Robot motion control system and robot

Abstract

The invention provides a robot motion control system and a robot, the robot is a wheel-foot composite robot, the robot comprises a robot body and a motion mechanism arranged on the robot body, and the motion mechanism comprises two rear rollers and two front leg mechanisms; the system comprises a motion planning module used for determining a first motion task, a second motion task and a third motion task according to a target task and latest actual state information of the wheel-foot composite robot; the motion control module is used for controlling the whole-body motion of the wheel-foot composite robot according to the first motion task, the second motion task and the third motion task; the state sensing module is used for determining actual state information of the wheel-foot composite robot and sending the actual state information of the wheel-foot composite robot to the motion planning module so that the motion planning module can conduct new motion planning according to the actual state information of the wheel-foot composite robot. Accurate closed-loop control can be carried out on the movement of the robot, and the task execution effect of the robot is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of robotics technology, and in particular to a robot motion control system and a robot. Background Art

[0002] With the development of robotics technology, robots are increasingly being used in a variety of fields, such as industrial and agricultural production, social services, and domestic services. They can assist or even replace humans in completing arduous and complex tasks, thereby improving work efficiency and quality. However, in actual applications, robots can suffer from poor performance due to inaccurate motion control or errors. Therefore, precisely controlling robot motion has become a pressing issue. Summary of the Invention

[0003] The embodiments of the present application provide a robot motion control system and a robot, which can perform precise closed-loop control of the robot's motion to ensure that the robot can perform tasks efficiently, smoothly and accurately, thereby improving the robot's task execution effect and enhancing the user experience.

[0004] In a first aspect, an embodiment of the present application provides a robot motion control system, wherein the robot is a wheel-leg composite robot, the wheel-leg composite robot comprising a robot body and four motion mechanisms disposed on the robot body, the four motion mechanisms comprising two rear rollers and two front leg mechanisms, the system comprising: a motion planning module, a motion control module, and a state perception module;

[0005] The motion planning module is configured to determine, based on the target task and the latest actual state information of the wheel-leg hybrid robot, a first motion task of the robot body, a second motion task of the two front leg mechanisms, and a third motion task of the two rear rollers;

[0006] The motion control module is configured to control the whole-body motion of the wheel-leg composite robot according to the first motion task, the second motion task, and the third motion task sent by the motion planning module;

[0007] The state perception module is used to determine the actual state information of the wheel-leg composite robot and send the actual state information of the wheel-leg composite robot to the motion planning module, so that the motion planning module performs new motion planning according to the actual state information of the wheel-leg composite robot.

[0008] In a second aspect, an embodiment of the present application provides a robot, comprising a robot motion control system as described in the embodiment of the first aspect.

[0009] The technical solution disclosed in the embodiment of the present application determines the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers through the motion planning module according to the target task and the latest actual state information of the wheel-foot composite robot. Then, the motion control module controls the whole-body motion of the wheel-foot composite robot according to the first motion task, the second motion task, and the third motion task output by the motion planning module. Then, the actual state information of the wheel-foot composite robot is determined through the state perception module, and the actual state information of the wheel-foot composite robot is sent to the motion planning module, so that the motion planning module performs new motion planning according to the actual state information of the wheel-foot composite robot. In this way, the robot's motion can be precisely closed-loop controlled to ensure that the robot can perform tasks efficiently, smoothly and accurately, thereby improving the robot's task execution effect and enhancing the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0011] Figure 1 A schematic structural diagram of a wheel-leg composite robot provided in an embodiment of the present application;

[0012] Figure 2 An exemplary block diagram of a robot motion control system provided in an embodiment of the present application;

[0013] Figure 3 A schematic diagram of the center point of a rear roller provided in an embodiment of the present application;

[0014] Figure 4 A schematic block diagram of another robot motion control system provided in an embodiment of the present application;

[0015] Figure 5 A schematic block diagram of another robot motion control system provided in an embodiment of the present application;

[0016] Figure 6 A schematic block diagram of another robot motion control system provided in an embodiment of the present application;

[0017] Figure 7 A schematic block diagram of a robot provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0020] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or solution described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or solutions. Specifically, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0021] In the description of the embodiments of the present application, it should also be noted that, unless otherwise clearly specified and limited, the terms "set", "layout", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to the specific circumstances and in combination with the existing technology. In addition, the features in the embodiments of the present application can be combined with each other unless there is a conflict. And one or more of the components in the diagram may be necessary or non-essential, and the relative positional relationship between the components in the above diagram can be adjusted according to actual needs.

[0022] In the description of the embodiments of the present application, unless otherwise specified, "a plurality of" refers to two or more than two, that is, at least two. "At least one" refers to one or more than one.

[0023] Since the robot may have poor task execution results due to inaccurate motion control or errors when performing tasks, the invention of this application is as follows: a motion planning module determines the first motion task of the robot body, the second motion task of the air leg mechanism in the two front leg mechanisms, and the third motion task of the center points of the two rear rollers according to the target task to be performed by the wheel-foot composite robot and the latest actual state information of the wheel-foot composite robot, and sends the first motion task, the second motion task, and the third motion task to the motion control module, so that the motion control module controls the whole-body motion of the wheel-foot composite robot according to the first motion task, the second motion task, and the third motion task. In addition, the actual state information of the wheel-foot composite robot is determined by the state perception module, and the actual state information of the wheel-foot composite robot is sent to the motion planning module, so that the motion planning module performs new motion planning according to the actual state information of the wheel-foot composite robot. In this way, precise closed-loop control of the robot's motion can be achieved to ensure that the robot can perform tasks efficiently, smoothly, and accurately, thereby improving the robot's task execution effect and enhancing user experience.

[0024] The technical solution of the present application is described in detail below through some embodiments. The embodiments described below can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0025] First, the wheel-foot composite robot structure in the embodiment of the present application is described in detail. Figure 1 As shown, the wheel-leg hybrid robot 10 includes: a robot body 11 and four motion mechanisms arranged on the robot body 11, wherein the four motion mechanisms include two rear rollers 121 and two front leg mechanisms 122.

[0026] In the present application, the two front leg mechanisms 122 include a left front leg mechanism and a right front leg mechanism. Each front leg mechanism 122 includes a front thigh mechanism 1221, a front calf mechanism 1222, and a roller 1223 disposed at the end of each front calf mechanism. The roller 1223 at the end of the front calf mechanism can be understood as a passive wheel. Figure 1 The connecting part between the middle front thigh mechanism 1221 and the robot body 11 is the hip joint 13, the connecting part between the front thigh mechanism 1221 and the front calf mechanism 1222 is the knee joint 14, and the connecting part between the front calf mechanism 1222 and the roller 1223 is the ankle joint 15.

[0027] In some optional embodiments, the two rear rollers 121 are provided at both ends of the rotatable connecting member 16, and the rotatable end of the connecting member 16 is connected to one end of the rear calf mechanism 123, the other end of the rear calf mechanism 123 is connected to one end of the rear thigh mechanism 124, and the other end of the rear thigh mechanism 124 is connected to the robot body 11. For details, see Figure 1 Among them, the two rear rollers 121 can be understood as driving wheels.

[0028] That is, the motion mechanism provided on the robot body 11 in the present application includes a hind leg mechanism, wherein the hind leg mechanism includes a hind thigh mechanism 124, a hind shank mechanism 123, a connecting component 16, and two rear rollers 121 provided at both ends of the connecting component 16. In the present application, the connection between the hind thigh mechanism 124 and the robot body 11 is the hip joint 13, the connection between the hind thigh mechanism 124 and the hind shank mechanism 123 is the knee joint 14, and the connection between the hind shank mechanism 124 and each rear roller 121 is the ankle joint 15. In other words, the hind leg mechanism in the present application includes a hip joint 13, a knee joint 14, and two ankle joints 15.

[0029] In addition, the above-mentioned hip joint 13, knee joint 14 and ankle joint 15 are all correspondingly provided with motors, specifically hip joint motors, knee joint motors and ankle joint motors, so that the corresponding torque is output by the hip joint motor, knee joint motor and / or ankle joint motor to drive the changes of the hip joint 13, knee joint 14 and / or ankle joint 15 to switch the motion mode of the wheel-foot composite robot.

[0030] Each of the aforementioned joints can be understood as indicating that at least two components of the wheel-leg hybrid robot can be flexibly connected. Furthermore, each joint can move under the control of the torque output by the joint motor. For example, rotating a joint by a certain angle can cause other joints and related mechanisms to move a certain amount within space, thereby achieving a change in the robot's motion mode.

[0031] Furthermore, in order to detect different motion environments and whether there are obstacles in the motion environments, the present application can set at least one environmental perception sensor at the front end of the robot body 11 of the wheel-leg composite robot 10, and set at least one environmental perception sensor at the rear end of the robot body 11, so that the wheel-leg composite robot can realize autonomous obstacle avoidance and mapping navigation and other functions based on the environmental perception sensors.

[0032] Among them, environmental perception sensors may include but are not limited to: laser sensors, visual sensors, infrared sensors, ultrasonic sensors and lidar sensors.

[0033] As an optional implementation, you can Figure 1A visual sensor is set in the middle position of the front end of the robot body 11 shown, a laser radar sensor is set on the left and right sides of the middle respectively, and a laser sensor is set in the middle position of the rear end of the robot body 11, and an ultrasonic sensor is set on the left and right sides of the middle respectively (not shown in the figure).

[0034] It should be noted that the number, position and type of environmental perception sensors set on the robot body 11 of this application are only exemplary and can be flexibly adjusted according to actual application needs. This application does not impose any restrictions on this.

[0035] After describing the structure of the wheel-leg hybrid robot, the robot motion control system and the robot provided in the embodiment of the present application are described in detail below. The robot here is the wheel-leg hybrid robot described above.

[0036] Figure 2 This is an exemplary block diagram of a robot motion control system provided in an embodiment of the present application. Figure 2 As shown, the robot motion control system 200 includes: a motion planning module 210 , a motion control module 220 and a state perception module 230 .

[0037] The motion planning module 210 is used to determine the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers based on the target task and the latest actual state information of the wheel-leg hybrid robot;

[0038] The motion control module 220 is used to control the whole body motion of the wheel-leg hybrid robot according to the first motion task, the second motion task and the third motion task sent by the motion planning module;

[0039] The state perception module 230 is used to determine the actual state information of the wheel-leg composite robot and send the actual state information of the wheel-leg composite robot to the motion planning module, so that the motion planning module 210 performs new motion planning according to the actual state information of the wheel-leg composite robot.

[0040] It should be understood that the motion planning module 210 of the present application performs new motion planning based on the actual state information of the wheel-leg composite robot, which means that in the next planning cycle, the motion planning module 210 determines the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers based on the latest actual state information of the wheel-leg composite robot.

[0041] Among them, the planning cycle of the motion planning module 210 can be comprehensively determined based on factors such as the motion planning requirements, hardware computing power, and algorithm complexity of the wheel-foot composite robot. In some optional implementations, if the real-time requirements for the motion planning of the wheel-foot composite robot are high and the hardware computing power is strong, the planning cycle can be set to a smaller value to achieve the effect of real-time planning. For example, the planning cycle can be set to 1 millisecond (ms), that is, a motion planning operation is performed every 1ms. If the real-time requirements for the motion planning of the wheel-foot composite robot are low and the hardware computing power is insufficient, the planning cycle can be set to a larger value. For example, the planning cycle can be set to 5ms, that is, a motion planning operation is performed every 5ms. This application does not impose any specific restrictions on the setting of the planning cycle.

[0042] It should be noted that when the motion planning module 210 performs motion planning according to a planning cycle, if the wheel-leg hybrid robot obtains a new task during a planning cycle, the motion planning module 210 performs motion planning based on the new task and the latest actual state information of the wheel-leg hybrid robot. If the wheel-leg hybrid robot is currently performing a previous task during a planning cycle (i.e., the previous task has not yet been completed and no new task has been obtained), the motion planning module 210 continues to perform motion planning based on the previous task and the latest actual state information of the wheel-leg hybrid robot.

[0043] In the embodiment of the present application, the target task refers to the task that the wheel-leg hybrid robot is about to perform, which can be a whole-body task, such as crawling, grabbing objects, or avoiding obstacles.

[0044] Furthermore, the target task may be any work task input by a user, or may be any work task obtained by the wheel-leg hybrid robot from a task allocation database. This application does not impose any restrictions on the method of obtaining the target task.

[0045] In the embodiment of the present application, the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers can be understood as motion tasks that need to be performed synchronously by different components of the wheel-foot composite robot when performing the target task.

[0046] according to Figure 1 As can be seen from the structure of the wheel-leg hybrid robot shown, the wheel-leg hybrid robot of the present application can support both wheeled motion modules and wheel-leg hybrid motion modules. Considering that the four motion mechanisms of the wheel-leg hybrid robot include two rear rollers and two front leg mechanisms, the rear rollers are always in a wheeled motion mode. And because each front leg mechanism is provided with a roller at the end, the front leg mechanism can include two motion modes: a wheeled motion mode and a footed motion mode.

[0047] Correspondingly, the wheel-foot composite motion modes supported by the wheel-foot composite robot can be divided into: the first wheel-foot composite motion mode obtained based on the wheel motion mode of the two rear rollers, the foot motion mode of the left front leg mechanism and the wheel motion mode of the right front leg mechanism; the second wheel-foot composite motion mode obtained based on the wheel motion mode of the two rear rollers, the wheel motion mode of the left front leg mechanism and the foot motion mode of the right front leg mechanism; and the third wheel-foot composite motion mode obtained based on the wheel motion mode of the two rear rollers and the foot motion mode of the two front leg mechanisms.

[0048] In the embodiments of the present application, the motion mode of the wheel-leg hybrid robot is specifically a wheel-leg hybrid motion mode, and this wheel-leg hybrid motion mode specifically refers to a third wheel-leg hybrid motion mode derived from the wheeled motion mode of the two rear rollers and the leg-leg motion mode of the two front leg mechanisms. That is, the rear leg mechanisms of the wheel-leg hybrid robot are in the wheeled motion mode, and the front leg mechanisms are in the leg-leg motion mode.

[0049] Among them, because the end of the front leg mechanism is provided with a roller, in order to make the front leg mechanism be in the foot-type motion mode, the present application can lock the roller in the front leg mechanism through the motor corresponding to the ankle joint in the front leg mechanism, so that the roller in the front leg mechanism is switched from the wheel-type motion mode to the supporting foot in contact with the ground, thereby ensuring that the front leg mechanism can be in the foot-type motion mode.

[0050] In some optional embodiments, the robot body can be understood as the robot torso. The robot torso occupies a certain physical space. Therefore, to determine the first motion task of the robot torso, the present application can reduce the difficulty of determining the first motion task by determining the first motion task of the robot torso center point. In other words, the first motion task of the robot body in the present application specifically refers to the first motion task of the robot body center point.

[0051] When the two front leg mechanisms of the wheel-leg hybrid robot in this application are in the foot-type motion mode, the motion of the left and right front leg mechanisms is similar to a human walking gait. That is, in each motion cycle, one front leg mechanism leaves the ground and is in a vacant state, while the other front leg mechanism contacts the ground and is in a supporting state. Among them, the front leg mechanism that leaves the ground and is in a vacant state can be referred to as a vacant leg mechanism, and the front leg structure that contacts the ground and is in a supporting state can be referred to as a supporting leg mechanism. Accordingly, the vacant leg mechanism corresponds to the vacant foot, and the supporting leg mechanism corresponds to the supporting foot.

[0052] In some optional embodiments, the walking gait of the two front leg mechanisms of the wheel-foot composite robot can be determined by a preset gait period t, a support time t stand , and the supporting phase t phase That is, according to different gait cycles t, support time t stand, and the supporting phase t phase , different gaits can be generated. It should be noted that the gait cycle t and support time t in this application are stand , and the supporting phase t phase These are all adjustable parameters, which can be flexibly adjusted according to the walking gait planning requirements of the front leg mechanism of the wheel-leg composite robot. This application does not impose any restrictions on this.

[0053] For example, assuming that the walking gait of the two front leg mechanisms of the wheel-foot hybrid robot is a cross gait, the gait of the left front leg mechanism of the two front leg mechanisms can be t=1 second (s), t stand = 0.5s and t phase = 0%; the gait of the right front leg mechanism can be t = 1 second (s), t stand = 0.5s and t phase =50%.

[0054] Considering that the supporting foot corresponding to the front leg mechanism in the supporting state of the two front leg mechanisms will always be in contact with the ground and assuming that there is no slippage, the position of the supporting foot is always constant in the world coordinate system and is known, while the landing point position of the front leg mechanism in the airborne state corresponding to the airborne foot is unknown. Therefore, in order to ensure that the wheel-leg composite robot maintains its body balance during movement, it is necessary to control the landing point position of the front leg mechanism in the airborne state corresponding to the airborne foot. That is, determining the second motion task of the two front leg mechanisms in this application is specifically determining the second motion task of the airborne leg mechanism of the two front leg mechanisms.

[0055] In some optional embodiments, since the two rear rollers of the wheel-foot composite robot are active wheels during movement, there may be factors such as control errors, causing the rear rollers to move too fast or too slow, thereby causing the position of the rear rollers of the wheel-foot composite robot to deviate from the preset position, that is, the rear rollers to rotate. Therefore, in order to make the movement process of the rear rollers fit the normal movement of the wheel-foot composite robot, it is necessary to ensure that the center points of the two rear rollers are always at a certain position directly behind the robot torso, so as to ensure the balance of the entire wheel-foot composite robot. That is, the third motion task of determining the two rear rollers in this application specifically refers to the third motion task of determining the center points of the two rear rollers.

[0056] In the present application, the center points of the two rear rollers mentioned above can be the center points of the connecting parts with rear rollers at both ends, such as Figure 3 shown.

[0057] In some optional embodiments, such as Figure 4As shown, the motion scale block 210 of the present application may include a trajectory determination unit 211 and a task determination unit 212, wherein the trajectory determination unit 211 is used to determine the whole-body motion trajectory of the wheel-leg composite robot according to the target task; the task determination unit 212 is used to determine the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers according to the whole-body motion trajectory and the first actual state information.

[0058] As an optional implementation, trajectory determination unit 211 may employ a preset robot motion trajectory planning method to plan the full-body motion trajectory of the wheel-leg hybrid robot for performing the target task based on the target task. It should be understood that the full-body motion trajectory is the expected full-body motion trajectory of the wheel-leg hybrid robot when performing the task to be performed.

[0059] In this application, the whole-body motion trajectory may include the robot's main motion trajectory, the lofted leg's corresponding lofted foot's motion trajectory, and the two rear roller center point's motion trajectory. The robot's main motion trajectory includes expected motion parameters for determining the robot's main motion task, such as expected position information, expected posture information, and expected speed information. Similarly, the lofted leg's corresponding lofted foot's motion trajectory includes expected motion parameters for determining the two front leg's lofted leg's second motion task. Furthermore, the two rear roller center point's motion trajectory includes expected motion parameters for determining the two rear roller center points' third motion task.

[0060] It should be noted that when planning the motion trajectories of the two rear roller center points, in order to maintain the balance of the entire wheel-leg hybrid robot, this application requires that the positions of the two rear roller center points always be controlled at a fixed distance behind the robot body (i.e., the robot torso). As for the posture of the two rear roller center points, the roll angle (roll) and pitch angle (pitch) are determined according to the terrain. On flat ground, the roll angle (roll) and pitch angle (pitch) are set to 0, and the yaw angle (yaw) is consistent with the yaw angle of the robot body. This achieves the purpose of planning the motion trajectories of the two rear roller center points.

[0061] The positions of the center points of the two rear rollers are located at a fixed distance behind the robot torso. This parameter can be determined based on the actual debugging of the physical structure of the wheel-foot composite robot.

[0062] As an optional implementation method, the present application plans the motion trajectory of the center points of the two rear rollers, specifically by obtaining the expected position, expected posture, expected speed and expected angular velocity of the robot body. Then, based on the forward kinematics (FK) algorithm, the expected position, expected posture, expected speed and expected angular velocity of the center points of the two rear rollers in the robot body coordinate system are obtained. Then, according to the conversion relationship between the robot body coordinate system and the world coordinate system, the expected position, expected posture, expected speed and expected angular velocity of the center points of the two rear rollers in the robot body coordinate system are converted into the expected position, expected posture, expected speed and expected angular velocity of the center points of the two rear rollers in the world coordinate system, thereby obtaining the motion trajectory of the center points of the two rear rollers. Among them, the conversion relationship between the robot body coordinate system and the world coordinate system is known information.

[0063] It is worth noting that the posture information and angular velocity information of the center points of the two rear rollers in this application are specifically the posture information and angular velocity information of the connecting rod where the center points of the two rear rollers are located.

[0064] After the trajectory determination unit 211 determines the whole-body motion trajectory of the wheel-leg compound robot to perform the target task, the task determination unit 212 of the present application can obtain the motion parameters required for determining the first motion task, the second motion task, and the third motion task from the whole-body motion trajectory. Then, the PD control law is used to determine the first motion task, the second motion task, and the third motion task based on the acquired motion parameters and the latest actual state information of the wheel-leg compound robot. The latest actual state information of the wheel-leg compound robot includes the actual motion parameters required for determining the first motion task of the robot body, the second motion task of the air leg mechanism in the two front leg mechanisms, and the third motion task of the center point of the two rear rollers, such as actual position information, actual posture information, and actual speed information.

[0065] The PD control law can be understood as the PD control algorithm. It should be understood that the PD control algorithm is a simplified form of the PID control algorithm, primarily regulating the system output value and output rate of change to achieve stable system operation. Compared to the PID control algorithm, the PD control algorithm is simpler, clearer, and easier to implement.

[0066] Typically, a PD control algorithm consists of two parts: a proportional component and a differential component. The proportional component calculates the difference between the current error and the setpoint and multiplies it by a proportional coefficient, Kp, to produce a first output signal. The differential component multiplies the rate of change between the current error and the previous error by a differential coefficient, Kd, ​​to produce a second output signal. The first and second output signals are then added together to produce the final output signal.

[0067] After determining the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers, the motion planning module 210 can send the above-mentioned first motion task, second motion task, and third motion task to the motion control module 220, so that the motion control module 220 can perform whole-body motion control of the wheel-foot compound robot according to the first motion task, the second motion task, and the third motion task, so that the wheel-foot compound robot can perform the target task efficiently, smoothly, and accurately.

[0068] In some optional embodiments, considering that each joint in each motion mechanism of the wheel-foot hybrid robot in the present application is provided with a motor, such as a hip joint provided with a hip joint motor, a knee joint provided with a knee joint motor, and an ankle joint provided with an ankle joint motor, the motion control module 220 of the present application may include a torque determination unit 221 and a motion control unit 222, see Figure 5 The torque determination unit 221 is used to determine the target torque of each joint in each motion mechanism according to the first motion task, the second motion task, and the third motion task; the motion control unit 222 is used to control the whole-body motion of the wheel-leg hybrid robot according to the target torque of each joint in each motion mechanism.

[0069] That is, the motion control module 220 of the present application controls the whole-body motion of the wheel-leg hybrid robot based on the first motion task, the second motion task, and the third motion task. Specifically, the motion control module 220 determines the target torque for each joint in each motion mechanism based on the first motion task, the second motion task, and the third motion task. Then, the whole-body motion of the wheel-leg hybrid robot is controlled based on the target torque for each joint in each motion mechanism.

[0070] In the present application, the torque determination unit 221 determines the target torque of each joint in each motion mechanism based on the first motion task, the second motion task and the third motion task. The first motion task, the second motion task and the third motion task can be edited and processed into a quadratic programming form, and the first motion task, the second motion task and the third motion task in the quadratic programming form can be solved using a convex optimization algorithm to obtain the target torque of each joint in each motion mechanism.

[0071] That is, this application converts the whole-body motion control problem of the wheel-leg composite robot into a standard convex optimization problem, solves the standard convex optimization problem to obtain the global optimal solution, and uses the global optimal solution as the target torque of each joint in each motion mechanism.

[0072] In some optional embodiments, the above standard convex optimization problem can be expressed as follows:

[0073]

[0074] Where x is the optimization variable and ||x|| is the cost function;

[0075] is a constraint set, where Constraints for each motion task; is the dynamic constraint; DF≤f μ is the friction cone constraint; is the control variable; lower≤x≤upper is the boundary constraint of the control variable, where lower and upper are adjustable parameters that can be set dynamically according to the robot hardware. The upper and lower limits of the parameter τ in x are the maximum torque of the joint motor.

[0076] above It is based on the robot dynamics equation: Obtained by deformation.

[0077] The meanings of the parameters in the above robot dynamics equation are as follows:

[0078] N is the degree of freedom of the robot. In this application, the degree of freedom is an adjustable parameter. For example, 18 dimensions can be selected, including 6 dimensions for the robot trunk and 12 dimensions for the joints. This application does not impose any restrictions on the degree of freedom of the wheel-leg composite robot.

[0079] q∈R N is the generalized position of the robot, where R N Represents an N-dimensional variable on the set of real numbers;

[0080] is the generalized speed of the robot;

[0081] is the generalized acceleration of the robot;

[0082] τ∈R N-6 is the joint driving torque of the robot;

[0083] f i ∈R 3 is the ground force at the i-th contact point in the world coordinate system;

[0084] H∈R N×N is the robot mass matrix, which can be quickly calculated using the Composite Rigid Body Algorithm (CRBA);

[0085] C∈R N×NTo describe the centrifugal force and Coriolis force of the robot, the Newton-Euler iterative algorithm (RNEA) is used for fast calculation;

[0086] G∈R N The gravity bias of the robot is calculated quickly using the Newton-Euler iterative algorithm (RNEA).

[0087] S∈R (N-6)×N To select the matrix used to distinguish active joints from undriven joints, record it as S = diag(0,0,0,0,0,0,a1,a2,...,an), where a i =0 means it is an undriven joint. The first six items can be directly set to 0 because there is no drive.

[0088] J i ∈R 3×N is the Jacobian matrix of the i-th contact point. It should be noted that because the i-th contact point only considers the Jacobian matrix of the position, and the actual Jacobian matrix is ​​6xN dimensions, here only 3xN dimensions are considered because only the position is considered.

[0089] The above friction cone constraint DF≤f μ It can be obtained by transforming the following formula:

[0090] where f x is the friction component of the contact point in the X-axis direction in the local coordinate system, f y is the friction component of the contact point in the Y-axis direction in the local coordinate system, f z is the friction component of the contact point in the Z-axis direction in the local coordinate system, f μ It should be understood that the friction force between each contact point and the ground in this application is three-dimensional, namely f x 、f y and f z .

[0091] In this application, the local coordinate system can be understood as the coordinate system of any local ground surface when the wheel-leg hybrid robot contacts the ground. For example, if the wheel-leg hybrid robot contacts the ground on a slope, the coordinate system of the slope is the local coordinate system.

[0092] Considering that when the wheel-foot hybrid robot moves to any local ground, the local ground can be a known ground or can be obtained by using a perception algorithm based on the environmental information collected by the environmental perception sensor, so the local coordinate system is obtained Afterwards, the The local coordinate system is transformed into the world coordinate system to obtain the friction cone matrix D in the friction cone constraint. In addition, F in the friction cone constraint can be flexibly set according to the mass of the wheel-foot hybrid robot, and this application does not impose any restrictions on this.

[0093] In this application, the above motion task constraints In the equation, J represents the Jacobian matrix of the contact point, q represents the generalized position, and [·] represents the first differential, that is, the velocity can be obtained based on the generalized position q.

··

[0094] Considering that the control variables in this application are in the joint space, and the first motion task, the second motion task and the third motion task corresponding to the wheel-foot composite robot are all in three-dimensional space (i.e., Cartesian space), in order to obtain the constraints of each motion task This application can establish the connection between the joint space and the three-dimensional space (i.e., Cartesian space) through the following formula (2):

[0095]

[0096] Next, by deriving formula (2), we can get the motion task constraints corresponding to each motion task: In this motion task constraint middle, is the acceleration of the robot body center point in the Cartesian coordinate system corresponding to the first motion task, the acceleration of the flight leg of the flight leg mechanism in the Cartesian coordinate system corresponding to the second motion task, or the acceleration of the two rear roller center points in the Cartesian coordinate system corresponding to the third motion task. J is the Jacobian matrix of the above motion tasks. In order to achieve the joint acceleration corresponding to the above motion tasks, is the current joint velocity. is a constant at every moment; and, The constraint can be set according to the maximum acceleration of the robot, and this application does not impose any specific restrictions on this.

[0097] In some optional embodiments, before determining the target torque of each joint in each motion mechanism of the wheel-foot hybrid robot based on the aforementioned formula (1), the present application optionally first calculates the Jacobian matrix of the first motion task, the Jacobian matrix of the second motion task, and the Jacobian matrix of the third motion task corresponding to each timestamp. Then, the Jacobian matrix of the first motion task, the Jacobian matrix of the second motion task, and the Jacobian matrix of the third motion task corresponding to the same timestamp are brought into the aforementioned motion task constraints. In the equation (3), we can get the three motion task constraints corresponding to the same timestamp, as shown in the following formula:

[0098]

[0099] in, is the first motion task constraint, is the second motion task constraint, is the third motion task constraint.

[0100] In this application, the Jacobian matrix of the first motion task, the Jacobian matrix of the second motion task, and the Jacobian matrix of the third motion task corresponding to each timestamp are calculated. Please refer to the prior art and will not be described in detail here.

[0101] Afterwards, the three motion task constraints corresponding to different time stamps described in the above formula (3) are substituted into the above formula (1) to obtain the target torque of each joint of the wheel-leg hybrid robot at different time stamps.

[0102] In this application, the joints of the wheel-leg composite robot may include: the left hip joint, left knee joint and left ankle joint in the left front leg mechanism, the right hip joint, right knee joint and right ankle joint in the right front leg mechanism, and the hind leg hip joint, hind leg knee joint and hind leg ankle joint in the hind leg mechanism. For details, please refer to Figure 1 part.

[0103] It should be noted that the Jacobian matrix of the above-mentioned first motion task is specifically the Jacobian matrix of the center point of the robot body, the Jacobian matrix of the second motion task is specifically the Jacobian matrix of the sole corresponding to the air leg mechanism in the front leg mechanism, and the Jacobian matrix of the third motion task is specifically the Jacobian matrix of the center points of the two rear rollers.

[0104] Furthermore, the Jacobian matrices for the first and third motion tasks can both be 6*18 dimensional matrices. The 6 represents the 6 degrees of freedom, specifically the 3 rotational dimensions and 3 translational dimensions along the X, Y, and Z axes. The 18 represents the degrees of freedom of the wheel-legged hybrid robot. Because the second task is the foot point, only 3D position is generally considered, meaning that the Jacobian matrix for the second motion task is a 3*18 dimensional matrix, where the 3 represents the 3 degrees of freedom, specifically the 3 translational dimensions along the X, Y, and Z axes.

[0105] In addition, each of the above motion tasks may be three-dimensional or six-dimensional, wherein the motion tasks include a first motion task, a second motion task, and a third motion task.

[0106] The above three dimensions can be understood as three degrees of freedom, specifically three rotation dimensions or three translation dimensions along the X-axis, Y-axis and Z-axis. The six dimensions can be understood as six degrees of freedom, specifically three rotation dimensions and three translation dimensions along the X-axis, Y-axis and Z-axis.

[0107] After determining the target torque of each joint in each motion mechanism, the present application can control the whole-body motion of the wheel-leg hybrid robot according to the target torque of each joint in each motion mechanism through the motion control unit 222.

[0108] Since each joint of the wheel-leg compound robot of the present application corresponds to a motor, after obtaining the target torque of each joint in each motion mechanism in the wheel-leg compound robot, the present application can control each motor to output the target torque to the corresponding joint according to the target torque of each joint, so as to drive each joint to rotate the corresponding angle, so that the other joints and related mechanisms associated with each joint produce a certain amount of movement in space, thereby realizing the whole-body motion control of the wheel-leg compound robot.

[0109] In an embodiment of the present application, when the motion control module 220 controls the whole-body motion of the wheel-leg composite robot, it can perform whole-body motion control of the wheel-leg composite robot according to a preset control cycle. The control cycle of the motion control module 220 can be determined based on factors such as the motion control requirements of the wheel-leg composite robot, hardware computing power, and algorithm complexity. In some optional implementations, if the real-time requirements for the motion control of the wheel-leg composite robot are high and the hardware computing power is strong, the control cycle can be set to a smaller value to achieve real-time control. For example, the control cycle can be set to 1 millisecond (ms), meaning that a motion control operation is performed every 1ms. If the real-time requirements for the motion control of the wheel-leg composite robot are low and the hardware computing power is insufficient, the control cycle can be set to a larger value. For example, the control cycle can be set to 5ms, meaning that a motion control operation is performed every 5ms. This application does not specifically limit the device with the control cycle.

[0110] It should be noted that when the motion control module 220 performs motion control according to a control cycle, if the first motion task, second motion task, and / or third motion task sent by the motion planning module 210 is the first motion task, second motion task, and / or third motion task of the previous control cycle, then the motion control module 220 continues to perform whole-body motion control of the wheel-leg hybrid robot based on the same first motion task, second motion task, and / or third motion task. If the motion planning module 210 sends a new first motion task, new second motion task, and / or new third motion task, then the motion control module 220 performs whole-body motion control of the wheel-leg hybrid robot based on the new first motion task, new second motion task, and / or new third motion task.

[0111] It is worth noting that as long as the wheel-foot composite robot is in working state, the actual state information of the wheel-foot composite robot can be determined through the state perception module 230, and the actual state information of the wheel-foot composite robot can be sent to the motion planning module 210, so that the motion planning module 210 can select the actual state information of a required wheel-foot composite robot for motion planning operation according to the planning requirements.

[0112] In this application, the wheel-leg hybrid robot also includes an inertial sensor, a foot odometer, and a wheel odometer. Optionally, the inertial sensor, foot odometer, and wheel odometer may be provided within the robot body 11, and this application imposes no limitations thereto. Therefore, the state perception module 230 can determine the actual state information of the wheel-leg hybrid robot based on the inertial sensor, foot odometer, and wheel odometer.

[0113] Among them, the inertial sensor can be selected as an inertial measurement unit (IMU) or a device including devices such as an accelerometer and a gyroscope, etc., and this application does not impose any restrictions on this.

[0114] The aforementioned foot-based and wheel-based odometers are two different types of odometers, both used to measure and estimate information such as the position, posture, and velocity of wheel-leg hybrid robots. Wheel rotation information or sensor measurements are typically used to calculate the robot's displacement and directional changes. By continuously accumulating and updating this motion information, an estimate of the robot's position, posture, and velocity relative to its starting position can be provided.

[0115] See Figure 6 In this application, the state perception module 230 may include: a prediction information determination unit 231, a measurement information determination unit 232, a first state determination unit 233 and a second state determination unit 234.

[0116] The prediction information determination unit 231 is used to collect inertial data through an inertial sensor and determine the predicted state information of the robot body according to the inertial data;

[0117] a measurement information determining unit 232 for determining measurement status information of the robot body based on the foot odometer and the wheel odometer;

[0118] A first state determination unit 233 is used to determine actual state information of the robot body according to the measured state information and the predicted state information;

[0119] The second state determination unit 234 is used to determine the actual state information of the flight foot corresponding to the flight leg mechanism in the two front leg mechanisms and the actual state information of the center points of the two rear rollers based on the actual state information of the robot body, and obtain the actual state information of the wheel-foot composite robot based on the actual state information of the robot body, the actual state information of the flight foot and the actual state information of the center points of the two rear rollers.

[0120] In this application, the above-mentioned predicted state information refers to the a priori predicted state information of the wheel-leg hybrid robot body. The above-mentioned measured state information refers to the information obtained by measuring the state of the wheel-leg hybrid robot body.

[0121] Considering that the inertial sensor can collect inertial data of the wheel-leg hybrid robot body in real time, the present application can collect inertial data through the inertial sensor during each sensing cycle and determine the measurement state information of the wheel-leg hybrid robot body based on the collected inertial data. The above-mentioned inertial data includes acceleration and angular velocity.

[0122] As an optional implementation, the prediction information determination unit 231 in the present application can specifically process the inertial data collected by the inertial sensor according to the state transition parameter to obtain the predicted state information of the wheel-foot composite robot body in the current perception cycle, as shown in the following formula (4):

[0123]

[0124] in, is the prior predicted state information of the wheel-leg hybrid robot in the current perception cycle k; k-1 is a state transfer parameter used to describe the prior predicted state of the current perception cycle k based on the posterior actual state information of the wheel-leg hybrid robot in the previous perception cycle k-1, and the state transfer parameter is specifically a state transfer matrix; B is the actual state information of the wheel-leg composite robot in the last sensing cycle k-1; k-1is the control matrix determined based on the external force, which can be understood as the input state transfer parameter, used to describe the conversion relationship between the input ambient noise and state information under the external force; ω k is the inertial data of the current sensing cycle k collected by the inertial sensor. k-1 The existing method can be used for determination, and this application does not impose any limitation on this.

[0125] It should be understood that the above-mentioned predicted state information may include: position information, posture information and speed information, wherein the speed information includes linear velocity and angular velocity.

[0126] Taking into account that the foot odometer and wheel odometer can obtain joint angle data from the joint encoders corresponding to the corresponding joints of the wheel-foot composite robot, the measurement information determination unit 232 of the present application can specifically determine the first measurement state information of the wheel-foot composite robot body in the current perception cycle k based on the joint angle data of the wheel-foot composite robot in the current perception cycle k obtained from the joint encoders of the corresponding joints through the foot odometer, and determine the second measurement state information of the wheel-foot composite robot body in the current perception cycle k based on the joint angle data of the wheel-foot composite robot in the current perception cycle k obtained from the joint encoders of the corresponding joints through the wheel odometer.

[0127] In this application, the measurement status information may include: position information, posture information and speed information, wherein the speed information includes linear velocity and angular velocity.

[0128] In some optional embodiments, it is considered that each joint in each front leg mechanism corresponds to a joint encoder, and each joint encoder stores the angle data of the corresponding joint. In addition, the supporting leg mechanism that is in contact with the ground and in a supporting state corresponds to the supporting foot on the ground and is assumed to be unchanged. Therefore, the measurement information determination unit 232 of the present application determines the first measurement state information of the wheel-foot compound robot body in the current perception cycle k based on the joint angle data of the wheel-foot compound robot in the current perception cycle k obtained from the joint encoder of the corresponding joint through the foot odometer. The specific process is as follows:

[0129] First, the first joint angle data of the wheel-leg hybrid robot in the current perception cycle k is obtained from the joint encoder corresponding to each joint in the supporting leg mechanism using the foot odometry. The first joint angle data may be multiple, specifically hip joint angle data, knee joint angle data, and ankle joint angle data. Then, based on the multiple first joint angle data, the first measurement state information of the wheel-leg hybrid robot body in the current perception cycle k is determined.

[0130] In some optional embodiments, the joint angle data stored in the joint encoder can be understood as the joint position. Therefore, the present application determines the first measurement state information of the wheel-foot composite robot body in the current perception cycle k based on multiple first joint angle data, and optionally obtains the actual speed (i.e., joint speed) of each first joint by performing a differential operation on each first joint angle data. Then, based on the joint position and joint speed of each joint in the supporting leg mechanism, the movement speed of the wheel-foot composite robot body relative to the corresponding supporting leg of the supporting leg mechanism in the current perception cycle k is calculated.

[0131] Furthermore, ankle joint angle data is obtained from the plurality of first joint angle data, and then, using the FK algorithm, the position information of the wheel-leg composite robot body relative to the corresponding supporting leg of the supporting leg mechanism in the current sensing period k is calculated based on the ankle joint angle data. The position information includes position information and posture information.

[0132] In the present application, the above-mentioned process of calculating the motion velocity of the wheel-leg composite robot body relative to the corresponding supporting foot of the supporting leg mechanism in the current perception period k based on the joint position and joint velocity of each joint in the supporting leg mechanism is as follows: a Jacobian matrix is ​​calculated based on the joint positions of all joints in the supporting leg mechanism, and then the product of the Jacobian matrix and the joint velocity of each joint is calculated to obtain the motion velocity of the corresponding supporting foot of the supporting leg mechanism relative to the wheel-leg composite robot body. Furthermore, the motion velocity of the corresponding supporting foot of the supporting leg mechanism relative to the wheel-leg composite robot body is multiplied by -1 to obtain the motion velocity of the wheel-leg composite robot body relative to the corresponding supporting foot of the supporting leg mechanism in the current perception period k.

[0133] Optionally, the above calculation of the movement speed of the wheel-leg composite robot body relative to the corresponding supporting foot of the supporting leg mechanism in the current sensing period k can be achieved by the following formula (5):

[0134]

[0135] Among them, J k is the Jacobian matrix calculated based on the joint positions of all joints in the supporting leg mechanism in the supporting state at the current sensing period k, is the joint velocity of each joint in the supporting leg mechanism in the supporting state during the current sensing period k, v k is the movement speed of the wheel-leg composite robot body relative to the corresponding supporting foot of the supporting leg mechanism in the current sensing period k.

[0136] In addition, the present application uses the FK algorithm to calculate the posture information of the wheel-foot composite robot body relative to the corresponding supporting foot of the supporting leg mechanism in the current sensing period k based on the ankle joint angle data, which can be achieved by the following formula (6):

[0137] body P foot =FK(q)........................(6)

[0138] in, body P foot is the position information of the wheel-leg composite robot body relative to the corresponding supporting foot of the supporting leg mechanism in the current perception period k, FK() is the forward kinematics algorithm, and q is the ankle joint angle data.

[0139] Considering that the conversion relationship between the supporting leg coordinate system and the world coordinate system is known or can be calculated through a conversion algorithm, after obtaining the measurement status information (position information, posture information and speed information) of the wheel-foot composite robot body relative to the supporting leg corresponding to the supporting leg mechanism, the present application can convert the measurement status information of the wheel-foot composite robot body relative to the supporting leg into the measurement status information of the wheel-foot composite robot body relative to the world coordinate system based on the conversion relationship between the supporting leg coordinate system and the world coordinate system, so as to obtain the first measurement status information of the wheel-foot composite robot body in the world coordinate system.

[0140] In some optional embodiments, the hind leg mechanism of the wheel-foot composite robot includes two rear rollers, as well as a rear thigh mechanism, a rear calf mechanism, and a connecting component for the two rear rollers, which are shared by the two rear rollers. For the rear rollers of the wheel-foot composite robot, since the rear rollers are always on the ground, assuming that there is pure rolling between the rear rollers and the ground, the measurement information determination unit 232 of the present application can obtain the second joint angle data corresponding to each rear roller in the current perception cycle k from the joint encoder corresponding to each rear roller through the wheel odometer. In addition, the radius of each rear roller is also obtained from the configuration information of the wheel-foot composite robot. Then, based on the second joint angle data corresponding to each rear roller and the radius of each rear roller, the second measurement state information of the wheel-foot composite robot body in the current perception cycle k is determined.

[0141] Among them, the number of second joint angle data corresponding to the first rear roller and the second rear roller respectively is multiple. Specifically, the second joint angle data corresponding to the first rear roller includes: hip joint angle data, knee joint angle data and ankle joint angle data corresponding to the first rear roller; the second joint angle data corresponding to the second rear roller includes: hip joint angle data, knee joint angle data and ankle joint angle data corresponding to the second rear roller.

[0142] In some optional embodiments, the second measurement state information of the wheel-foot composite robot body in the current perception period k is determined based on the second joint angle data corresponding to each rear roller and the radius of each rear roller. The specific implementation process is as follows: the rolling distance of each rear roller on the ground can be first calculated by the following formula (7):

[0143] D m =q m ×r m …………………………………(7)

[0144] Among them, D m is the rolling distance of the mth rear roller on the ground in the current sensing cycle k, q m is the second joint angle data corresponding to the mth rear roller in the current sensing cycle k, r m is the radius of the mth rear roller, where m∈[1,2].

[0145] Because the two rear rollers are constrained by the differential gear system, the actual distance traveled by the center points of the two rear rollers on the ground is calculated based on the rolling distance of each rear roller. Furthermore, the speed of the two rear roller center points relative to the wheel-foot hybrid robot is calculated based on the actual distance traveled by the center points of the two rear rollers and the difference between the previous sensing cycle k and the current sensing cycle k. The speed of the two rear roller center points relative to the wheel-foot hybrid robot is then multiplied by -1 to obtain the speed of the wheel-foot hybrid robot relative to the two rear roller center points during the current sensing cycle k.

[0146] In addition, the present application can also use the FK algorithm to calculate the position information of the wheel-foot composite robot body relative to the two rear roller center points in the current perception period k based on the actual movement distance of the two rear roller center points on the ground.

[0147] Considering that the conversion relationship between the coordinate system corresponding to the center points of the two rear rollers (local coordinate system) and the world coordinate system is known or can be calculated through a conversion algorithm, after obtaining the measurement status information of the wheel-foot composite robot body relative to the center points of the two rear rollers, the present application can convert the measurement status information of the wheel-foot composite robot body relative to the two rear roller center points into the measurement status information of the wheel-foot composite robot body relative to the world coordinate system based on the conversion relationship between the local coordinate system and the world coordinate system, so as to obtain the second measurement status information of the wheel-foot composite robot body in the world coordinate system.

[0148] After obtaining the predicted state information and measured state information (first measured state information and second measured state information), the first state determination unit 233 can first use the state transition parameters, control matrix, and state transition noise corresponding to the wheel-leg hybrid robot to process the posterior estimated covariance of the wheel-leg hybrid robot body in the previous perception cycle k-1 to obtain the prior estimated covariance of the wheel-leg hybrid robot body in the current perception cycle k. Next, based on the observation matrix, the observation noise covariance matrix, and the observation noise variance, the prior estimated covariance of the current perception cycle k is processed to obtain the filter gain of the current perception cycle k. Furthermore, using the Extended Kalman Filter (EKF) algorithm, based on the filter gain of the current perception cycle k and the observation matrix, the measured state information and predicted state information of the wheel-leg hybrid robot body in the current perception cycle k are fused to obtain the actual state information of the wheel-leg hybrid robot body in the current perception cycle k.

[0149] As an optional implementation method, the above-mentioned state transition parameters, control matrix, and state transition noise corresponding to the wheel-leg composite robot are used to process the posterior estimated covariance of the wheel-leg composite robot body in the previous perception cycle k-1 to obtain the prior estimated covariance of the wheel-leg composite robot body in the current perception cycle k, which can be achieved by the following formula (8):

[0150]

[0151] in, is the prior estimated covariance of the wheel-leg hybrid robot in the current perception cycle k, F k-1 is the state transition parameter of the previous sensing cycle k-1, is the posterior estimated covariance of the last sensing cycle k-1, is the transpose of the state transition parameter of the previous sensing cycle k-1, B k-1 is the control matrix determined based on the external force, which can be understood as the input state transfer parameter, used to describe the conversion relationship between the input ambient noise and state information under the external force, Q k is the uncertainty of the motion environment corresponding to the current perception period k, is the transpose of the control matrix.

[0152] Moreover, based on the observation matrix, the observation noise covariance matrix and the observation noise variance, the prior estimated covariance of the wheel-leg hybrid robot body in the current perception period k is processed to obtain the filter gain of the current perception period k, which can be achieved by the following formula (9):

[0153]

[0154] Among them, K k is the filter gain of the current sensing period k, is the prior estimated covariance of the wheel-leg hybrid robot in the current perception cycle k, G k is the observation matrix of the current perception period k, which is used to map the state matrix to the matrix of the observation space and describe the relationship between the state variables and the observation variables. is the observation matrix G k The transpose of C k is the observation noise covariance matrix of the current sensing period k, which is used to describe the statistical characteristics of the observation noise, including the variance and covariance of the observation error. is the observation noise covariance matrix C k The transpose of R k is the observation noise variance.

[0155] In addition, the EKF algorithm is used to fuse the measured state information and predicted state information of the wheel-leg hybrid robot in the current perception period k based on the filter gain and observation matrix of the current perception period k, and the actual state information of the wheel-leg hybrid robot in the current perception period k is obtained, which can be achieved by the following formula (10):

[0156]

[0157] in, is the actual state information of the wheel-foot composite robot in the current perception cycle k, K is the prior predicted state information of the wheel-leg hybrid robot in the current perception cycle k, k is the filter gain of the current sensing period k, y k G is the measurement state information of the wheel-foot composite robot in the current perception cycle k. In this application, the measurement state information includes the first measurement state determined by the foot odometer and the second measurement state determined by the wheel odometer. k is the observation matrix of the current perception period k.

[0158] That is, the present application obtains the actual state information of the wheel-leg composite robot body in the current perception period k by fusing the measured state information and predicted state information of the wheel-leg composite robot body in the current perception period k.

[0159] In this application, the above-mentioned actual state information may include: position information, posture information and speed information. Among them, the speed information includes linear velocity and angular velocity.

[0160] In some optional embodiments, when the second state determination unit 234 in the present application determines the actual state information of the wheel-leg hybrid robot, the following steps may be included:

[0161] Step 1: Using the forward kinematics (FK) algorithm, the actual state information of the flying foot relative to the robot body and the actual state information of the center points of the two rear rollers relative to the robot body are calculated based on the actual state information of the robot body.

[0162] Step 2: Based on the conversion relationship between the robot body coordinate system and the world coordinate system, the actual state information of the airborne foot relative to the robot body is converted into the actual state information of the airborne foot in the world coordinate system, and the actual state information of the center points of the two rear rollers relative to the robot body is converted into the actual state information of the center points of the two rear rollers in the world coordinate system.

[0163] Among them, the conversion relationship between the robot body coordinate system and the world coordinate system is a known quantity.

[0164] In this application, the actual position information of the robot body includes: actual position information, actual posture information, actual speed information and actual angular velocity information; the actual state information of the air leg mechanism corresponding to the air leg in the two front leg mechanisms includes: actual position information and actual speed information; the actual state information of the center points of the two rear rollers includes: actual position information, actual posture information, actual speed information and actual angular velocity information.

[0165] After obtaining the actual state information of the aired foot corresponding to the aired leg mechanism in the two front leg mechanisms and the actual state information of the center points of the two rear rollers, the second state determination unit 234 can determine the actual state information of the robot body, the actual state information of the aired foot corresponding to the aired leg mechanism in the two front leg mechanisms and the actual state information of the center points of the two rear rollers as the actual state information of the wheel-foot composite robot, and send the actual state information of the wheel-foot composite robot to the motion planning module 210, so that the motion planning module 210 performs motion planning for the wheel-foot composite robot for the next planning cycle according to the actual state information of the wheel-foot composite robot perceived by the state perception module 230.

[0166] In an embodiment of the present application, when the state perception module 230 determines the actual state information of the wheel-leg hybrid robot, it may determine the actual state information of the wheel-leg hybrid robot based on a preset perception period. The perception period of the state perception module 230 may be determined based on a combination of factors such as the state perception requirements of the wheel-leg hybrid robot, hardware computing power, and algorithm complexity. In some optional implementations, if the real-time requirements for the actual state perception of the wheel-leg hybrid robot are high or the hardware computing power is strong, the perception period may be set to a smaller value to achieve real-time perception. For example, the perception period may be set to 0.5 milliseconds (ms), meaning that a state perception operation is performed every 0.5 ms. If the real-time requirements for the actual state perception of the wheel-leg hybrid robot are low or the hardware computing power is weak, the perception period may be set to a larger value. For example, the perception period may be set to 5 ms, meaning that a state perception operation is performed every 5 ms. This application does not impose any specific restrictions on the setting of the perception period.

[0167] It should be noted that the state perception module 230 determines the actual state information of the wheel-foot composite robot, which is not related to whether the wheel-foot composite robot performs the work task. That is, when the wheel-foot composite robot is in the working state, as long as the perception cycle is reached, the operation of determining the actual state of the wheel-foot composite robot is executed regardless of whether the wheel-foot composite robot performs the work task.

[0168] The technical solution disclosed in the embodiment of the present application determines the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers through the motion planning module according to the target task and the latest actual state information of the wheel-foot composite robot. Then, the motion control module controls the whole-body motion of the wheel-foot composite robot according to the first motion task, the second motion task, and the third motion task output by the motion planning module. Then, the actual state information of the wheel-foot composite robot is determined through the state perception module, and the actual state information of the wheel-foot composite robot is sent to the motion planning module, so that the motion planning module performs new motion planning according to the actual state information of the wheel-foot composite robot. In this way, the robot's motion can be precisely closed-loop controlled to ensure that the robot can perform tasks efficiently, smoothly and accurately, thereby improving the robot's task execution effect and enhancing the user experience.

[0169] As an optional implementation method in the present application, the robot body motion trajectory of the wheel-leg hybrid robot of the present application includes the expected position information, expected posture information, expected speed information, and expected angle information of the robot body. The lofted leg motion trajectory corresponding to the lofted leg mechanism of the wheel-leg hybrid robot includes the expected position information and expected speed information of the lofted leg. The motion trajectory of the center points of the two rear rollers of the wheel-leg hybrid robot includes the expected position information, expected posture information, expected speed information, and expected angular velocity information of the center points of the two rear rollers.

[0170] In addition, the latest actual status information of the wheel-leg composite robot includes: the latest actual position information, the latest actual posture information, the latest actual speed information and the latest actual angular velocity information of the robot body; the latest actual position information and the latest actual speed information corresponding to the air leg mechanism in the two front leg mechanisms, and the latest actual position information, the latest actual posture information, the latest actual speed information and the latest actual angular velocity of the center points of the two rear rollers.

[0171] Therefore, the task determination unit 212 of the present application determines the first motion task of the robot body, the second motion task of the air leg mechanism in the two front leg mechanisms, and the motion task of the center point of the two rear rollers, which may include the following parts:

[0172] The first step is to determine the first motion task of the robot body, which may include the following steps:

[0173] Step 11: Determine the position motion subtask of the robot body according to the expected position information, the latest actual position information, the expected speed information and the latest actual speed information of the robot body.

[0174] Step 12: Determine the posture motion subtask of the robot body according to the expected posture information, the latest actual posture information, the expected angular velocity information and the latest actual angular velocity information of the robot body.

[0175] Step 13: Obtain the first motion task of the robot body according to the position motion subtask and the posture motion subtask of the robot body.

[0176] It should be understood that the first motion task of the robot body includes: a position motion subtask and a posture motion subtask of the robot body.

[0177] As an optional implementation method, the position motion subtask of the robot body can be determined by the following formula (11):

[0178]

[0179] in, is the position motion subtask of the robot body, and in Corresponding to the dynamic equation 4 to 6 dimensions; p des is the expected position information of the robot body; p act is the latest actual position information of the robot body; des is the expected velocity information of the robot body; v act kp is the latest actual speed information of the robot body; pis the robot body position control gain, kd p is the robot body speed control gain, and kp p and kd p They are adjustable parameters.

[0180] Furthermore, the posture motion subtask of the robot body can be determined by the following formula (12):

[0181]

[0182] in, is the posture motion subtask of the robot body, and in Corresponding to the dynamic equation 1 to 3 dimensions; rpy des is the expected posture information of the robot body, where rpy is the abbreviation of roll-pitch-yaw; rpy act is the latest actual posture information of the robot body; ω des is the expected angular velocity information of the robot body; ω act kp is the latest actual angular velocity information of the robot body; rpy is the robot body posture control gain, kd rpy is the robot body speed control gain, and kp rpy and kd rpy They are adjustable parameters.

[0183] Considering that posture can include multiple representations such as quaternions, rotation matrices, and rpy posture angles. As an optional implementation, the present application uses rpy posture angles to represent posture. That is, the posture motion subtask of the robot body is determined based on the posture of the robot body represented by the rpy posture angle. Of course, the present application can also be determined based on the posture of the robot body represented by quaternions, or based on the posture of the robot body represented by rotation matrices, and there is no limitation on this.

[0184] It should be noted that the execution order of the above steps 11 and 12 can be to execute step 11 first and then step 12; or, to execute step 12 first and then step 11; or, to execute step 11 and step 12 in parallel. This application does not impose any restrictions on this.

[0185] The second part, determining the second motion task of the flight leg mechanism, may specifically include the following steps: determining according to the expected position information, the latest actual position information, the expected speed information and the latest actual speed information of the flight leg.

[0186] In some optional embodiments, the second motion task of the flying leg mechanism in the two front leg mechanisms can be determined by the following formula (13):

[0187]

[0188] in, It is the second motion task of the two front leg mechanisms; des is the expected position information of the airborne foot; act The latest actual position information of the empty foot; des is the expected velocity of the airborne foot; v act is the latest actual speed information of the air foot; kp is the air foot position control gain, kd is the air foot speed control gain, and kp and kd are adjustable parameters.

[0189] It should be noted that in the present application, the expected position of the flying foot of the flying leg mechanism can also be calculated by first calculating the angle that the robot body can rotate within the remaining swing time of the flying leg mechanism through the motion planning module 210 based on the latest actual angular velocity information of the robot body. Then, the position information hhip_pos_local of the flying foot corresponding to the flying leg mechanism in the hip joint coordinate system corresponding to the flying leg mechanism is calculated. After that, the position change of the position information hhip_pos_local caused by the rotation of the robot body at the end of the swing of the flying leg mechanism is calculated. Finally, the expected position of the flying foot of the flying leg mechanism is calculated based on the position information of the robot body in the world coordinate system, the position information hhip_pos_local of the flying foot corresponding to the flying leg mechanism in the hip joint coordinate system corresponding to the flying leg mechanism, the translation speed of the robot body relative to the world coordinate system and the remaining swing time of the flying leg. It should be understood that the expected position of the flying foot of the flying leg mechanism refers to the expected position of the flying foot of the flying leg mechanism at the moment of landing in the world coordinate system.

[0190] The third part, determining the third motion task of the center points of the two rear rollers, may specifically include the following steps:

[0191] Step 21 : determining the position movement subtask of the center points of the two rear rollers according to the expected position information, the latest actual position information, the expected speed information and the latest actual speed information of the center points of the two rear rollers.

[0192] Step 22: determining the posture motion subtasks of the two rear roller center points based on the expected posture information, the latest actual posture information, the expected angular velocity information, and the latest actual angular velocity information of the two rear roller center points.

[0193] Step 23: Obtain a third motion task of the center points of the two rear rollers according to the position motion subtask and the posture motion subtask of the center points of the two rear rollers.

[0194] It should be understood that the third motion task of the center points of the two rear rollers includes: a position motion subtask and a posture motion subtask of the center points of the two rear rollers.

[0195] In some optional implementations, the subtask of determining the position of the center points of the two rear rollers can be implemented by the following formula (14):

[0196]

[0197] in, The position movement subtask for the center points of the two rear rollers; p des is the expected position information of the center points of the two rear rollers; p act is the latest actual position information of the center points of the two rear rollers; des is the desired velocity of the center points of the two rear rollers; v act kp is the latest actual speed information of the center point of the two rear rollers; center kd is the control gain of the center point position of the two rear rollers, center is the speed control gain of the center point of the two rear rollers, and kp center and kd center They are adjustable parameters.

[0198] Furthermore, the posture motion subtasks of the two rear roller center points can be determined by the following formula (15):

[0199]

[0200] in, The posture motion subtask for the center points of the two rear rollers; rpy des is the expected posture information of the center points of the two rear rollers; rpy act is the latest actual posture information of the center points of the two rear rollers; ω des is the expected angular velocity information of the center points of the two rear rollers; ω act kp is the latest actual angular velocity information of the center points of the two rear rollers; center is the robot body posture control gain, kd center is the robot body speed control gain, and kp center and kd center They are adjustable parameters.

[0201] Considering that the posture can include multiple representations such as quaternions, rotation matrices, and rpy posture angles. As an optional implementation method, the present application uses the rpy posture angle to represent the posture. That is, the posture motion subtask of the two rear roller center points is determined based on the posture of the two rear roller center points represented by the rpy posture angle. Of course, the present application can also be determined based on the posture of the two rear roller center points represented by quaternions, or based on the posture of the two rear roller center points represented by rotation matrices, and there is no limitation on this here.

[0202] It should be noted that the execution order of the above steps 21 and 22 can be to execute step 21 first and then step 22; or, to execute step 22 first and then step 21; or, to execute step 21 and step 22 in parallel. This application does not impose any restrictions on this.

[0203] The technical solution disclosed in the embodiments of the present application can achieve precise closed-loop control of the robot's movement to ensure that the robot can perform tasks efficiently, smoothly and accurately, thereby improving the robot's task execution effect and enhancing the user experience.

[0204] Please refer to the attached Figure 7 , describes a robot proposed in an embodiment of the present application. In this application, the robot is the aforementioned Figure 1 The wheel-leg compound robot shown in the figure comprises a robot body and four motion mechanisms arranged on the robot body, wherein the four motion mechanisms include two rear rollers and two front leg mechanisms. Figure 7 As shown, the robot 300 includes the robot motion control system 200 provided in the first embodiment of the present application.

[0205] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0206] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0207] Modules described as separate components may or may not be physically separate, and components displayed as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected based on actual needs to achieve the purpose of the present embodiment. For example, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module.

[0208] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program that has a predetermined function and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0209] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A robot motion control system, characterized in that: The robot is a wheel-leg composite robot, comprising a robot body and four motion mechanisms arranged on the robot body, wherein the four motion mechanisms include two rear rollers and two front leg mechanisms. The system comprises: a motion planning module, a motion control module and a state perception module; The motion planning module is configured to determine, based on the target task and the latest actual state information of the wheel-leg hybrid robot, a first motion task of the robot body, a second motion task of the two front leg mechanisms, and a third motion task of the two rear rollers; The motion control module is configured to control the whole-body motion of the wheel-leg composite robot according to the first motion task, the second motion task, and the third motion task sent by the motion planning module; The state perception module is used to determine the actual state information of the wheel-leg composite robot and send the actual state information of the wheel-leg composite robot to the motion planning module, so that the motion planning module performs new motion planning according to the actual state information of the wheel-leg composite robot.

2. The system according to claim 1, wherein: The motion planning module includes: a trajectory determination unit and a task determination unit; The trajectory determination unit is configured to determine the whole-body motion trajectory of the wheel-leg hybrid robot according to the target task; The task determination unit is used to determine the first motion task of the robot body, the second motion task of the two front leg mechanisms, and the third motion task of the two rear rollers based on the whole-body motion trajectory and the latest actual state information.

3. The system according to claim 2, characterized in that The motion mode of the wheel-foot composite robot is a wheel-foot composite motion mode; Correspondingly, the first motion task is the motion task of the center point of the robot body; The second motion task is a motion task of the air leg mechanism in the two front leg mechanisms; The third motion task is the motion task of the center points of the two rear rollers.

4. The system according to claim 3, characterized in that The whole-body motion trajectory includes: the robot body motion trajectory, the motion trajectory of the soaring leg mechanism corresponding to the soaring foot, and the motion trajectory of the center points of the two rear rollers; Correspondingly, the task determination unit is specifically used to obtain the expected position information, expected posture information, expected speed information and expected angular velocity information of the robot body according to the motion trajectory of the robot body; obtain the expected position information and expected speed information of the soaring foot according to the motion trajectory of the soaring foot corresponding to the soaring leg mechanism; and obtain the expected position information, expected posture information, expected speed information and expected angular velocity information of the center points of the two rear rollers according to the motion trajectories of the center points of the two rear rollers.

5. The system according to claim 4, characterized in that The latest actual status information includes: the latest actual position information, the latest actual posture information, the latest actual speed information and the latest actual angular velocity information of the robot body, the latest actual position information and the latest actual speed information of the corresponding flight legs of the flight leg mechanisms in the two front leg mechanisms, and the latest actual position information, the latest actual posture information, the latest actual speed information and the latest actual angular velocity of the center points of the two rear rollers.

6. The system according to claim 5, characterized in that The task determination unit is further configured to: The position motion subtask of the robot body is determined based on the expected position information, the latest actual position information, the expected speed information and the latest actual speed information of the robot body; the posture motion subtask of the robot body is determined based on the expected posture information, the latest actual posture information, the expected angular velocity information and the latest actual angular velocity information of the robot body; the first motion task of the robot body is obtained based on the position motion subtask and the posture motion subtask of the robot body.

7. The system according to claim 5, characterized in that The task determination unit is further configured to: The second motion task of the flight leg mechanism of the two front leg mechanisms is determined according to the expected position information, the latest actual position information, the expected speed information and the latest actual speed information of the flight leg.

8. The system according to claim 5, wherein: The task determination unit is further configured to: According to the expected position information, the latest actual position information, the expected speed information and the latest actual speed information of the two rear roller center points, the position movement subtask of the two rear roller center points is determined; according to the expected posture information, the latest actual posture information, the expected angular velocity information and the latest actual angular velocity information of the two rear roller center points, the posture movement subtask of the two rear roller center points is determined; according to the position movement subtask and the posture movement subtask of the two rear roller center points, the third movement task of the two rear roller center points is obtained.

9. The system according to claim 1, wherein: The motion control module includes: a torque determination unit and a motion control unit; The torque determination unit is configured to determine a target torque for each joint in each motion mechanism according to the first motion task, the second motion task, and the third motion task; The motion control unit is used to control the whole-body motion of the wheel-leg composite robot according to the target torque of each joint in each of the motion mechanisms.

10. The system according to claim 1, wherein: The wheel-foot hybrid robot further includes an inertial sensor, a foot odometer and a wheel odometer, and the state perception module includes: a prediction information determination unit, a measurement information determination unit, a first state determination unit and a second state determination unit; The prediction information determination unit is configured to collect inertial data through the inertial sensor and determine the predicted state information of the robot body according to the inertial data; The measurement information determining unit is configured to determine the measurement status information of the robot body according to the foot odometer and the wheel odometer respectively; The first state determination unit is configured to determine actual state information of the robot body based on the measured state information and the predicted state information; The second state determination unit is used to determine the actual state information of the soaring leg corresponding to the soaring leg mechanism in the two front leg mechanisms and the actual state information of the center points of the two rear rollers based on the actual state information of the robot body, and obtain the actual state information of the wheel-foot composite robot based on the actual state information of the robot body, the actual state information of the soaring leg and the actual state information of the center points of the two rear rollers.

11. A robot, characterized in that: The robot motion control system comprises the robot motion control system according to claims 1-10.

Images