Robot control method and apparatus, device and storage medium

The wheel-leg robot uses the alternate swing and whole-body dynamic model of mechanical leg sets in the same plane of the hip rotation center to obtain joint torque, which solves the problem of insufficient movement efficiency and accuracy of the wheel-leg robot on obstacles, and achieves dynamic balance Fast movement and precise control under.

WO2024234594A1PCT designated stage expired Publication Date: 2024-11-21TENCENT TECHNOLOGY (SHENZHEN) CO LTD

Patent Information

Application Number
PCT/CN2023/131610
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-17
Filing Date
2023-11-14
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

The existing wheel-leg robots are not efficient when moving on obstacles such as climbing stairs and lack control accuracy, which is mainly due to the restriction of the center of gravity projection in the support area, resulting in slow movement.

Method used

The robot control method performed by a computer device is used to swing alternately with the first and second mechanical leg groups with the hip rotation center located in the same vertical plane, and combine the whole-body dynamic model to obtain the desired joint moment set to realize the robot in a dynamic equilibrium state Fast movement and precise control.

Benefits of technology

The robot's movement efficiency and control accuracy are improved, so that the center of gravity can exceed the support area, achieve rapid movement, and ensure the effective completion of the task through the accurate calculation of the expected joint moment set.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2023131610_21112024_PF_FP_ABST
    Figure CN2023131610_21112024_PF_FP_ABST
Patent Text Reader

Abstract

A robot control method and apparatus, a device, and a storage medium, relating to the technical field of artificial intelligence. The method comprises: for a robot having a first mechanical leg group and a second mechanical leg group, acquiring an expected operation space task of the robot on a supporting surface, the expected operation space task being used to guide the robot to alternately swing the first mechanical leg group and the second mechanical leg group so as to move on the supporting surface (601); according to the expected operation space task and a whole-body dynamics model of the robot, acquiring an expected joint torque set corresponding to the expected operation space task, the expected joint torque set comprising expected joint torque used to control each part of the robot (602); and controlling the robot to move under the guidance of the expected operation space task according to the expected joint torque set (603). The control accuracy of the robot is improved by means of the whole-body dynamics model, and the movement efficiency of the robot is improved by means of the expected operation space task.
Need to check novelty before this filing date? Find Prior Art

Description

Robot control method, device, equipment and storage medium

[0001] This application claims priority to Chinese patent application No. 202310558206.5 filed on May 17, 2023, entitled “Robot control method, device, equipment and storage medium”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] The embodiments of the present application relate to the field of artificial intelligence technology, and in particular to a robot control method, device, equipment, and storage medium. Background Art

[0003] With the development of robot control technology, some organizations and research institutions have successively launched wheel-legged robots with wheels as feet. These wheel-legged robots can not only slide quickly on their wheels, but also climb stairs and overcome obstacles.

[0004] Taking wheel-legged robots climbing stairs as an example, related technologies use heuristic methods to control wheel-legged robots, such as planning the position of the wheel-legged robot's mechanical legs on the stairs and controlling the projection of the wheel-legged robot's center of gravity to move slowly within its support area to achieve stair climbing. This type of method is static stair climbing, that is, the projection of the robot's center of gravity will not exceed the robot's support area, resulting in low efficiency of the robot climbing stairs.

[0005] Summary of the Invention

[0006] The embodiments of the present application provide a robot control method, apparatus, device, and storage medium, which can improve the robot's movement efficiency and the robot's control accuracy. The technical solution is as follows:

[0007] According to one aspect of an embodiment of the present application, a method for controlling a robot is provided, the method being executed by a computer device. The robot includes a body, a first mechanical leg group and a second mechanical leg group connected to the body via hip joints, at least one of the first mechanical leg group and the second mechanical leg group includes at least two mechanical legs, and the rotation centers of the hip joints corresponding to the first mechanical leg group and the rotation centers of the hip joints corresponding to the second mechanical leg group are located in the same vertical plane. The method includes:

[0008] Obtaining an expected operation space task of the robot on a support surface, the expected operation space task including expected accelerations of various parts of the robot in the operation space of the robot, the expected operation space task being used to guide the robot to alternately swing the first and second mechanical leg groups to move on the support surface;

[0009] According to the expected operation space task and the whole-body dynamics model of the robot, obtaining an expected joint torque set corresponding to the expected operation space task, the expected joint torque set including expected joint torques for controlling various parts of the robot;

[0010] According to the expected joint torque set, the robot is controlled to move under the guidance of the expected operation space task.

[0011] According to one aspect of an embodiment of the present application, a control device for a robot is provided, wherein the robot includes a body, a first mechanical leg group and a second mechanical leg group connected to the body via hip joints, at least one of the first mechanical leg group and the second mechanical leg group includes at least two mechanical legs, and the rotation center of the hip joint corresponding to the first mechanical leg group and the rotation center of the hip joint corresponding to the second mechanical leg group are located in the same vertical plane; the device includes:

[0012] an expected task acquisition module, configured to acquire an expected operation space task of the robot on a support surface, the expected operation space task including expected accelerations of various parts of the robot in the robot's operation space, the expected operation space task being used to guide the robot to alternately swing the first and second mechanical leg groups to move on the support surface;

[0013] an expected torque acquisition module, configured to acquire an expected joint torque set corresponding to the expected operation space task based on the expected operation space task and a whole-body dynamics model of the robot, wherein the expected joint torque set includes expected joint torques for controlling various parts of the robot;

[0014] The robot movement module is used to control the robot to move under the guidance of the expected operation space task according to the expected joint torque set.

[0015] According to one aspect of an embodiment of the present application, a computer device is provided, wherein the robot includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the above-mentioned robot control method.

[0016] According to one aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored. The computer program is loaded and executed by a processor to implement the above-mentioned robot control method.

[0017] According to one aspect of an embodiment of the present application, a computer program product is provided, the computer program product including a computer program stored in a computer-readable storage medium. A processor of a computer device reads the computer program from the computer-readable storage medium and executes the computer program, causing the computer device to perform the above-described robot control method.

[0018] The technical solutions provided by the embodiments of the present application may have the following beneficial effects:

[0019] For robots with a first set of mechanical legs and a second set of mechanical legs, since the rotation centers of their corresponding hip joints lie in the same vertical plane, by planning the desired operational space task, the robot can maintain a standing position with one set of mechanical legs while swinging with the other set of mechanical legs, allowing for rapid movement in a dynamic equilibrium state (i.e., the robot's center of gravity can exceed the robot's support area), thereby improving the robot's mobility efficiency. Furthermore, controlling the robot's movement by alternating the swinging of the first and second sets of mechanical legs can further improve the robot's mobility efficiency.

[0020] In addition, by adopting the robot's whole-body dynamics model, the expected joint torque set corresponding to the expected operation space task can be accurately obtained according to the expected operation space task, and then the expected operation space task can be accurately achieved according to the expected joint torque set, thereby improving the control accuracy of the robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] FIG1 is a schematic diagram of an implementation environment of a solution provided by an embodiment of the present application;

[0022] FIG2 is a schematic diagram of a quadruped wheel-legged robot provided by one embodiment of the present application;

[0023] FIG3 is a schematic diagram of a wheel-legged robot climbing stairs provided by one embodiment of the present application;

[0024] FIG4 is a schematic diagram of a wheel-legged robot crossing a road shoulder provided by one embodiment of the present application;

[0025] FIG5 is a schematic diagram of a wheel-legged robot crossing a pit provided by one embodiment of the present application;

[0026] FIG6 is a schematic diagram of a robot control method provided by one embodiment of the present application;

[0027] FIG7 is a flowchart of a method for obtaining a desired operation space task according to an embodiment of the present application;

[0028] FIG8 is a schematic diagram of an inverted pendulum model of a wheel-legged robot provided in one embodiment of the present application;

[0029] FIG9 is a flowchart of a method for obtaining a desired joint torque set provided by one embodiment of the present application;

[0030] Figures 10-12 are schematic diagrams showing a quadruped wheel-legged robot climbing stairs;

[0031] FIG13 is a block diagram of a robot control device provided by one embodiment of the present application;

[0032] FIG14 is a block diagram of a robot control device provided by another embodiment of the present application;

[0033] FIG15 is a simplified structural block diagram of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0035] Artificial Intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also studies the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0036] Artificial intelligence (AI) technology is a comprehensive discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0037] The technical solution of this application mainly relates to robotics technology in artificial intelligence technology, mainly to intelligent control of robots. A robot is a mechanical and electronic device that can imitate certain human skills by combining mechanical transmission and modern microelectronics technology. Robots are developed on the basis of electronic, mechanical and information technology. A robot does not necessarily have to look like a human. As long as it can independently complete the tasks and commands assigned to it by humans, it belongs to the robot family. A robot is an automated machine that has some intelligent capabilities similar to those of humans or biological organisms, such as perception, planning, movement and coordination. It is an automated machine with high flexibility. With the development of computer technology and artificial intelligence technology, robots have been greatly improved in terms of function and technical level. Mobile robots and robot vision and touch technologies are typical representatives.

[0038] In the technical solution provided in the embodiments of the present application, the execution entity of each step can be a computer device, which refers to an electronic device with data calculation, processing and storage capabilities.

[0039] In some embodiments, the computer device can be a PC (Personal Computer) device such as a desktop computer or laptop computer for controlling the robot; it can also be a server for controlling the robot. The server can be an independent physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server providing cloud computing services. The computer device and the robot can be connected via physical lines, a network, or the like. For example, referring to FIG1 , the computer device 101 can calculate the expected joint torque set corresponding to the robot 103 based on the expected operation space task of the robot 103 (the expected operation space task includes the expected acceleration of each part of the robot 103 in the operation space of the robot 103), and control the movement of the robot 103 through the network 102 based on the expected joint torque set. For example, the computer device 101 can control the first mechanical leg group 104 and the second mechanical leg group 105 of the robot 103 to swing alternately based on the expected joint torque set corresponding to the expected operation space task of the robot 103, so that the robot 103 moves on the support surface.

[0040] In some embodiments, the computer device may also be the robot itself, that is, the execution subject of each step in the technical solution provided in the embodiments of the present application is the robot. For example, referring to Figure 1, the computer device 101 may send the expected operation space task of the robot 103 to the robot 103 via the network 102. The robot 103 calculates the corresponding expected joint torque set based on the expected operation space task and then moves according to the expected joint torque set. In some embodiments, the robot 103 may also automatically plan the expected operation space task according to the real environment to perform different tasks in the real environment, and the embodiments of the present application are not limited to this.

[0041] The robot in the embodiments of the present application may refer to a wheel-leg robot, a leg-foot robot, etc. A wheel-leg robot refers to a robot with wheels as feet, and a leg-foot robot refers to a robot with feet as feet. The embodiments of the present application do not limit this.

[0042] In some embodiments, the robot may include a body, a first mechanical leg group and a second mechanical leg group connected to the body through a hip joint, at least one of the first mechanical leg group and the second mechanical leg group includes at least two mechanical legs, such as the first mechanical leg group includes at least two mechanical legs, the first mechanical leg group may also include at least two mechanical legs, there are at least two mechanical legs in the first mechanical leg group, respectively located on both sides of the central axis (i.e., the sagittal plane) of the robot, and there are at least two mechanical legs in the second mechanical leg group, respectively located on both sides of the central axis of the robot, and the first mechanical leg group and the second mechanical leg group are distributed side by side, that is, the rotation center of the hip joint corresponding to the first mechanical leg group and the rotation center of the hip joint corresponding to the second mechanical leg group are located in the same vertical plane.

[0043] Exemplarily, the robot in the embodiments of the present application is described by taking a wheel-legged robot as an example.

[0044] The wheel-legged robot in the embodiment of the present application may include a body, an outer mechanical leg group (i.e., a first mechanical leg group) and an inner mechanical leg group (i.e., a second mechanical leg group) connected to the body through a hip joint. The outer mechanical leg group may include two outer mechanical legs, and the inner mechanical leg group may include at least one inner mechanical leg. Exemplarily, the wheel-legged robot is a four-legged wheel-legged robot, i.e., the wheel-legged robot includes two outer mechanical legs and two inner mechanical legs; the wheel-legged robot is also a three-legged wheel-legged robot, i.e., the wheel-legged robot includes two outer mechanical legs and one inner mechanical leg. Wherein, the hip joint corresponding to at least one inner mechanical leg in the inner mechanical leg group is located between the hip joints corresponding to the two outer mechanical legs in the outer mechanical leg group, and the rotation center of the hip joint corresponding to the outer mechanical leg and the rotation center of the hip joint corresponding to the inner mechanical leg are located in the same vertical plane. The wheel-legged robot can stand on the support surface through the outer mechanical legs or the inner mechanical legs. The wheel-legged robot can also slide on the support surface through the wheel on the outer mechanical legs or the wheel on the inner mechanical legs. It can also move (i.e. walk) on the support surface by controlling the outer mechanical leg group and the inner mechanical leg group to swing alternately.

[0045] For example, as shown in Figure 2, which shows a schematic structural diagram of a quadruped wheel-legged robot, the quadruped wheel-legged robot 200 may include: a body, hip joints, and mechanical legs.

[0046] The quadruped wheel-legged robot 200 has four mechanical legs: two outer mechanical legs 201 (i.e., a first mechanical leg group) and two inner mechanical legs 202 (i.e., a second mechanical leg group). The two inner mechanical legs 202 are located between the two outer mechanical legs 201, and the four mechanical legs can be independently extended and retracted in the direction shown in the figure. A wheel 203 is installed at the end of each of the four mechanical legs, and each wheel 203 can be driven independently. The quadruped wheel-legged robot 200 can stand up by using the two inner mechanical legs 202 or the two outer mechanical legs 201 to be in a double-wheeled support state; the quadruped wheel-legged robot 200 can stand up by using the two inner mechanical legs 202 and the two outer mechanical legs 201 at the same time to be in a quadruped wheeled support state, which is not limited in this embodiment of the present application.

[0047] In some embodiments, the two inner mechanical legs 202 may be implemented as a whole, that is, the quadruped wheel-legged robot 200 may be implemented as a tripod wheel-legged robot having only one inner mechanical leg.

[0048] The other end of each of the four mechanical legs is connected to a hip joint 204. Each mechanical leg can rotate around its respective hip joint 204 and maintain a coordinated motion. In this embodiment of the present application, the rotation centers of the hip joints 204 corresponding to the quadruped wheeled leg robot 200 are located in the same vertical plane 205, and the rotation planes of the mechanical legs corresponding to the quadruped wheeled leg robot 200 are parallel. The hip joints 204 corresponding to the two inner mechanical legs 202 are located between the hip joints 204 corresponding to the two outer mechanical legs 201.

[0049] In some embodiments, the hip joints 204 corresponding to the quadruped wheeled legged robot 200 can be coaxial, that is, the rotation centers of the hip joints 204 are located on the same straight line. The hip joints 204 corresponding to the quadruped wheeled legged robot 200 can also be non-coaxial, for example, the hip joints 204 corresponding to the two inner mechanical legs 202 are coaxial, and the hip joints 204 corresponding to the two outer mechanical legs 201 are coaxial, but the hip joints 204 corresponding to the two inner mechanical legs 202 are not coaxial with the hip joints 204 corresponding to the two outer mechanical legs 201.

[0050] In some embodiments, the hip joints 204 corresponding to the two outer mechanical legs 201 share a drive motor to enable synchronous movement of the two outer mechanical legs 201; and the hip joints 204 corresponding to the two inner mechanical legs 202 share a drive motor to enable synchronous movement of the two inner mechanical legs 202. In one feasible example, each hip joint 204 corresponding to the quadruped wheeled leg robot 200 can also be independently driven by its own corresponding drive motor, which is not limited in this embodiment of the application.

[0051] The body of the quadruped wheeled-legged robot 200 may include a waist 206 , a trunk 207 , a head 208 and upper limbs 209 .

[0052] Each hip joint 204 of the quadruped wheeled-legged robot 200 is connected to the same end of the waist 206, and the other end of the waist 206 is connected to one end of the torso 207. The waist 206 has two rotation centers: a pitch rotation center that causes the torso 207 to pitch, and a roll rotation center that causes the torso 207 to swing. The roll rotation center is designed to be connected in series with the pitch rotation center and is located above the pitch rotation center, connecting to the torso 207.

[0053] The other end of the torso 207 is connected to the head 208 and the upper limb 209, and the upper limb 209 can be an upper limb with multiple degrees of freedom. In some embodiments, an end effector, such as a robotic gripper, a suction cup, etc., is deployed on the upper limb 209. Data acquisition equipment can be deployed in the head 208 to perceive the real environment, such as an image acquisition device, a video shooting device, an IMU (Inertial Measurement Unit), etc. Among them, the IMU can be placed at the geometric center of the torso 207, the center point of the hip joint (i.e., the rotation center of the hip joint), etc., which can be used to measure the actual acceleration, actual posture angular velocity, actual Euler angle, etc. of the torso 207.

[0054] In the technical solution provided in the embodiment of the present application, the wheels, mechanical legs, hip joints, waist and torso (including IMU) of the quadruped wheeled robot 200 are necessary hardware for the control algorithm, and the rest are non-essential hardware.

[0055] Compared with bipedal wheeled-legged robots, quadrupedal wheeled-legged robots have a more stable structure and are more resistant to external impact disturbances. Compared with hexapod wheeled-legged robots, they have fewer redundant joints and lower design complexity. They can bear large loads, pass through narrow spaces, and perform tasks on objects of different heights. Quadrupedal wheeled-legged robots have strong adaptability to the environment.

[0056] The robot control method provided in the embodiments of the present application is applicable to a variety of scenarios, such as climbing stairs, crossing thresholds, crossing curbs, crossing pits, and any other obstacle-crossing scenario. Robots employing the technical solutions provided in the embodiments of the present application have enhanced environmental adaptability. The robot control method provided in the embodiments of the present application can improve the robot's mobility efficiency and control accuracy.

[0057] The following will take a wheel-legged robot as an example to illustrate the application scenarios of the technical solutions provided in the embodiments of this application.

[0058] In some embodiments, referring to FIG3 , when the wheeled-legged robot 301 needs to climb stairs, it can first plan the desired operation space task according to the stairs, then calculate the desired joint torque set corresponding to the desired operation space task, and then control the outer mechanical leg group 302 (i.e., the first mechanical leg group) and the inner mechanical leg group 303 (i.e., the second mechanical leg group) to swing alternately according to the desired joint torque set to complete the stair climbing. For example, the outer mechanical leg group 302 is first used as the supporting mechanical leg group and the inner mechanical leg group 303 is used as the swinging mechanical leg group, so that the robot 301 climbs the first step. Then, the inner mechanical leg group 303 is used as the supporting mechanical leg group and the outer mechanical leg group 302 is used as the swinging mechanical leg group, so that the robot 301 climbs the second step. The outer mechanical leg group 302 and the inner mechanical leg group 303 swing alternately in sequence to complete the stair climbing task.

[0059] In some embodiments, referring to FIG4 , when the wheel-legged robot 401 needs to cross a road shoulder, it can first plan a desired operation space task based on the road shoulder, then calculate the desired joint torque set corresponding to the desired operation space task, and then control the outer mechanical leg group 402 and the inner mechanical leg group 403 to swing alternately according to the desired joint torque set to complete the road shoulder crossing. For example, the inner mechanical leg group 403 is first used as the supporting mechanical leg group and the outer mechanical leg group 402 is used as the swinging mechanical leg group, so that the outer mechanical leg group 402 of the robot 401 climbs onto the road shoulder. Then, the outer mechanical leg group 402 is used as the supporting mechanical leg group and the inner mechanical leg group 403 is used as the swinging mechanical leg group, so that the robot 401 completely crosses the road shoulder. The inner mechanical leg group 403 and the outer mechanical leg group 402 swing alternately in sequence to complete the road shoulder crossing task.

[0060] In some embodiments, referring to FIG5 , when the wheel-legged robot 501 needs to cross a pit, it can first plan a desired operation space task based on the pit, then calculate the desired joint torque set corresponding to the desired operation space task, and then control the outer mechanical leg group 502 and the inner mechanical leg group 503 to swing alternately according to the desired joint torque set to complete the crossing of the pit. For example, the inner mechanical leg group 503 is first used as the supporting mechanical leg group and the outer mechanical leg group 502 is used as the swinging mechanical leg group, so that the outer mechanical leg group 502 of the robot 501 crosses the pit. Then, the outer mechanical leg group 502 is used as the supporting mechanical leg group and the inner mechanical leg group 503 is used as the swinging mechanical leg group, so that the robot 501 completely crosses the pit. The inner mechanical leg group 503 and the outer mechanical leg group 502 swing alternately in sequence to complete the crossing of the pit.

[0061] The following will use a method embodiment to illustrate the control method of the robot provided in the embodiment of the present application.

[0062] Please refer to Figure 6, which shows a flow chart of a robot control method provided by an embodiment of the present application. In this embodiment of the present application, the robot control method is described by taking the robot as an example of the execution subject of each step. The method may include at least one of the following steps (601-603):

[0063] Step 601, obtain the expected operation space task of the robot on the support surface, which includes the expected acceleration of each part of the robot in the robot's operation space. The expected operation space task is used to guide the robot to alternately swing the first mechanical leg group and the second mechanical leg group to move on the support surface.

[0064] The expected operation space task refers to the task of the robot in the operation space, which can be planned based on the robot and the real environment in which the robot is located. For example, by planning the expected movement of the robot on the support surface, the expected operation space task corresponding to the scene can be obtained.

[0065] The robot can control the movement of various parts of the robot using the desired acceleration in the desired operating space task to achieve movement of the robot on the support surface. For example, the robot can alternately swing the first and second mechanical leg groups to move on the support surface in the form of walking. The robot in the embodiment of the present application is the same as that described in the above embodiment and will not be repeated here.

[0066] The desired acceleration refers to the acceleration derived from the real environment and used to control the robot. The reference acceleration, described below, is a planned value used to guide the robot. The robot follows this desired acceleration to achieve the corresponding action. For example, if the desired acceleration of a robotic leg is used to guide the leg's swing, the robot can control the leg to follow this desired acceleration to achieve the desired swing.

[0067] In some embodiments, the desired operational space task includes the desired acceleration of each robot component at each control moment, so that the robot as a whole moves according to the planned trajectory. The control moment refers to the time at which the robot is controlled by a control signal. The control moments are arranged at specified time intervals, which can be set and adjusted based on actual usage requirements.

[0068] In the embodiments of the present application, the support surface may include only a single plane, such as a flat ground surface or a road, or may include multiple planes at different heights, such as stairs, a road surface with a shoulder, or a ground surface with a pothole, and the embodiments of the present application are not limited thereto. The desired acceleration in the desired operational space task can then be used to guide the robot to alternately swing the first and second mechanical leg groups to perform tasks such as walking, climbing stairs, crossing a shoulder, and crossing a pothole.

[0069] The robot's operating space is the Cartesian space corresponding to the robot. In task-oriented whole-body control (TWBC), the Cartesian space corresponding to the robot can be referred to as the robot's operating space. In the embodiments of the present application, each position in the robot's operating space can be represented based on the robot's world coordinate system.

[0070] For example, the robot's world coordinate system can be constructed with the contact point between the robot's foot and the support surface in the initial state as the origin, the horizontal direction as the x-axis, the vertical direction as the z-axis, and the direction perpendicular to both the horizontal and vertical directions as the y-axis. The robot's position in the operating space can be represented based on the robot's three-dimensional coordinates in the world coordinate system. In some embodiments, the calculation processes in the embodiments of the present application all occur in the robot's world coordinate system.

[0071] In some embodiments, the above-mentioned parts may include the first mechanical leg group, the second mechanical leg group and the body of the robot, and the expected operation space task includes the expected accelerations corresponding to the first mechanical leg group, the second mechanical leg group and the body respectively.

[0072] In some embodiments, during the movement of the robot, the mechanical leg group used for swinging is the swinging mechanical leg group, and the mechanical leg group used for supporting is the supporting mechanical leg group. For example, if the first mechanical leg group is used as support and the second mechanical leg group is swung in the forward direction of the robot, the first mechanical leg group is the supporting mechanical leg group and the second mechanical leg group is the swinging mechanical leg group; if the second mechanical leg group is used as support and the first mechanical leg group is swung in the forward direction of the robot, the second mechanical leg group is the supporting mechanical leg group and the first mechanical leg group is the swinging mechanical leg group.

[0073] The supporting mechanical leg group and the swinging mechanical leg group corresponding to each step of the robot can be determined from the first mechanical leg group and the second mechanical leg group based on the robot's gait information. This gait information can be used to indicate whether the robot's mechanical legs are swinging mechanical legs during each step cycle. This gait information can be planned in real time or pre-set, and this application does not limit this. The step cycle is used to indicate the time it takes for the first mechanical leg group or the second mechanical leg group to complete a swing.

[0074] For example, referring to Figure 3, during the process of robot 301 climbing stairs, the first mechanical leg group 302 and the second mechanical leg group 303 are alternately used for swinging, and at the same time, the second mechanical leg group 303 and the first mechanical leg group 302 are synchronously alternately used for support, so that the robot 301 completes the stair climbing.

[0075] In some embodiments, the expected acceleration included in the expected operation space task can be calculated by a feedback controller based on the reference movement trajectory of each part of the robot and the actual state of the robot, such as a PD (Proportional Derivative) feedback controller and other feedback controllers.

[0076] Referring to FIG7 , the above step 601 may further include the following sub-steps:

[0077] Step 601a: Obtain a first expected acceleration of the swinging mechanical leg group in the operating space according to a swinging reference movement trajectory corresponding to the swinging mechanical leg group. The swinging reference movement trajectory is obtained by planning the movement trajectory of the swinging mechanical leg group according to the support surface.

[0078] The swing reference trajectory refers to the reference trajectory of the swinging mechanical legs in the swinging mechanical leg group, such as the reference trajectory of the foot of the swinging mechanical leg. The reference trajectory is used to guide the movement of the robot. For example, the reference trajectory may include the reference position of the robot at various control moments. The reference position refers to the position that the robot is expected to reach.

[0079] For example, the initial position and end position of the swinging mechanical leg within a stride cycle can be planned based on the size information of the supporting surface, and then the spline curve interpolation method can be used to interpolate the initial position and end position within the stride cycle to obtain the swinging reference movement trajectory within the stride cycle. The swinging reference movement trajectory within each stride cycle is then spliced ​​to obtain the final swinging reference movement trajectory.

[0080] For example, taking a four-legged wheeled robot climbing stairs as an example, based on the size information of each step of the stairs (such as width, height, length, etc.), the reference position of the swinging mechanical leg on each step can be planned (represented by three-dimensional coordinates in the world coordinate system), and then interpolation is performed between each reference position to obtain the swinging reference movement trajectory.

[0081] The first expected acceleration refers to the expected acceleration of the swinging mechanical leg at each control moment. For example, the expected acceleration of the foot of the swinging mechanical leg at each control moment can be determined as the first expected acceleration.

[0082] For example, when the robot stops moving after a plurality of control moments, the process of obtaining the first expected acceleration may be as follows:

[0083] 1. For each control moment, obtain the reference position, reference velocity, and reference acceleration of the swinging robotic leg group at the control moment according to the swinging reference movement trajectory.

[0084] In some embodiments, the reference position of the swinging robotic leg group at a certain control moment can be directly determined based on the swing reference movement trajectory. Furthermore, based on the reference position of the swinging robotic leg group at that control moment, the reference velocity and reference acceleration of the swinging robotic leg group at that control moment can be obtained. For example, by taking the first-order and second-order derivatives of the reference position of the swinging robotic leg group at that control moment, the reference velocity and reference acceleration of the swinging robotic leg group at that control moment can be obtained. The reference velocity refers to the desired velocity, and the reference acceleration refers to the desired acceleration, and they are not directly used to control the robot.

[0085] 2. Using a PD feedback controller, the first expected acceleration at the control moment is calculated based on the reference position, reference speed and reference acceleration of the swinging mechanical leg group at the control moment, as well as the actual position and actual speed of the swinging mechanical leg group at the control moment.

[0086] In some embodiments, based on the IMU on the fuselage, the acceleration, angular velocity, and attitude (i.e., actual Euler angles) of the fuselage can be measured. Then, combined with the actual angles and actual accelerations of all joints of the robot, as well as the contact information between the robot and the support surface, the actual position, actual velocity, actual attitude, and actual angular velocity of the fuselage in the world coordinate system can be obtained using a state estimation algorithm. Furthermore, if the relative position, relative velocity, and relative angular velocity between the hip joint center point and the IMU of the hip joint are known, the state of the hip joint center point can be calculated based on the state of the fuselage (i.e., the actual position, actual velocity, actual attitude, and actual angular velocity of the fuselage in the world coordinate system).

[0087] In an embodiment of the present application, the center point of the hip joint is the origin of the floating base coordinate system of the robot. Based on the state of the center point of the hip joint, the state of the floating base coordinate system can be obtained. Combined with the actual angles and actual angular velocities of all the joints of the robot, the forward kinematics of the full model of the robot can be used to calculate the actual position, actual speed, actual posture and actual angular velocity of all the links corresponding to the robot in the world coordinate system. All the links include the swinging mechanical legs, and the actual position and actual speed of the swinging mechanical legs in the operating space can be obtained. Among them, the actual position refers to the actual position of the swinging mechanical legs, which is a true value, and the actual speed refers to the actual speed of the swinging mechanical legs, which is a true value. The forward dynamics model can be used to indicate the relationship between the acceleration of the robot in the operating space and the speed and acceleration of the robot in the joint space of the robot.

[0088] For example, the first expected acceleration can be expressed as follows:

[0089] in, and are the reference position, reference velocity and reference acceleration of the swinging robotic leg in the operating space at the control time t, and are the actual position and actual speed of the swinging robot leg in the operating space at the control time t, k p,swing and k d,swing They correspond to the proportional coefficient and differential coefficient of the swinging mechanical leg respectively.

[0090] In some examples, the mechanical legs in the first mechanical leg group move synchronously, and the mechanical legs in the second mechanical leg group move synchronously, that is, the swinging mechanical legs in the swinging mechanical leg group move synchronously. If the robot has no displacement in the y-axis direction, then, ignoring the y-axis direction, the swinging reference movement trajectory of each swinging mechanical leg in the swinging mechanical leg group is the same. By calculating the first desired acceleration for any swinging mechanical leg, the first desired acceleration corresponding to the swinging mechanical leg group can be obtained. This reduces the computational effort required to calculate the first desired acceleration, thereby improving the control efficiency of the robot.

[0091] Step 601b: Obtain a second expected acceleration of the supporting mechanical leg group in the operating space according to the supporting reference movement trajectory corresponding to the supporting mechanical leg group. The supporting reference movement trajectory is planned based on the movement trajectory of the supporting mechanical leg group by the supporting surface.

[0092] The support reference movement trajectory refers to the reference movement trajectory of the support mechanical leg in the support mechanical leg group, such as the reference movement trajectory of the foot on the support mechanical leg. The second expected acceleration includes the expected acceleration of the support mechanical leg at each control moment.

[0093] In an embodiment of the present application, in each step cycle, the reference position of the supporting mechanical leg on the supporting surface is a fixed value, and during the supporting process of the supporting mechanical leg, there is no relative sliding between the foot of the supporting mechanical leg and the supporting surface, so the value of the second expected acceleration is always zero.

[0094] For example, the second expected acceleration can be expressed as follows:

[0095] Among them, N c is the number of supporting mechanical legs (the feet of each supporting mechanical leg in the supporting mechanical leg group are in contact with the supporting surface, N c It can also be recorded as the number of support contact points).

[0096] Step 601c: Obtain a third expected acceleration of the center of mass in the operating space according to a center of mass reference movement trajectory corresponding to the center of mass of the robot. The center of mass reference movement trajectory is planned based on the movement trajectory of the support surface relative to the center of mass.

[0097] The center of mass reference movement trajectory refers to the reference movement trajectory of the center of mass of the robot. The third expected acceleration includes the reference position of the center of mass of the robot at each control moment.

[0098] In some embodiments, during the movement of the robot, the center of mass of the robot not only needs to move continuously along the forward direction, but also needs to help the robot maintain dynamic balance. Therefore, a balance controller needs to be built to calculate the expected acceleration of the center of mass in the forward direction.

[0099] For example, the center of mass reference movement trajectory may be divided into a sub-reference movement trajectory in the forward direction and a sub-reference movement trajectory in the vertical direction. The third sub-expected acceleration in the forward direction is determined based on the sub-reference movement trajectory in the forward direction, and the third sub-expected acceleration in the vertical direction is determined based on the sub-reference movement trajectory in the vertical direction. This process may include the following:

[0100] 1. According to the sub-reference movement trajectory of the center of mass reference movement trajectory in the first direction, obtain the third sub-expected acceleration of the center of mass in the first direction.

[0101] The first direction may be the forward direction of the robot, such as the x-axis direction of the world coordinate system.

[0102] In some embodiments, the robot can be converted into an inverted pendulum model for building a balance controller.

[0103] For example, since the robot always moves in the first direction (such as no displacement in the y-axis direction of the above-mentioned world coordinate system), and the mechanical legs in the supporting mechanical leg group move synchronously, the mechanical legs in the swinging mechanical leg group move synchronously, and the hip joints of the robot are coaxial, the robot can be simplified into a plane model under the sagittal plane (such as the sagittal plane in Figure 2).

[0104] Referring to Figure 8, when the gravitational field in which the robot is located is uniform, the mass of the robot is concentrated at the center of mass 802 of the robot. The center of the line between the feet of the two supporting mechanical legs is set as the virtual support contact point 801 between the robot's inverted pendulum model and the supporting surface. By connecting the center of mass 802 and the virtual support contact point 801, the robot's plane model can be converted into an inverted pendulum model 800 of the robot.

[0105] After constructing the inverted pendulum model of the robot, a balance controller can be constructed based on the inverted pendulum model. Then, the third sub-expected acceleration in the first direction can be calculated based on the balance controller. The process can be as follows:

[0106] 1) The robot's inverted pendulum dynamics equation is constructed using the position of the center of mass, the velocity of the center of mass, the distance between the center of mass and the support contact point in the first direction, and the derivative of the distance as state variables, and the acceleration of the center of mass relative to the support contact point in the first direction as the control variable. The support contact point refers to the contact point between the foot corresponding to the supporting mechanical leg group and the support surface.

[0107] In the inverted pendulum model, the support contact point is represented by a virtual support contact point. For example, the dynamic equation of the inverted pendulum of the robot can be expressed as follows:

[0108] Among them, Δx and are the distance between the center of mass and the support contact point in the first direction, and the derivative of the distance, x com and are the position and velocity of the center of mass in the first direction, is the acceleration of the center of mass relative to the virtual support contact point in the first direction, is the position of the center of mass in the second direction, and g is the acceleration due to gravity.

[0109] 2) Use the linear quadratic programming regulator to calculate the feedback gain matrix of the inverted pendulum dynamics equation.

[0110] The Linear Quadratic Regulator (LQR) is built based on the Quadratic Programming optimization method. Its essence is to find a multidimensional vector under linear constraints so that the quadratic objective function of the multidimensional vector is minimized (or maximized).

[0111] For example, a linear quadratic programming regulator is used to construct an objective function of the inverted pendulum dynamics equation, and then the feedback gain matrix of the inverted pendulum dynamics equation is iteratively obtained with the goal of minimizing the objective function.

[0112] 3) For each control moment, according to the sub-reference movement trajectory in the first direction, obtain the reference position, reference speed, reference distance and reference speed between the center of mass and the support contact point at the control moment and in the first direction.

[0113] In some embodiments, the sub-reference movement trajectory of the center of mass in the first direction can be obtained based on the reference position planning of the supporting mechanical leg using heuristic (i.e., the trajectory is given based on experience and the real environment) or inverted pendulum model (the trajectory is planned using the inverted pendulum model), etc. The embodiments of the present application do not limit this.

[0114] Exemplarily, based on the sub-reference movement trajectory in the first direction, the reference position of the center of mass at the control moment and in the first direction can be directly obtained, and then the reference position of the center of mass at the control moment and in the first direction is taken as the first-order derivative to obtain the reference speed of the center of mass at the control moment and in the first direction.

[0115] According to the support reference movement trajectory, the reference position and reference speed of the support contact point at the control moment and the first direction can be obtained, and then the reference position of the virtual support contact point can be obtained by taking the geometric average of the reference position of the support contact point at the control moment and the first direction, and the reference speed of the virtual support contact point can be obtained by taking the geometric average of the reference speed of the support contact point at the control moment and the first direction. Then, the reference position of the virtual support contact point at the control moment and the first direction is subtracted from the reference position of the center of mass at the control moment and the first direction to obtain the reference distance between the center of mass and the virtual support contact point, and the reference speed of the virtual support contact point at the control moment and the first direction is subtracted from the reference speed of the center of mass at the control moment and the first direction to obtain the reference speed between the center of mass and the virtual support contact point.

[0116] 4) According to the feedback gain matrix, the reference position, reference speed, reference distance and reference speed between the center of mass and the support contact point at the control moment and the first direction, the third sub-expected acceleration at the control moment and the first direction is obtained.

[0117] In some embodiments, the reference position of the center of mass at the control time in the first direction, the reference speed, the reference distance between the center of mass and the support contact point, and the reference speed are used as reference values ​​of the control variables. Then, the control variables of the inverted pendulum dynamics equation can be obtained based on the reference values ​​of the control variables through LQR. This process can be expressed as follows:

[0118] in, is the actual value of the state variable.

[0119] Control variables of the inverted pendulum dynamics equation This is the third sub-expected acceleration in the first direction, recorded as The third sub-expected acceleration in the first direction obtained in this way can not only ensure the following of the reference movement trajectory of the center of mass in the first direction, but also maintain the dynamic balance of the robot during the movement, thereby improving the movement stability of the robot.

[0120] In some embodiments, the third sub-expected acceleration in the first direction can also be calculated using a PD feedback controller based on the reference position, reference velocity and reference acceleration of the center of mass in the first direction at the control moment, and the actual position and actual velocity of the center of mass in the first direction at the control moment. This embodiment of the present application is not limited to this.

[0121] 2. According to the sub-reference movement trajectory of the center of mass reference movement trajectory in the second direction, obtain the third sub-expected acceleration of the center of mass in the second direction, where the second direction is perpendicular to the first direction.

[0122] In some embodiments, the second direction may refer to a vertical direction, such as the z-axis direction of the world coordinate system.

[0123] In an embodiment of the present application, the height between the center of mass of the robot and the foot of the robot's supporting mechanical leg is set to a constant value. This constant value can refer to the distance between the center of mass of the robot and the center of the foot corresponding to the supporting mechanical leg (such as the wheel center) in the z-axis direction. Therefore, based on the supporting reference movement trajectory and this constant value, a sub-reference movement trajectory of the center of mass in the second direction can be planned. For example, using a spline curve interpolation method, based on the sum of the reference position in the supporting reference movement trajectory and the constant value, the sub-reference movement trajectory of the center of mass in the second direction is interpolated.

[0124] In some embodiments, the sub-reference movement trajectory in the second direction may be equivalent to the reference movement trajectory of the robot's body in the second direction. By following the sub-reference movement trajectory in the second direction, the body can be moved in the vertical direction to complete the adjustment of the center of mass position. Especially in the stair climbing scenario, by following the sub-reference movement trajectory in the second direction, the robot can climb the stairs in the vertical direction.

[0125] For example, when the robot stops moving after a plurality of control moments, the process of obtaining the third sub-expected acceleration in the second direction may be as follows:

[0126] 1. For each control moment, obtain the reference position, reference velocity, and reference acceleration of the center of mass at the control moment and in the second direction according to the sub-reference movement trajectory in the second direction.

[0127] In some embodiments, the reference position of the center of mass at a certain control moment and in the second direction can be determined directly based on the sub-reference movement trajectory in the second direction, and then the first-order derivative and second-order derivative of the reference position of the center of mass at the control moment and in the second direction can be performed respectively to obtain the reference velocity and reference acceleration of the center of mass at the control moment and in the second direction.

[0128] 2. Using a PD feedback controller, the third sub-expected acceleration in the second direction at the control moment is calculated based on the reference position, reference velocity and reference acceleration of the center of mass in the second direction at the control moment, as well as the actual position and actual velocity of the center of mass in the second direction at the control moment.

[0129] In some embodiments, based on the IMU on the fuselage, the acceleration, angular velocity and posture (i.e., the actual Euler angle) of the fuselage can be measured. Combined with the actual angles, actual accelerations of all joints of the robot, and the contact information between the robot and the supporting surface, the state estimation algorithm can be used to obtain the actual position, actual speed, actual posture and actual angular velocity of the fuselage in the world coordinate system. Finally, the actual position and actual speed of the fuselage in the world coordinate system are directly determined as the actual position and actual speed of the center of mass in the world coordinate system, and the actual position and actual speed of the center of mass in the second direction are obtained.

[0130] For example, the third sub-expected acceleration in the second direction can be expressed as follows:

[0131] in, and are the reference position, reference velocity and reference acceleration of the center of mass at the control time t and the second direction, respectively. and The actual position and actual speed of the center of mass at the control time t and the second direction, k p,base and k d,base They correspond to the proportional coefficient and differential coefficient of the center of mass respectively.

[0132] 3. Obtain a third expected acceleration based on the third sub-expected acceleration of the center of mass in the first direction and the third sub-expected acceleration of the center of mass in the second direction.

[0133] When determining the third expected acceleration, the embodiment of the present application comprehensively considers the third sub-expected accelerations corresponding to the center of mass in the first direction and the second direction, respectively, thereby improving the accuracy of the determined third expected acceleration.

[0134] For each control moment, the third expected acceleration of the center of mass in the first direction and the third expected acceleration of the center of mass in the second direction are combined to obtain the third expected acceleration.

[0135] Step 601d: Obtain a fourth expected acceleration of the robot body in the operating space according to a reference posture change trajectory of the robot body, wherein the reference posture change trajectory is obtained by planning the change trajectory of the body.

[0136] A reference posture change trajectory refers to a reference change trajectory of the posture of the robot's body, which can be used to describe the posture changes of the robot's body. For example, it can include the reference posture of the robot's body at various control moments. In some embodiments, the body's posture can be represented by Euler angles, such as the Euler angles of the body's roll, pitch, and yaw. The reference posture change trajectory can then refer to a corresponding Euler angle reference movement trajectory of the robot's body, which includes reference posture angles (i.e., reference Euler angles) at various control moments.

[0137] The fourth expected acceleration includes the expected acceleration of the robot body at each control moment. When the posture of the body is expressed using Euler angles, the expected acceleration may refer to the expected posture angular acceleration (ie, the expected Euler angle acceleration).

[0138] In some embodiments, when the robot stops moving after a plurality of control moments, the fourth expected acceleration may be obtained as follows:

[0139] 1. For each control moment, according to the attitude reference change trajectory, obtain the reference attitude angle, reference attitude angular velocity and reference attitude angular acceleration of the fuselage at the control moment.

[0140] In the embodiment of the present application, the robot body remains vertical during the movement of the robot, and the values ​​of the reference posture angle, reference posture angular velocity, and reference posture angular acceleration corresponding to the body are all zero. For example, the values ​​of the reference posture angle, reference posture angular velocity, and reference posture angular acceleration can be expressed as:

[0141] and

[0142] 2. Using a PD feedback controller, the fourth desired acceleration at the control moment is calculated based on the reference attitude angle, reference attitude angular velocity, and reference attitude angular acceleration of the fuselage at the control moment, as well as the actual attitude angle and actual attitude angular velocity of the fuselage at the control moment.

[0143] In some embodiments, based on the IMU on the fuselage, the acceleration, angular velocity and posture of the fuselage (i.e., the actual Euler angle) can be measured. Combined with the actual angles, actual accelerations of all joints of the robot, and the contact information between the robot and the supporting surface, the state estimation algorithm can be used to obtain the actual position, actual speed, actual posture and actual angular velocity (i.e., the actual state) of the fuselage in the world coordinate system. Finally, based on the actual state of the fuselage in the world coordinate system, the actual posture angle (i.e., the actual Euler angle) and actual posture angular velocity (i.e., the actual Euler angular velocity) of the fuselage can be obtained.

[0144] For example, the fourth expected acceleration can be expressed as follows:

[0145] in, and They are the actual attitude angle and actual attitude angular velocity of the fuselage in the operating space at the control time t, k p,euler and k d,euler They correspond to the proportional coefficient and differential coefficient of the fuselage attitude respectively.

[0146] Step 601e: Obtain the expected operation space task according to the first expected acceleration, the second expected acceleration, the third expected acceleration, and the fourth expected acceleration.

[0147] In the embodiment of the present application, the expected operation space task is acquired through the first expected acceleration, the second expected acceleration, the third expected acceleration, and the fourth expected acceleration, thereby improving the accuracy of the acquired expected operation space task.

[0148] In some embodiments, the desired action space task may be expressed as follows:

[0149] Step 602 : acquiring an expected joint torque set corresponding to the expected operation space task according to the expected operation space task and the whole-body dynamics model of the robot. The expected joint torque set includes expected joint torques for controlling various parts of the robot.

[0150] In some embodiments, the above-mentioned expected operation space task can be used for task-oriented whole body control of the robot, that is, considering the mass, inertia and other dynamic information of all the links of the robot, mobilizing all the degrees of freedom of the robot, and controlling the robot to complete one or more set tasks.

[0151] For example, based on the rigid body dynamics of the robot, the whole body dynamics model of the robot in the joint space system can be expressed as follows:

[0152] in, represents the robot's joint space inertia matrix, Represents the joint space offset force vector of the robot, which is the sum of the Coriolis force, centrifugal force and gravity corresponding to the robot. represents the robot's selection matrix, represents the Jacobian matrix of the contact points between the robot and the support surface, represents the active joint torque vector of the robot, represents the contact force vector of the robot, q, Represents the generalized position vector, generalized velocity vector and generalized acceleration vector at each degree of freedom of the robot, N G Represents the total degree of freedom of the robot, that is, the floating basis degree of freedom N F and the active joint degrees of freedom N J The sum, N C Indicates the number of contact forces corresponding to the robot, that is, the number of contact points n C With the dimension N of a single contact force D ∈{0, 1, 2, 3}.

[0153] Taking the quadruped wheeled leg robot in Figure 2 as an example, the connecting rods of the quadruped wheeled leg robot can include four mechanical legs, four wheels, the waist of the fuselage (a separate connecting root), and the torso, upper limbs, and head of the fuselage (the three are integrated). Among them, the active joints of the quadruped wheeled leg robot can include two hip joints, linear extension joints corresponding to the four mechanical legs, rotation joints of the four wheels, and one pitch rotation joint (pitch + yaw). In some embodiments, the robot's active joint degrees of freedom are the degrees of freedom corresponding to the above 11 active joints.

[0154] The floating base degrees of freedom (DFOs) refer to the robot's floating base coordinate system's six degrees of freedom (position, px, py, pz, and attitude, yaw, roll, and pitch) in the world coordinate system. The floating base coordinate system is constructed with the hip joint's center of rotation as its origin. Initially, the coordinate axes of the floating base coordinate system are aligned with those of the world coordinate system. The number of contact points can refer to the number of supporting robotic legs. For example, during the swinging motion of a swinging robotic leg group, the number of contact points is two.

[0155] In some embodiments, referring to FIG9 , step 602 may further include the following sub-steps:

[0156] Step 602a: Obtain a forward dynamics model of the robot. The forward dynamics model is used to indicate the relationship between the acceleration of the robot in the operation space and the velocity and acceleration of the robot in the joint space of the robot.

[0157] In some embodiments, based on rigid body dynamics, the forward dynamics model of the robot can be expressed as follows:

[0158] in, represents the acceleration of the robot in the operating space at the control time t, and They represent the velocity and acceleration of the robot in the joint space, that is, the generalized position vector and generalized velocity vector at each degree of freedom of the robot, J t and Represents the Jacobian matrix and the first-order derivative of the Jacobian matrix corresponding to the operation space.

[0159] Step 602b: The whole-body dynamics model and the forward dynamics model are combined to obtain a dynamics equation to be solved. The dynamics equation to be solved uses the acceleration of the robot in the joint space as an unknown variable.

[0160] The whole-body dynamics model and the forward dynamics model are combined and simplified to obtain:

[0161] in, is an unknown variable, and the rest are known variables.

[0162] Step 602c: Substitute the desired operation space task into the dynamic equation to be solved, and obtain the desired joint torque set.

[0163] In some embodiments, the robot is also subject to physical limitations of the robot's body structure and drive motors during motion. To improve the rationality and accuracy of the expected joint torque set, the present application further sets constraints during the process of solving the expected joint torque set. Step 602c may further include the following:

[0164] 1. Replace the desired operation space task with the acceleration of the robot in the operation space in the dynamic equation to be solved to obtain the intermediate dynamic equation.

[0165] In some embodiments, the above The acceleration of the robot in the operating space in the dynamic equation to be solved By making substitutions, we can obtain the intermediate dynamic equation.

[0166] 2. Construct the joint physical constraint expression and friction constraint expression of the robot. The joint physical constraint expression is used to constrain each joint of the robot. The contact force between the robot and the support surface under the constraint of the friction constraint expression satisfies the friction cone constraint.

[0167] In some embodiments, the active joint torque in the unknown variable is limited according to the actual physical characteristics of the robot's driving motor, that is, the joint physical constraint expression can be: lb ≤τ≤τ ub ; where τ lb and τ ub Respectively represent the minimum and maximum values ​​of the driving motor torque.

[0168] In some embodiments, the contact force corresponding to the foot of the robot's mechanical leg should satisfy the friction cone constraint. In order to reduce nonlinearity, the friction cone can be approximated as a friction angle cone. Then the friction constraint expression can be:

[0169] Among them, n x 、n y and n z They represent the unit orthogonal basis along the contact surface in the world coordinate system, μ i represents the friction coefficient corresponding to the i-th contact force, f i Expression i contact force, f z,lb and f z,ub They represent the minimum and maximum values ​​of the non-negative positive pressure perpendicular to the contact surface, respectively.

[0170] 3. Under the constraints of the robot's joint physical constraint expressions and friction constraint expressions, the intermediate dynamic equations are solved to obtain the desired joint torque set.

[0171] In some embodiments, the intermediate kinetic equation can be rewritten in the form of AX=B;

[0172] in,

[0173] The essence of the solution process of AX=B is to find the solution of the linear equation system. Here, the quadratic programming optimization method can be used to construct the objective function of the intermediate dynamic equation.

[0174] In some embodiments, the objective function of the intermediate kinetic equation can be expressed as follows:

[0175] J = min(AX-B) T QAX-B)+X T RX;

[0176] Among them, Q and R represent weight matrices.

[0177] Under the constraints of the robot's joint physical and frictional constraint expressions, the desired joint torque set is obtained by minimizing the objective function. For example, using a quadratic programming optimizer, under the constraints of the robot's joint physical and frictional constraint expressions, minimizing the objective function is used as the optimization objective. The unknown variable X can be obtained, and the active joint torque τ in the unknown variable X can be directly determined as the desired joint torque set. This method refines the method for determining the desired joint torque set and improves the efficiency of obtaining the desired joint torque set.

[0178] Taking the quadruped wheel-legged robot in Figure 2 as an example, the expected joint torque set can include the expected joint torques corresponding to two hip joints, the linear telescopic joints corresponding to the four mechanical legs, the rotation joints of the four wheels, and one pitch rotation joint (pitch + yaw).

[0179] In some embodiments, if the robot model structure is relatively simple, the robot dynamic equation is also relatively simple, and the intermediate dynamic equation can be solved directly by using the matrix inversion method, that is, X = A -1 B, to obtain the desired joint torque set.

[0180] Step 603 : According to the expected joint torque set, the robot is controlled to move under the guidance of the expected operation space task.

[0181] The expected joint torque set includes the expected joint torques of each active joint of the robot at each control moment. At any control moment, it is only necessary to drive the drive motors corresponding to each active joint according to the expected joint torque of each active joint at that control moment to achieve the following of the expected operation control task. That is, the robot can move under the guidance of the expected operation space task and follow the various reference movement trajectories corresponding to the expected operation space task to achieve robot control.

[0182] In summary, the technical solution provided by the embodiments of the present application improves the robot's mobility by allowing it to maintain a standing position with one set of mechanical legs and rapidly move in a dynamic equilibrium state (i.e., the robot's center of gravity can exceed the robot's support area) by swinging the other set of mechanical legs while maintaining a standing position. Furthermore, controlling the robot's movement by alternating the swinging of the first and second mechanical leg groups can further improve the robot's mobility.

[0183] The robot's center of gravity refers to the point where the robot's gravity is concentrated, and the robot's center of mass is the weighted average of the mass position with respect to its mass. If the gravity is uniform, the center of mass and center of gravity can coincide. The robot's support area refers to the area enclosed by the contact points between the feet of each of the robot's mechanical legs and the support surface. Related technologies require that the projection of the center of gravity always be controlled within the support area, resulting in a very low center of mass speed and slow overall movement. However, the robot in the embodiment of the present application can have its center of gravity exceed the robot's support area, making the overall movement very fast, thereby improving the robot's mobility.

[0184] In addition, by adopting the robot's whole-body dynamics model, the expected joint torque set corresponding to the expected operation space task can be accurately obtained according to the expected operation space task, and then the expected operation space task can be accurately achieved according to the expected joint torque set, thereby improving the control accuracy of the robot.

[0185] In addition, under the constraints of the robot's joint physical constraint expressions and friction constraint expressions, solving the intermediate dynamic equations can obtain a reasonable and accurate set of expected joint torques, thereby further improving the robot's control accuracy.

[0186] Furthermore, using an inverted pendulum model to calculate the third sub-desired acceleration of the center of mass in the forward direction allows the robot's center of mass to continuously move in the forward direction while also helping the robot maintain dynamic balance, enabling stable and rapid movement. This further improves the robot's control stability and movement efficiency. Simultaneously, using a PD feedback controller to calculate the third sub-desired acceleration of the center of mass in the second direction, as well as the fourth desired acceleration, makes the robot more stable when controlled using the final desired acceleration.

[0187] In some embodiments, referring to Figures 10-12, the technical solutions provided by the embodiments of the present application are described by taking a quadruped wheel-legged robot climbing stairs as an example, which may include the following contents:

[0188] The quadruped wheel-legged robot 1000 climbs a step during each step cycle. Based on the size information of each step and the radius of the wheel, the reference position of the quadruped wheel-legged robot 1000 on each step (including the coordinates in the x-axis and z-axis directions) can be planned. Based on the reference position, the support movement reference trajectory of the support mechanical leg group, the swing movement reference trajectory of the swing mechanical leg group, and the robot's center of mass reference movement trajectory are planned. Since the quadruped wheel-legged robot 1000 always remains upright during movement, a reference trajectory for the posture change of the quadruped wheel-legged robot 1000 can be planned.

[0189] The quadruped wheeled-legged robot 1000 adopts the technical solution provided by the above embodiment. According to the swing reference movement trajectory, the first expected acceleration of the swing mechanical leg group at each control moment is obtained. According to the support reference movement trajectory, the second expected acceleration of the support mechanical leg group at each control moment is obtained. According to the center of mass reference movement trajectory, the third expected acceleration of the center of mass at each control moment is obtained. According to the posture reference change trajectory, the fourth expected acceleration of the fuselage at each control moment is obtained. Then, based on the first expected acceleration, second expected acceleration, third expected acceleration and fourth expected acceleration at each control moment, the expected operation space task of the quadruped wheeled-legged robot 1000 is obtained.

[0190] The quadruped wheeled-legged robot 1000 continues to adopt the technical solution provided by the above embodiment to obtain the expected joint torque set corresponding to the expected operation space task. The expected joint torque set includes the expected joint torques of the two hip joints at each control moment, the expected joint torques of the linear telescopic joints corresponding to the four mechanical legs, the expected joint torques of the rotation joints of the four wheels, and the expected joint torque of one pitch rotation joint.

[0191] The quadruped wheeled legged robot 1000 is controlled to climb stairs according to the desired joint torque set. The specific contents may be as follows:

[0192] The initial state of the quadruped wheeled legged robot 1000 when climbing stairs is an overlapping standing state, that is, in the initial state, the four mechanical legs of the quadruped wheeled legged robot 1000 are synchronously close together in the forward direction (e.g., horizontally to the right). The embodiments of the present application do not limit the initial state of the quadruped wheeled legged robot 1000; for example, it can be a four-legged standing state or a two-legged standing state.

[0193] 1. In the first step cycle (i.e., FIG10 ), the quadruped wheeled leg robot 1000 uses the outer mechanical leg group 1001 (i.e., the first mechanical leg group) as support and swings the inner mechanical leg group 1002 (i.e., the second mechanical leg group) in the forward direction, so that the quadruped wheeled leg robot 1000 climbs the first step of the stairs.

[0194] Among them, for the two inner mechanical legs in the inner mechanical leg group 1002, the inner hip joint is controlled to rotate according to the desired joint torque of its corresponding hip joint (hereinafter referred to as the inner hip joint), so that the two inner mechanical legs swing synchronously in the forward direction, and at the same time, the inner linear telescopic joint is controlled to contract according to the desired joint torque of its corresponding linear telescopic joint (hereinafter referred to as the inner linear telescopic joint), and the leg length of the two inner mechanical legs is shortened synchronously, so that the two inner mechanical legs are higher than the first step and do not contact the first step.

[0195] Then, the desired joint torque of the inner hip joint is continued to control the rotation of the inner hip joint, so that the two inner mechanical legs continue to swing synchronously in the forward direction. At the same time, the desired joint torque of the inner linear telescopic joint is used to control the extension of the inner linear telescopic joint, and the leg length of the two inner mechanical legs is synchronously extended, so that the inner mechanical leg group 1002 reaches the reference position of the first step.

[0196] During this process, the quadruped wheeled-legged robot 1000 controls the adaptive and synchronous extension of the two outer mechanical legs in the outer mechanical leg group 1001 through the desired joint torque of the outer linear extension joint, and keeps the two outer mechanical legs standing through the desired joint torque of the outer hip joint. The quadruped wheeled-legged robot 1000 controls the body to remain upright through the desired joint torque of the pitch rotation joint.

[0197] 2. In the second step cycle (i.e., FIG11 ), the quadruped wheeled leg robot 1000 uses the inner mechanical leg group 1002 as support and swings the outer mechanical leg group 1001 in the forward direction, so that the quadruped wheeled leg robot 1000 climbs the second step of the stairs.

[0198] Among them, for the two outer mechanical legs in the outer mechanical leg group 1001, the rotation of the outer hip joint is controlled by the desired joint torque of the outer hip joint, so that the two outer mechanical legs swing synchronously in the forward direction. At the same time, the contraction of the outer linear telescopic joint is controlled by the desired joint torque of the outer linear telescopic joint, and the leg length of the two outer mechanical legs is controlled to be shortened synchronously, so that the two outer mechanical legs are higher than the first step and do not contact the first step.

[0199] Then, the desired joint torque of the outer hip joint is continued to be used to control the rotation of the outer hip joint, so that the two outer mechanical legs continue to swing synchronously in the forward direction. At the same time, the desired joint torque of the outer linear telescopic joint is continued to be used to control the contraction of the outer linear telescopic joint, and the leg length of the two outer mechanical legs is controlled to be shortened synchronously, so that both outer mechanical legs are higher than the second step and do not contact the second step.

[0200] Finally, the desired joint torque of the outer hip joint is continued to control the rotation of the outer hip joint, so that the two outer mechanical legs continue to swing synchronously in the forward direction. At the same time, the desired joint torque of the outer linear telescopic joint is used to control the extension of the outer linear telescopic joint, and the leg length of the two outer mechanical legs is synchronously extended, so that the outer mechanical leg group 1001 reaches the reference position of the second step.

[0201] During this process, the quadrupedal wheeled-legged robot 1000 controls the adaptive and synchronous extension of the two inner mechanical legs in the inner mechanical leg group 1002 through the desired joint torque of the inner linear telescopic joint, and maintains the two inner mechanical legs in an upright position through the desired joint torque of the inner hip joint. The quadrupedal wheeled-legged robot 1000 controls the body to maintain an upright position through the desired joint torque of the pitch rotation joint. The body of the quadrupedal wheeled-legged robot 1000 follows the extension of the two inner mechanical legs, moving simultaneously in the vertical and forward directions to achieve adjustment of the center of mass of the quadrupedal wheeled-legged robot 1000.

[0202] 3. In the third step cycle (i.e., FIG12 ), the quadruped wheeled leg robot 1000 uses the outer mechanical leg group 1001 as support and swings the inner mechanical leg group 1002 in the forward direction, so that the quadruped wheeled leg robot 1000 climbs the third step of the stairs.

[0203] Among them, for the two inner mechanical legs in the inner mechanical leg group 1002, the rotation of the inner hip joint is controlled by the desired joint torque of the inner hip joint, so that the two inner mechanical legs swing synchronously in the forward direction. At the same time, the contraction of the inner linear telescopic joint is controlled by the desired joint torque of the inner linear telescopic joint, and the leg length of the two inner mechanical legs is controlled to be shortened synchronously, so that both inner mechanical legs are higher than the second step and do not contact the second step.

[0204] Then, the desired joint torque of the inner hip joint is continued to be used to control the rotation of the inner hip joint, so that the two inner mechanical legs continue to swing synchronously in the forward direction. At the same time, the desired joint torque of the inner linear telescopic joint is continued to be used to control the contraction of the inner linear telescopic joint, and the leg lengths of the two inner mechanical legs are controlled to be shortened synchronously, so that both inner mechanical legs are higher than the third step and do not contact the third step.

[0205] Finally, the desired joint torque of the inner hip joint is continued to control the rotation of the inner hip joint, so that the two inner mechanical legs continue to swing synchronously in the forward direction. At the same time, the desired joint torque of the inner linear telescopic joint is used to control the extension of the inner linear telescopic joint, and the leg length of the two inner mechanical legs is synchronously extended, so that the inner mechanical leg group 1002 reaches the reference position of the third step.

[0206] During this process, the quadrupedal wheeled-legged robot 1000 controls the adaptive and synchronous extension of the two outer mechanical legs in the outer mechanical leg group 1001 through the desired joint torque of the outer linear telescopic joint, and maintains the two outer mechanical legs in an upright position through the desired joint torque of the outer hip joint. The quadrupedal wheeled-legged robot 1000 controls the body to maintain an upright position through the desired joint torque of the pitch rotation joint. The body of the quadrupedal wheeled-legged robot 1000 follows the extension of the two outer mechanical legs, moving simultaneously in the vertical and forward directions to achieve adjustment of the center of mass of the quadrupedal wheeled-legged robot 1000.

[0207] In the subsequent stepping cycle, the inner mechanical leg group 1002 and the outer mechanical leg group 1001 swing alternately, so that the quadruped wheeled-legged robot 1000 climbs the stairs.

[0208] In summary, the technical solution provided by the embodiments of the present application allows a quadruped wheeled-legged robot with an inner and outer mechanical leg groups to maintain a standing position by using one set of mechanical legs while swinging the other set of mechanical legs, thereby improving the robot's mobility. Furthermore, controlling the robot's stair climbing by alternating the swinging of the inner and outer mechanical leg groups can further improve the robot's stair climbing efficiency.

[0209] In addition, by adopting the robot's whole-body dynamics model, the expected joint torque set corresponding to the expected operation space task can be accurately obtained according to the expected operation space task, and then the expected operation space task can be accurately achieved according to the expected joint torque set, thereby improving the control accuracy of the quadruped wheeled-legged robot.

[0210] The following are device embodiments of the present application, which can be used to implement the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0211] Referring to Figure 13, a block diagram of a robot control device provided by one embodiment of the present application is shown. The device has the function of implementing the control method of the above-mentioned robot, and the function can be implemented by hardware or by hardware executing corresponding software. The device can be a computer device introduced above (such as a wheel-legged robot), or it can be placed in a computer device. As shown in Figure 13, the device 1300 includes: a desired task acquisition module 1301, a desired torque acquisition module 1302 and a robot movement module 1303.

[0212] The expected task acquisition module 1301 is used to obtain the expected operating space task of the robot on the support surface. The expected operating space task includes the expected acceleration of each part of the robot in the operating space of the robot. The expected operating space task is used to guide the robot to alternately swing the first mechanical leg group and the second mechanical leg group to move on the support surface.

[0213] The expected torque acquisition module 1302 is used to obtain the expected joint torque set corresponding to the expected operation space task based on the expected operation space task and the whole-body dynamics model of the robot. The expected joint torque set includes the expected joint torques for controlling various parts of the robot.

[0214] The robot movement module 1303 is used to control the robot to move under the guidance of the expected operation space task according to the expected joint torque set.

[0215] In some embodiments, as shown in FIG14 , the desired torque acquisition module 1302 includes: a forward model acquisition submodule 1302 a , a dynamic model simultaneous submodule 1302 b , and a desired torque acquisition submodule 1302 c .

[0216] The forward model acquisition submodule 1302a is used to obtain the forward dynamics model of the robot, where the forward dynamics model is used to indicate the relationship between the acceleration of the robot in the operation space and the velocity and acceleration of the robot in the joint space of the robot.

[0217] The dynamic model simultaneous submodule 1302b is used to simultaneously establish the whole-body dynamic model and the forward dynamic model to obtain a dynamic equation to be solved, wherein the dynamic equation to be solved uses the acceleration of the robot in the joint space as an unknown variable.

[0218] The expected torque acquisition submodule 1302c is used to bring the expected operation space task into the dynamic equation to be solved to obtain the expected joint torque set.

[0219] In some embodiments, the expected torque acquisition submodule 1302c is configured to:

[0220] Substituting the desired operation space task with the acceleration of the robot in the operation space in the dynamic equation to be solved to obtain an intermediate dynamic equation;

[0221] Constructing joint physical constraint expressions and friction constraint expressions of the robot, wherein the joint physical constraint expressions are used to constrain the joints of the robot, and the contact force between the robot and the support surface under the constraints of the friction constraint expressions satisfies a friction cone constraint;

[0222] Under the constraints of the joint physical constraint expressions and friction constraint expressions of the robot, the intermediate dynamic equations are solved to obtain the desired joint torque set.

[0223] In some embodiments, the desired torque acquisition submodule 1302c is further configured to:

[0224] Adopting a quadratic programming optimization method to construct an objective function of the intermediate dynamics equation;

[0225] Under the constraints of the joint physical constraint expressions and the friction constraint expressions of the robot, the desired joint torque set is obtained with minimization of the objective function as the optimization goal.

[0226] In some embodiments, during the movement of the robot, the mechanical leg group used for swinging is a swinging mechanical leg group, and the mechanical leg group used for supporting is a supporting mechanical leg group; as shown in Figure 14, the expected task acquisition module 1301 includes: a first acceleration acquisition submodule 1301a, a second acceleration acquisition submodule 1301b, a third acceleration acquisition submodule 1301c, a fourth acceleration acquisition submodule 1301d and an expected task acquisition submodule 1301e.

[0227] The first acceleration acquisition submodule 1301a is used to obtain the first expected acceleration of the swinging mechanical leg group in the operating space according to the swinging reference movement trajectory corresponding to the swinging mechanical leg group, and the swinging reference movement trajectory is obtained by planning the movement trajectory of the swinging mechanical leg group according to the support surface.

[0228] The second acceleration acquisition submodule 1301b is used to obtain the second expected acceleration of the supporting mechanical leg group in the operating space according to the supporting reference movement trajectory corresponding to the supporting mechanical leg group, and the supporting reference movement trajectory is obtained by planning the movement trajectory of the supporting mechanical leg group according to the supporting surface.

[0229] The third acceleration acquisition submodule 1301c is used to obtain the third expected acceleration of the center of mass in the operating space according to the center of mass reference movement trajectory corresponding to the center of mass of the robot, and the center of mass reference movement trajectory is planned according to the movement trajectory of the support surface to the center of mass.

[0230] The fourth acceleration acquisition submodule 1301d is used to obtain a fourth expected acceleration of the robot body in the operating space according to a reference attitude change trajectory of the robot body, where the reference attitude change trajectory is obtained by planning the change trajectory of the body.

[0231] The expected task acquisition submodule 1301e is configured to acquire the expected operation space task according to the first expected acceleration, the second expected acceleration, the third expected acceleration, and the fourth expected acceleration.

[0232] In some embodiments, the robot stops moving after a plurality of control moments; the first acceleration acquisition submodule 1301a is configured to:

[0233] For each of the control moments, obtaining a reference position, a reference velocity, and a reference acceleration of the swinging mechanical leg group at the control moment according to the swinging reference movement trajectory;

[0234] A proportional differential PD feedback controller is used to calculate the first expected acceleration at the control moment based on the reference position, reference speed and reference acceleration of the swinging mechanical leg group at the control moment, as well as the actual position and actual speed of the swinging mechanical leg group at the control moment.

[0235] In some embodiments, during the supporting process of the supporting mechanical leg, there is no relative sliding between the foot of the supporting mechanical leg and the supporting surface, and the value of the second expected acceleration is always zero.

[0236] In some embodiments, the robot moves in a first direction; the third acceleration acquisition submodule 1301c is configured to:

[0237] acquiring a third sub-expected acceleration of the center of mass in the first direction according to a sub-reference movement trajectory of the center of mass reference movement trajectory in the first direction;

[0238] Obtaining a third sub-expected acceleration of the center of mass in a second direction according to a sub-reference movement trajectory of the center of mass reference movement trajectory in a second direction, where the second direction is perpendicular to the first direction;

[0239] The third expected acceleration is obtained according to the third sub-expected acceleration of the center of mass in the first direction and the third sub-expected acceleration of the center of mass in the second direction.

[0240] In some embodiments, the robot stops moving after a plurality of control moments; the third acceleration acquisition submodule 1301c is further configured to:

[0241] The position of the center of mass, the velocity of the center of mass, the distance between the center of mass and the support contact point in the first direction, and the derivative of the distance are used as state variables, and the acceleration of the center of mass relative to the support contact point in the first direction is used as a control variable to construct the inverted pendulum dynamic equation of the robot, where the support contact point refers to the contact point between the foot corresponding to the supporting mechanical leg group and the support surface;

[0242] Calculating the feedback gain matrix of the inverted pendulum dynamics equation using a linear quadratic programming regulator;

[0243] For each of the control moments, according to the sub-reference movement trajectory in the first direction, obtaining a reference position, a reference speed, a reference distance, and a reference speed between the center of mass and a support contact point in the first direction at the control moment;

[0244] According to the feedback gain matrix, the reference position, reference speed, reference distance and reference speed between the center of mass and the support contact point at the control moment and the first direction, the third sub-expected acceleration at the control moment and the first direction is obtained.

[0245] In some embodiments, the robot stops moving after a plurality of control moments; the third acceleration acquisition submodule 1301c is further configured to:

[0246] For each of the control moments, obtaining, according to the sub-reference movement trajectory in the second direction, a reference position, a reference velocity, and a reference acceleration of the center of mass in the second direction at the control moment;

[0247] A proportional differential PD feedback controller is used to calculate the third sub-expected acceleration at the control moment and in the second direction based on the reference position, reference velocity and reference acceleration of the center of mass at the control moment and in the second direction, as well as the actual position and actual velocity of the center of mass at the control moment and in the second direction.

[0248] In some embodiments, the robot stops moving after a plurality of control moments; the fourth acceleration acquisition submodule 1301d is configured to:

[0249] For each of the control moments, obtaining a reference attitude angle, a reference attitude angular velocity, and a reference attitude angular acceleration of the fuselage at the control moment according to the attitude reference change trajectory;

[0250] A proportional differential (PD) feedback controller is used to calculate the fourth desired acceleration at the control moment based on the reference attitude angle, reference attitude angular velocity, and reference attitude angular acceleration of the fuselage at the control moment, as well as the actual attitude angle and actual attitude angular velocity of the fuselage at the control moment.

[0251] In some embodiments, the body of the robot remains vertical during the movement of the robot, and the values ​​of the reference posture angle, reference posture angular velocity, and reference posture angular acceleration corresponding to the body are all zero.

[0252] In some embodiments, the hip joints of the robot are coaxial.

[0253] In some embodiments, the robotic legs in the first robotic leg group move synchronously, and the robotic legs in the second robotic leg group move synchronously.

[0254] In summary, the technical solution provided by the embodiments of the present application improves the robot's mobility by allowing it to maintain a standing position with one set of mechanical legs and rapidly move in a dynamic equilibrium state (i.e., the robot's center of gravity can exceed the robot's support area) by swinging the other set of mechanical legs while maintaining a standing position. Furthermore, controlling the robot's movement by alternating the swinging of the first and second mechanical leg groups can further improve the robot's mobility.

[0255] In addition, by adopting the robot's whole-body dynamics model, the expected joint torque set corresponding to the expected operation space task can be accurately obtained according to the expected operation space task, and then the expected operation space task can be accurately achieved according to the expected joint torque set, thereby improving the control accuracy of the robot.

[0256] It should be noted that the apparatus provided in the above embodiments, when implementing its functions, is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0257] Please refer to Figure 15, which shows a simplified block diagram of a computer device provided in one embodiment of the present application. The computer device may refer to any electronic device with data computing, processing, and storage capabilities.

[0258] In some embodiments, as shown in FIG15 , the computer device includes a processor 1501 and a memory 1502. The processor 1501 includes, but is not limited to, any one of the following: a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and an FPGA (Field Programmable Gate Array). The memory 1502 may include storage devices such as RAM (Random-Access Memory) and ROM (Read-Only Memory). The processor 1501 and the memory 1502 may be connected via a system bus.

[0259] In some embodiments, a computer program is stored in the memory 1502 , and the computer program is loaded and executed by the processor 1501 to implement the above-mentioned robot control method.

[0260] In some embodiments, a computer-readable storage medium is further provided, wherein the storage medium stores a computer program, and the computer program implements the above-mentioned robot control method when executed by a processor of a computer device.

[0261] In some embodiments, the computer-readable storage medium may include: ROM, RAM, SSD (Solid State Drives) or optical disks, etc. Among them, random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0262] In some embodiments, a computer program product is further provided, comprising a computer program stored in a computer-readable storage medium. A processor of a computer device reads the computer program from the computer-readable storage medium and executes the computer program, causing the computer device to perform the above-described robot control method.

[0263] It should be noted that, in the embodiment of the present application, before collecting the relevant data of the user and during the process of collecting the relevant data of the user, a prompt interface, pop-up window or voice prompt information can be displayed. The prompt interface, pop-up window or voice prompt information is used to remind the user that its relevant data is currently being collected, so that the present application only starts to execute the relevant steps of obtaining the relevant data of the user after obtaining the confirmation operation issued by the user on the prompt interface or pop-up window. Otherwise (that is, when the confirmation operation issued by the user on the prompt interface or pop-up window is not obtained), the relevant steps of obtaining the relevant data of the user are terminated, that is, the relevant data of the user is not obtained. In other words, all user data collected by this application are processed strictly in accordance with the requirements of relevant national laws and regulations. The informed consent or separate consent of the personal information subject is obtained only when the user agrees and authorizes the collection, and the subsequent data use and processing behavior is carried out within the scope of the authorization of the laws and regulations and the personal information subject. The collection, use and processing of relevant user data need to comply with the relevant laws, regulations and standards of the relevant countries and regions. For example, the real environment involved in this application is obtained with full authorization.

[0264] It should be understood that the "multiple" mentioned in this article refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent three situations: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the previous and subsequent associated objects are in an "or" relationship. In addition, the step numbers described in this article only illustrate a possible execution sequence between the steps. In some other embodiments, the above steps may not be executed in the order of the numbers, such as two steps with different numbers are executed at the same time, or two steps with different numbers are executed in the opposite order to the diagram. The embodiments of the present application do not limit this.

[0265] The above description is merely an exemplary embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A control method for a robot, the method being executed by a computer device, the robot comprising a body, a first mechanical leg group and a second mechanical leg group connected to the body via a hip joint, at least one of the first mechanical leg group and the second mechanical leg group comprising at least two mechanical legs, the rotation center of the hip joint corresponding to the first mechanical leg group and the rotation center of the hip joint corresponding to the second mechanical leg group being located in the same vertical plane; the method include: Acquire an expected operation space task of the robot on a support surface, wherein the expected operation space task includes expected accelerations of various parts of the robot in the operation space of the robot, and the expected operation space task is used to guide the robot to alternately swing the first mechanical leg group and the second mechanical leg group to move on the support surface; According to the expected operation space task and the whole body dynamics model of the robot, an expected joint torque set corresponding to the expected operation space task is acquired, wherein the expected joint torque set includes expected joint torques for controlling various parts of the robot; According to the expected joint torque set, the robot is controlled to move under the guidance of the expected operation space task.

2. The method according to claim 1, in, The step of obtaining an expected joint torque set corresponding to the expected operation space task according to the expected operation space task and the whole body dynamics model of the robot includes: Acquire a forward dynamics model of the robot, wherein the forward dynamics model is used to indicate a relationship between an acceleration of the robot in the operation space and a velocity and acceleration of the robot in a joint space of the robot; The whole-body dynamics model and the forward dynamics model are combined to obtain a dynamics equation to be solved, wherein the dynamics equation to be solved uses the acceleration of the robot in the joint space as an unknown variable; The desired operation space task is introduced into the dynamic equation to be solved, and the desired joint torque set is obtained by solving the equation.

3. The method according to claim 2, in, The step of bringing the expected operation space task into the dynamic equation to be solved and solving the expected joint torque set includes: Substituting the desired operation space task with the acceleration of the robot in the operation space in the dynamic equation to be solved to obtain an intermediate dynamic equation; Constructing a joint physical constraint expression and a friction constraint expression of the robot, wherein the joint physical constraint expression is used to constrain each joint of the robot, and the contact force between the robot and the support surface constrained by the friction constraint expression satisfies a friction cone constraint; Under the constraints of the joint physical constraint expressions and friction constraint expressions of the robot, the intermediate dynamics equations are solved to obtain the desired joint torque set.

4. The method according to claim 3, in, The intermediate dynamic equation is solved under the constraints of the joint physical constraint expression and the friction constraint expression of the robot to obtain the desired joint torque set, including: Using a quadratic programming optimization method to construct an objective function of the intermediate dynamics equation; Under the constraints of the joint physical constraint expressions and the friction constraint expressions of the robot, the desired joint torque set is obtained by taking minimization of the objective function as the optimization goal.

5. The method according to any one of claims 1 to 4, in, During the movement of the robot, the mechanical leg group used for swinging is the swinging mechanical leg group, and the mechanical leg group used for supporting is the supporting mechanical leg group; The obtaining of the expected operation space task of the robot on the support surface comprises: According to the swing reference movement trajectory corresponding to the swing mechanical leg group, a first expected acceleration of the swing mechanical leg group in the operation space is obtained, wherein the swing reference movement trajectory is obtained by planning the movement trajectory of the swing mechanical leg group according to the support surface; According to the support reference movement trajectory corresponding to the support mechanical leg group, a second expected acceleration of the support mechanical leg group in the operation space is obtained, wherein the support reference movement trajectory is obtained by planning the movement trajectory of the support mechanical leg group according to the support surface; According to the center of mass reference movement trajectory corresponding to the center of mass of the robot, the first 3. Expected acceleration, the reference moving trajectory of the center of mass is planned according to the moving trajectory of the support surface to the center of mass; acquiring a fourth expected acceleration of the body in the operation space according to a reference change trajectory of the posture of the body of the robot, wherein the reference change trajectory of the posture is obtained by planning the change trajectory of the body; The expected operation space task is acquired according to the first expected acceleration, the second expected acceleration, the third expected acceleration and the fourth expected acceleration.

6. The method according to claim 5, in, The robot stops moving after a plurality of control moments; The step of obtaining a first expected acceleration of the swinging mechanical leg group in the operating space according to the swinging reference movement trajectory corresponding to the swinging mechanical leg group includes: For each of the control moments, according to the swing reference movement trajectory, a reference position, a reference speed and a reference acceleration of the swing mechanical leg group at the control moment are obtained; A proportional differential PD feedback controller is used to calculate the first expected acceleration at the control moment according to the reference position, reference speed and reference acceleration of the swinging mechanical leg group at the control moment, as well as the actual position and actual speed of the swinging mechanical leg group at the control moment.

7. The method according to claim 5 or 6, in, During the supporting process of the supporting mechanical leg, there is no relative sliding between the foot of the supporting mechanical leg and the supporting surface, and the value of the second expected acceleration is always zero.

8. The method according to any one of claims 5 to 7, in, The robot moves in a first direction; The obtaining, according to the center of mass reference movement trajectory corresponding to the center of mass of the robot, a third expected acceleration of the center of mass in the operation space includes: Acquire a third sub-expected acceleration of the center of mass in the first direction according to a sub-reference moving trajectory of the center of mass reference moving trajectory in the first direction; acquiring, according to a sub-reference moving trajectory of the reference moving trajectory of the center of mass in a second direction, a third sub-expected acceleration of the center of mass in the second direction, wherein the second direction is perpendicular to the first direction; The third expected acceleration is obtained according to the third sub-expected acceleration of the center of mass in the first direction and the third sub-expected acceleration of the center of mass in the second direction.

9. The method according to claim 8, in, The robot stops moving after a plurality of control moments; The acquiring, according to the sub-reference moving trajectory of the center of mass reference moving trajectory in the first direction, a third sub-expected acceleration of the center of mass in the first direction comprises: The position of the center of mass, the speed of the center of mass, the distance between the center of mass and the support contact point in the first direction, and the derivative of the distance are used as state variables, and the acceleration of the center of mass relative to the support contact point in the first direction is used as a control variable to construct the inverted pendulum dynamics equation of the robot, wherein the support contact point refers to the contact point between the foot corresponding to the supporting mechanical leg group and the support surface; Utilizing a linear quadratic programming regulator, calculating a feedback gain matrix of the inverted pendulum dynamics equation; For each of the control moments, according to the sub-reference movement trajectory in the first direction, obtain a reference position, a reference speed, a reference distance and a reference speed between the center of mass and a support contact point in the first direction at the control moment; According to the feedback gain matrix, the reference position, reference speed, reference distance and reference speed between the center of mass and the support contact point at the control moment and the first direction, the third sub-expected acceleration at the control moment and the first direction is obtained.

10. The method according to claim 8 or 9, in, The robot stops moving after a plurality of control moments; The acquiring, according to the sub-reference moving trajectory of the center of mass reference moving trajectory in the second direction, a third sub-expected acceleration of the center of mass in the second direction comprises: For each of the control moments, according to the sub-reference movement trajectory in the second direction, obtaining a reference position, a reference velocity, and a reference acceleration of the center of mass in the second direction at the control moment; A proportional differential PD feedback controller is used, according to the reference of the center of mass at the control time and the second direction The third sub-expected acceleration at the control moment and the second direction is calculated based on the position, reference speed and reference acceleration, as well as the actual position and actual speed of the center of mass at the control moment and the second direction.

11. The method according to any one of claims 5 to 10, in, The robot stops moving after a plurality of control moments; The acquiring, according to the reference change trajectory of the posture of the robot body, a fourth expected acceleration of the body in the operation space includes: For each of the control moments, according to the reference attitude change trajectory, obtaining a reference attitude angle, a reference attitude angular velocity, and a reference attitude angular acceleration of the fuselage at the control moment; A proportional differential PD feedback controller is adopted to calculate the fourth expected acceleration at the control moment according to the reference attitude angle, reference attitude angular velocity and reference attitude angular acceleration of the fuselage at the control moment, and the actual attitude angle and actual attitude angular velocity of the fuselage at the control moment.

12. The method according to claim 11, in, The body of the robot remains vertical during the movement of the robot, and the value of the reference posture angle, the value of the reference posture angular velocity and the value of the reference posture angular acceleration corresponding to the body are all zero.

13. The method according to any one of claims 1 to 12, in, The hip joints of the robot are coaxial.

14. The method according to any one of claims 1 to 13, in, The mechanical legs in the first mechanical leg group move synchronously, and the mechanical legs in the second mechanical leg group move synchronously.

15. A control device for a robot, the robot comprising a body, a first mechanical leg group and a second mechanical leg group connected to the body via hip joints, at least one of the first mechanical leg group and the second mechanical leg group comprises at least two mechanical legs, and the rotation center of the hip joint corresponding to the first mechanical leg group and the rotation center of the hip joint corresponding to the second mechanical leg group are located in the same vertical plane; the device include: an expected task acquisition module, used to acquire an expected operation space task of the robot on a support surface, wherein the expected operation space task includes expected accelerations of various parts of the robot in the operation space of the robot, and the expected operation space task is used to guide the robot to alternately swing the first mechanical leg group and the second mechanical leg group to move on the support surface; An expected torque acquisition module, used for acquiring an expected joint torque set corresponding to the expected operation space task according to the expected operation space task and the whole body dynamics model of the robot, wherein the expected joint torque set includes expected joint torques for controlling various parts of the robot; The robot movement module is used to control the robot to move under the guidance of the expected operation space task according to the expected joint torque set.

16. A computer device, comprising a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the robot control method according to any one of claims 1 to 14.

17. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program is loaded and executed by a processor to implement the robot control method according to any one of claims 1 to 14.

18. A computer program product, comprising a computer program, wherein the computer program is stored in a computer-readable storage medium, and a processor reads and executes the computer program from the computer-readable storage medium to implement the robot control method according to any one of claims 1 to 14.

Citation Information

Patent Citations

  • Compliant control method of wheel foot type robot

    CN104108433A

  • Quadruped robot pacing gait planning method, device and equipment and readable medium

    CN111208826A

  • Robot balance control method and device, robot control equipment and storage medium

    CN113290559A

  • Control method, device and equipment of wheel-legged robot and readable storage medium

    CN113753150A

  • Four-leg mobile robot

    JP2005046950A

Cited By

  • Attitude control method and system for tandem type four-foot wheel-leg robot

    CN119328772A

  • Tracking control method and system for incomplete four-wheel mobile robot

    CN119668272A

  • Surgical robot arm joint torque self-adaptive compensation and dynamic balance control method

    CN119927934A