图13示出了本申请一示例性实施例示出的计算机设备1300的结构框图。该计算机设备可以实现为本申请上述方案中的服务器。所述图像计算机设备1300包括中央处理单元(Central Processing Unit，CPU)1301、包括随机存取存储器(Random Access Memory，RAM)1302和只读存储器(Read-Only Memory，ROM)1303的系统存储器1304，以及连接系统存储器1304和中央处理单元1301的系统总线1305。所述图像计算机设备1300还包括用于存储操作系统1309、应用程序1310和其他程序模块1311的大容量存储设备1306。Figure 13 shows a structural block diagram of a computer device 1300 according to an exemplary embodiment of the present application. The computer device can be implemented as the server in the above solution of this application. The image computer device 1300 includes a central processing unit (Central Processing Unit, CPU) 1301, a system memory 1304 including a random access memory (Random Access Memory, RAM) 1302 and a read-only memory (Read-Only Memory, ROM) 1303, and a system bus 1305 connecting the system memory 1304 and the central processing unit 1301. The imaging computer device 1300 also includes a mass storage device 1306 for storing an operating system 1309, applications 1310 and other program modules 1311.
所述大容量存储设备1306通过连接到系统总线1305的大容量存储控制器(未示出)连接到中央处理单元1301。所述大容量存储设备1306及其相关联的计算机可读介质为图像计算机设备1300提供非易失性存储。也就是说，所述大容量存储设备1306可以包括诸如硬盘或者只读光盘(Compact Disc Read-Only Memory，CD-ROM)驱动器之类的计算机可读介质(未示出)。The mass storage device 1306 is connected to the central processing unit 1301 through a mass storage controller (not shown) connected to the system bus 1305 . The mass storage device 1306 and its associated computer-readable media provide non-volatile storage for the image computing device 1300 . That is, the mass storage device 1306 may include a computer-readable medium (not shown) such as a hard disk or a Compact Disc Read-Only Memory (CD-ROM) drive.
不失一般性，所述计算机可读介质可以包括计算机存储介质和通信介质。计算机存储介质包括以用于存储诸如计算机可读指令、数据结构、程序模块或其他数据等信息的任何方法或技术实现的易失性和非易失性、可移动和不可移动介质。计算机存储介质包括RAM、ROM、可擦除可编程只读寄存器(Erasable Programmable Read Only Memory，EPROM)、电子抹除式可复写只读存储器(Electrically-Erasable Programmable Read-OnlyMemory，EEPROM)闪存或其他固态存储其技术，CD-ROM、数字多功能光盘(DigitalVersatile Disc，DVD)或其他光学存储、磁带盒、磁带、磁盘存储或其他磁性存储设备。当然，本领域技术人员可知所述计算机存储介质不局限于上述几种。上述的系统存储器1304和大容量存储设备1306可以统称为存储器。Without loss of generality, the computer-readable media may include computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media include RAM, ROM, Erasable Programmable Read Only Memory (EPROM), Electronically Erasable Programmable Read-Only Memory (EEPROM) flash memory or other solid state Storage technology, CD-ROM, Digital Versatile Disc (DVD) or other optical storage, tape cassette, magnetic tape, disk storage or other magnetic storage device. Of course, those skilled in the art will know that the computer storage media is not limited to the above types. The above-mentioned system memory 1304 and mass storage device 1306 may be collectively referred to as memory.
根据本公开的各种实施例，所述图像计算机设备1300还可以通过诸如因特网等网络连接到网络上的远程计算机运行。也即图像计算机设备1300可以通过连接在所述系统总线1305上的网络接口单元1307连接到网络1308，或者说，也可以使用网络接口单元1307来连接到其他类型的网络或远程计算机系统(未示出)。According to various embodiments of the present disclosure, the image computer device 1300 may also operate on a remote computer connected to a network through a network such as the Internet. That is, the image computer device 1300 can be connected to the network 1308 through the network interface unit 1307 connected to the system bus 1305, or the network interface unit 1307 can also be used to connect to other types of networks or remote computer systems (not shown). out).
所述存储器还包括至少一段计算机程序，所述至少一段计算机程序存储于存储器中，中央处理器1301通过执行该至少一段程序来实现上述各个实施例所示的四足机器人的控制方法。The memory also includes at least one section of a computer program. The at least one section of the computer program is stored in the memory. The central processor 1301 executes the at least one section of the program to implement the control method of the quadruped robot shown in the above embodiments.
本申请实施例还提供一种四足机器人，该四足机器人包括处理器和存储器，该存储器中存储有至少一条指令，该至少一条指令由处理器加载并执行以实现上述各方法实施例提供的四足机器人的控制方法。Embodiments of the present application also provide a quadruped robot. The quadruped robot includes a processor and a memory. At least one instruction is stored in the memory. The at least one instruction is loaded and executed by the processor to implement the methods provided by the above method embodiments. Control methods for quadruped robots.
本申请实施例还提供一种计算机可读存储介质，该存储介质中存储有至少一条程序，该至少一条程序由处理器加载并执行以实现上述各方法实施例提供的四足机器人的控制方法。Embodiments of the present application also provide a computer-readable storage medium, which stores at least one program. The at least one program is loaded and executed by a processor to implement the control method of a quadruped robot provided by the above method embodiments.
本申请实施例还提供一种计算机程序产品，所述计算机程序产品包括计算机程序，所述计算机程序存储在计算机可读存储介质中；所述计算机程序由计算机设备的处理器从所述计算机可读存储介质读取并执行，使得所述计算机设备执行上述各方法实施例提供的四足机器人的控制方法。Embodiments of the present application also provide a computer program product. The computer program product includes a computer program. The computer program is stored in a computer-readable storage medium. The computer program is readable by a processor of a computer device from the computer. The storage medium is read and executed, so that the computer device executes the control method of the quadruped robot provided by the above method embodiments.
本申请实施例还提供一种芯片，所述芯片包括可编程逻辑电路或程序，安装有所述芯片的设备用于实现如上所述的四足机器人的控制方法。An embodiment of the present application also provides a chip, which includes a programmable logic circuit or program, and a device equipped with the chip is used to implement the control method of a quadruped robot as described above.
可以理解的是，在本申请的具体实施方式中，涉及到的数据，历史数据，以及画像等与对象身份或特性相关的对象数据处理等相关的数据，当本申请以上实施例运用到具体产品或技术中时，需要获得对象许可或者同意，且相关数据的收集、使用和处理需要遵守相关国家和地区的相关法律法规和标准。It can be understood that in the specific implementation of this application, the data involved, historical data, and portraits and other data related to object data processing related to the identity or characteristics of the object, when the above embodiments of this application are applied to specific products or technology, the permission or consent of the subject needs to be obtained, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions.
应当理解的是，在本文中提及的“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。字符“/”一般表示前后关联对象是一种“或”的关系。It should be understood that "plurality" mentioned in this article means two or more. "And/or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and/or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the related objects are in an "or" relationship.
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。Those of ordinary skill in the art can understand that all or part of the steps to implement the above embodiments can be completed by hardware, or can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium. The above-mentioned The storage media mentioned can be read-only memory, magnetic disks or optical disks, etc.
以上所述仅为本申请的可选实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同切换、改进等，均应包含在本申请的保护范围之内。The above are only optional embodiments of this application and are not intended to limit this application. Any modifications, equivalent switches, improvements, etc. made within the spirit and principles of this application shall be included in the protection of this application. within the range.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210877945.6ACN116985111B (en) | 2022-07-25 | 2022-07-25 | Control method and device of quadruped robot, robot and storage medium |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210877945.6ACN116985111B (en) | 2022-07-25 | 2022-07-25 | Control method and device of quadruped robot, robot and storage medium |
| Publication Number | Publication Date |
|---|---|
| CN116985111Atrue CN116985111A (en) | 2023-11-03 |
| CN116985111B CN116985111B (en) | 2025-07-25 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210877945.6AActiveCN116985111B (en) | 2022-07-25 | 2022-07-25 | Control method and device of quadruped robot, robot and storage medium |
| Country | Link |
|---|---|
| CN (1) | CN116985111B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106695791A (en)* | 2017-01-11 | 2017-05-24 | 山东科技大学 | Generation method of continuous free tread of four-foot biomimetic robot |
| CN107562052A (en)* | 2017-08-30 | 2018-01-09 | 唐开强 | A kind of Hexapod Robot gait planning method based on deeply study |
| CN107644441A (en)* | 2017-08-30 | 2018-01-30 | 南京大学 | Multi-foot robot complex road condition based on three-dimensional imaging is separated into point methods of stopping over |
| US20210039731A1 (en)* | 2019-08-06 | 2021-02-11 | Boston Dynamics, Inc. | Leg Swing Trajectories |
| CN112731941A (en)* | 2020-12-29 | 2021-04-30 | 深圳市优必选科技股份有限公司 | Biped robot path planning method and device and biped robot |
| CN114474034A (en)* | 2020-10-26 | 2022-05-13 | 腾讯科技（深圳）有限公司 | Method, device, equipment and medium for controlling motion of foot type robot |
| CN114489092A (en)* | 2020-10-26 | 2022-05-13 | 腾讯科技（深圳）有限公司 | Foot robot motion control method, device, equipment and medium |
| CN114723807A (en)* | 2021-01-04 | 2022-07-08 | 腾讯科技（深圳）有限公司 | Centroid trajectory determination method and device, foot type robot, equipment and medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106695791A (en)* | 2017-01-11 | 2017-05-24 | 山东科技大学 | Generation method of continuous free tread of four-foot biomimetic robot |
| CN107562052A (en)* | 2017-08-30 | 2018-01-09 | 唐开强 | A kind of Hexapod Robot gait planning method based on deeply study |
| CN107644441A (en)* | 2017-08-30 | 2018-01-30 | 南京大学 | Multi-foot robot complex road condition based on three-dimensional imaging is separated into point methods of stopping over |
| US20210039731A1 (en)* | 2019-08-06 | 2021-02-11 | Boston Dynamics, Inc. | Leg Swing Trajectories |
| CN114474034A (en)* | 2020-10-26 | 2022-05-13 | 腾讯科技（深圳）有限公司 | Method, device, equipment and medium for controlling motion of foot type robot |
| CN114489092A (en)* | 2020-10-26 | 2022-05-13 | 腾讯科技（深圳）有限公司 | Foot robot motion control method, device, equipment and medium |
| CN112731941A (en)* | 2020-12-29 | 2021-04-30 | 深圳市优必选科技股份有限公司 | Biped robot path planning method and device and biped robot |
| CN114723807A (en)* | 2021-01-04 | 2022-07-08 | 腾讯科技（深圳）有限公司 | Centroid trajectory determination method and device, foot type robot, equipment and medium |
| Publication number | Publication date |
|---|---|
| CN116985111B (en) | 2025-07-25 |
| Publication | Publication Date | Title |
|---|---|---|
| Fu et al. | From language to goals: Inverse reinforcement learning for vision-based instruction following | |
| Shen et al. | Multimode locomotion via SuperBot reconfigurable robots | |
| Mishkin et al. | Benchmarking classic and learned navigation in complex 3d environments | |
| Zhu et al. | Reinforcement and imitation learning for diverse visuomotor skills | |
| Tan et al. | Scorpio: A biomimetic reconfigurable rolling–crawling robot | |
| Knopp et al. | Formation control using GQ (λ) reinforcement learning | |
| CN109605345B (en) | Modular robot structure sensing method | |
| Ranaweera et al. | Bridging the reality gap between virtual and physical environments through reinforcement learning | |
| Allali et al. | Rhoban football club: Robocup humanoid kid-size 2016 champion team paper | |
| Zhao et al. | Guard: A safe reinforcement learning benchmark | |
| Kherici et al. | Using PSO for a walk of a biped robot | |
| JP2004298975A (en) | Robot device and obstacle searching method | |
| CN108762726B (en) | Basic framework development platform and method for designing game through platform | |
| CN116985111A (en) | Control method and device of quadruped robot, robot and storage medium | |
| CN114723807B (en) | Method, device, footed robot, equipment and medium for determining center of mass trajectory | |
| CN118123838A (en) | A robotic arm control system calibration method, storage device and electronic device based on deep reinforcement learning | |
| Huang et al. | NavigationNet: A large-scale interactive indoor navigation dataset | |
| Zhang et al. | Learning based gaits evolution for an AIBO dog | |
| JP2018171448A (en) | Learning device, learning method, learning program, moving image distribution device, object action device, object action program, and moving image generation device | |
| JP6326167B1 (en) | Learning device, learning method, learning program, moving image distribution device, activity device, activity program, and moving image generation device | |
| Ray | Cooperative robotics using wireless communication | |
| Bongard et al. | Avoiding local optima with interactive evolutionary robotics | |
| Sabe | Development of entertainment robot and its future | |
| Xu | From simulation to reality: Migration of humanoid robot control | |
| Liu et al. | A Reinforcement Learning Toolkit for Quadruped Robots With Pybullet |
| 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 |