Motion state control method and device, electronic equipment and storage medium

Abstract

The application provides a motion state control method and device, electronic equipment and a storage medium, and is applied to a wheel-legged robot. The method comprises the following steps: acquiring various kinetic parameters of the wheel-legged robot; constructing a kinetic model based on a total wheel torque, a tracking linear velocity, a pitch angle, a pitch angular velocity, a characteristic parameter, a gravitational acceleration and an intermediate parameter; obtaining a wheel target angular acceleration sum and a wheel target angular acceleration difference based on a linear velocity parameter, a pitch angle parameter, a weight coefficient, a coordination variable, a cost function, the kinetic model, a preset proportional gain coefficient and a yaw angle parameter; and obtaining a left wheel target angular velocity and a right wheel target angular velocity; obtaining a target joint angle based on a preset leg height and a left-right leg target height difference; and controlling the wheel-legged robot to complete a preset motion task based on the target joint angle, the left wheel target angular velocity and the right wheel target angular velocity, thereby improving the accuracy of the gravity center balance control and taking into account the compliant control of the wheel-legged robot.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of robot control, and particularly relates to a motion state method and device, an electronic device and a storage medium. BACKGROUND

[0002] A wheel-legged robot controls a motion state of a robot body through a wheel-legged structure. Since the wheel-legged robot is an unstable under-actuated system, there is a problem of balance control. A current control method of a motion state of a wheel-legged robot adopts a trolley inverted pendulum model or a Proportion Integral Differential (PID) control method. The control effect of the balance of a center of gravity of the wheel-legged robot under a complex motion state cannot meet the demand.

[0003] The prior art has a problem that the control effect of the balance of a center of gravity of a wheel-legged robot under a complex motion state cannot meet the demand. SUMMARY

[0004] Embodiments of the application provide a motion state method and device, an electronic device and a storage medium, which can solve the problem that the control effect of the balance of a center of gravity of a wheel-legged robot under a complex motion state cannot meet the demand.

[0005] In a first aspect, embodiments of the application provide a control method of a motion state, applied to a wheel-legged robot, the wheel-legged robot comprising a body, a leg part and a wheel, and the control method comprising:

[0006] obtaining a total torque of the wheel, a linear velocity parameter, a pitch angle parameter, a yaw angle parameter, a weight coefficient, a co-state variable, an intermediate parameter, a characteristic parameter, a preset proportional gain coefficient, a preset leg part height, a left-right leg target height difference, a wheel angular velocity estimation value, a discrete time period and a cost function, wherein the linear velocity parameter comprises a target tracking linear velocity and a tracking linear velocity, the pitch angle parameter comprises a target pitch angle, a pitch angle and a pitch angle velocity, and the yaw angle parameter comprises a target yaw angle velocity and a yaw angle velocity;

[0007] constructing a dynamics model based on the total torque of the wheel, the tracking linear velocity, the pitch angle, the pitch angle velocity, the characteristic parameter, a gravity acceleration and the intermediate parameter, wherein the dynamics model represents a dynamics state of the wheel-legged robot in a balance motion, and the intermediate parameter represents an intermediate motion state of the wheel-legged robot based on the characteristic parameter;

[0008] obtaining the wheel target angular acceleration and value based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the co-state variable, the cost function and the dynamics model;

[0009] obtain a wheel target angular acceleration difference value based on the preset proportional gain coefficient and the yaw angle parameter;

[0010] obtain a left wheel target angular velocity and a right wheel target angular velocity based on the wheel target angular acceleration sum value, the wheel target angular acceleration difference value, the discrete time period and the wheel angular velocity estimation value;

[0011] obtain a target joint angle based on the preset leg height and the left-right leg target height difference value;

[0012] control the wheel-legged robot to complete a preset motion task based on the target joint angle, the left wheel target angular velocity and the right wheel target angular velocity.

[0013] In a second aspect, an embodiment of the present application provides a motion state control device, which comprises:

[0014] an acquisition module, configured to acquire a wheel total torque, a linear velocity parameter, a pitch angle parameter, a yaw angle parameter, a weight coefficient, a co-state variable, an intermediate parameter, a characteristic parameter, a preset proportional gain coefficient, a preset leg height, a left-right leg target height difference value, a wheel angular velocity estimation value, a discrete time period and a cost function, wherein the linear velocity parameter comprises a target tracking linear velocity and a tracking linear velocity, the pitch angle parameter comprises a target pitch angle, a pitch angle and a pitch angle velocity, and the yaw angle parameter comprises a target yaw angle velocity and a yaw angle velocity;

[0015] a construction module, configured to construct a dynamics model based on the wheel total torque, the tracking linear velocity, the pitch angle, the pitch angle velocity, the characteristic parameter, a gravitational acceleration and the intermediate parameter, wherein the dynamics model represents a dynamics state of the wheel-legged robot in a balance motion, and the intermediate parameter represents an intermediate motion state of the wheel-legged robot based on the characteristic parameter;

[0016] a first obtaining module, configured to obtain a wheel target angular acceleration sum value based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the co-state variable, the cost function and the dynamics model;

[0017] a second obtaining module, configured to obtain a wheel target angular acceleration difference value based on the preset proportional gain coefficient and the yaw angle parameter;

[0018] a third obtaining module, configured to obtain a left wheel target angular velocity and a right wheel target angular velocity based on the wheel target angular acceleration sum value, the wheel target angular acceleration difference value, a discrete time period and the wheel angular velocity estimation value;

[0019] a fourth obtaining module, configured to obtain a target joint angle based on the preset leg height and the left-right leg target height difference;

[0020] a control module, configured to control the wheel-legged robot to complete a preset motion task based on the target joint angle, the left wheel target angular velocity and the right wheel target angular velocity.

[0021] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the computer program, the method according to any one of the first aspect is implemented.

[0022] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the method according to any one of the first aspect is implemented.

[0023] In a fifth aspect, a computer program product is provided. When the computer program product is run on a terminal device, the terminal device executes the method according to any one of the first aspect.

[0024] It can be understood that the beneficial effects of the second aspect to the fifth aspect can be referred to the related description of the first aspect, which will not be repeated here.

[0025] Compared with the prior art, the embodiments of the present application have at least the following beneficial effects:

[0026] The motion state control method of the application is applied to a wheel-legged robot, the wheel-legged robot comprising a body, a leg part and a wheel, the method comprising the following steps: obtaining a total torque of the wheel, a linear velocity parameter, a pitch angle parameter, a yaw angle parameter, a weight coefficient, a coordination variable, an intermediate parameter, a characteristic parameter, a preset proportional gain coefficient, a preset leg height, a left-right leg target height difference, a wheel angular velocity estimation value, a discrete time period and a cost function, wherein the linear velocity parameter comprises a target tracking linear velocity and a tracking linear velocity, the pitch angle parameter comprises a target pitch angle, a pitch angle and a pitch angle velocity, and the yaw angle parameter comprises a target yaw angle velocity and a yaw angle velocity; constructing a dynamics model based on the total torque of the wheel, the tracking linear velocity, the pitch angle, the pitch angle velocity, the characteristic parameter, a gravitational acceleration and the intermediate parameter, wherein the dynamics model represents a dynamics state of the wheel-legged robot in a balance motion, and the intermediate parameter represents an intermediate motion state of the wheel-legged robot based on the characteristic parameter; obtaining a wheel target angular acceleration sum value based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the coordination variable, the cost function and the dynamics model; obtaining a wheel target angular acceleration difference value based on the preset proportional gain coefficient and the yaw angle parameter; obtaining a left wheel target angular velocity and a right wheel target angular velocity based on the wheel target angular acceleration sum value, the wheel target angular acceleration difference value, the discrete time period and the wheel angular velocity estimation value; obtaining a target joint angle based on the preset leg height and the left-right leg target height difference; and controlling the wheel-legged robot to complete a preset motion task based on the target joint angle, the left wheel target angular velocity and the right wheel target angular velocity, so as to improve the accuracy of the center of gravity balance control of the wheel-legged robot in a complex motion state, and the input of the target joint angle takes into account the compliant control of the wheel-legged robot, so that the wheel-legged robot can compliantly complete complex actions such as balance driving motion, center of gravity follow-up steering motion and height adjustment motion. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without any creative effort based on these drawings.

[0028] Figure 1 is a structural schematic diagram of a wheel-legged robot provided by an embodiment of the application;

[0029] Figure 2 is a flowchart of a motion state control method provided by an embodiment of the application;

[0030] Figure 3 FIG. 1 is a flowchart illustrating a process for obtaining a wheel angular velocity estimation value according to an embodiment of the present application;

[0031] Figure 4 FIG. 2 is a flowchart illustrating a process for obtaining a left-right leg target height difference according to an embodiment of the present application;

[0032] Figure 5 FIG. 3 is a flowchart illustrating a process for constructing a dynamic model based on a wheel total torque, a tracking linear velocity, a pitch angle, a pitch angular velocity, a characteristic parameter, a gravity acceleration, and an intermediate parameter according to an embodiment of the present application;

[0033] Figure 6 FIG. 4 is a flowchart illustrating a process for obtaining a wheel target angular acceleration and value based on a linear velocity parameter, a pitch angle parameter, a weight coefficient, a cost function, and a dynamic model according to an embodiment of the present application;

[0034] Figure 7 FIG. 5 is a flowchart illustrating a process for obtaining a left wheel target angular velocity and a right wheel target angular velocity based on a wheel target angular acceleration and value, a wheel target angular acceleration difference value, a discrete time period, and a wheel angular velocity estimation value according to an embodiment of the present application;

[0035] Figure 8 FIG. 6 is a flowchart illustrating a process for obtaining a target joint angle based on a preset leg height and a left-right leg target height difference value according to an embodiment of the present application;

[0036] Figure 9 FIG. 7 is a structural diagram of a motion state control device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] In the following description, for purposes of explanation and not limitation, specific details are set forth, such as a particular sequence of actions, in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, and circuits are omitted so as not to obscure the description of the present application with unnecessary detail.

[0038] Artificial intelligence (AI) is the theory, method, technology and application system of using digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive environment, acquire knowledge and use knowledge to obtain optimal results. That is, artificial intelligence is a comprehensive technology of computer science, which tries to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence studies the design principles and implementation methods of various intelligent machines, so that the machines have the functions of perception, reasoning and decision making.

[0039] The present application mainly relates to the robot technology in the artificial intelligence technology, and mainly relates to robot intelligent control. The robot is a kind of mechanical electronic equipment that can imitate the skill of human being by using mechanical transmission and modern microelectronic technology, and the robot is developed on the basis of electronics, machinery and information technology. The appearance of the robot does not necessarily have to be like a person, as long as it can autonomously complete the task and command given by human being, it belongs to the members of the robot family. The robot is an automatic machine, and the machine has some intelligent capabilities similar to human beings or living beings, such as perception ability, planning ability, action ability and cooperation ability, and is a kind of automatic machine with high flexibility. With the development of computer technology and artificial intelligence technology, the robot has been greatly improved in function and technical level, for example, mobile robot and robot vision and touch.

[0040] Wheel-legged robot: the wheel-legged robot is a kind of robot structure that controls the motion of the robot body through the wheel-leg structure, which combines the advantages of wheeled robot and legged robot, has the high efficiency of wheeled robot, and inherits the strong terrain adaptability of legged robot, and can overcome uneven terrain and obstacles. Since the contact points of the wheel-legged robot with the ground only include the contact points of the wheels with the ground, the wheel-legged robot needs to be balanced in various motions.

[0041] In the embodiment, as shown in Figure 1 The wheel-legged robot includes a body, a leg and a wheel, the leg includes a left leg and a right leg, both ends of the left leg and the right leg are movably connected to the body and the wheel respectively, the left leg and the right leg each include a first leg and a second leg movably connected, the first leg is also called thigh, and the second leg is also called calf, the first leg is driven by a first driving motor and can rotate around a hip joint, the second leg is driven by a second driving motor and can rotate around a knee joint, the leg further includes a first connecting rod and a second connecting rod movably connected, one end of the first connecting rod is connected to the second driving motor, one end of the second connecting rod is connected to the second component, and the wheel is driven by a built-in third driving motor, which is a hub driving motor. It can be understood that in the present application, the specific structure of the wheel-legged robot is not limited, and the wheel-legged robot is not limited to the above structure, and the specific structure of the wheel-legged robot is determined according to the requirements of the application scene. The wheel-legged robot should be understood as a robot containing a wheeled structure.

[0042] Wheel-legged robots control the motion state of the robot body through the wheel-legged structure. Since the wheel-legged robot is an unstable under-actuated system, there is a problem of balance control. At present, the control method of the motion state of the wheel-legged robot mainly includes a model-based control method and a non-model-based control method. In the model-based control method, a car-inverted pendulum model is constructed, and model predictive control and nonlinear control are adopted. In the non-model-based control method, a proportion integral differential (PID) control method is used, and wheel torque or thrust is used as a control amount of balance and speed.

[0043] In the prior art, when the model-based control is used, the car-inverted pendulum model cannot accurately describe the influence of the wheel rotational inertia on the robot pitch. The non-model-based PID control parameter setting has high complexity, and the effect of the center of gravity balance control cannot meet the application requirements of various scenes. When the wheel torque or thrust is used as the control amount, since the data of the wheel friction and rotational inertia are difficult to obtain, the dynamics model from the wheel torque or thrust to the pitch angle is not accurate, so that the balance control effect cannot meet the application requirements of various scenes.

[0044] Therefore, the prior art has a problem that the effect of the center of gravity balance control of the wheel-legged robot under a complex motion state cannot meet the requirements.

[0045] The motion state control method of the application is applied to a wheel-leg robot. The wheel total torque, linear velocity parameter, pitch angle parameter, yaw angle parameter, weight coefficient, cooperative state variable, intermediate parameter, characteristic parameter, preset proportional gain coefficient, preset leg height, left-right leg target height difference, wheel angular velocity estimation value, discrete time period and cost function are obtained. The linear velocity parameter includes target tracking linear velocity and tracking linear velocity. The pitch angle parameter includes target pitch angle, pitch angle and pitch angle velocity. The yaw angle parameter includes target yaw angle velocity and yaw angle velocity. A dynamics model is constructed based on the wheel total torque, tracking linear velocity, pitch angle, pitch angle velocity, characteristic parameter, gravitational acceleration and intermediate parameter. The dynamics model represents the dynamics state of the wheel-leg robot in balanced motion, and the intermediate parameter represents the intermediate motion state of the wheel-leg robot based on the characteristic parameter. The wheel target angular acceleration sum and wheel target angular acceleration difference are obtained based on the linear velocity parameter, pitch angle parameter, weight coefficient, cooperative state variable, cost function and dynamics model. The left wheel target angular velocity and right wheel target angular velocity are obtained based on the preset proportional gain coefficient and yaw angle parameter, wheel target angular acceleration sum, wheel target angular acceleration difference, discrete time period and wheel angular velocity estimation value. The target joint angle is obtained based on the preset leg height and left-right leg target height difference. The wheel-leg robot completes the preset motion task based on the target joint angle, left wheel target angular velocity and right wheel target angular velocity. Since the dynamics model with the wheel target angular acceleration as the control variable is used, and the left wheel target angular velocity, right wheel target angular velocity and target joint angle related to the leg height are used as the input of the wheel-leg robot controller, the accuracy of the center of gravity balance control of the wheel-leg robot in complex motion states is improved. At the same time, the input of the target joint angle takes into account the compliant control of the wheel-leg robot, which can control the wheel-leg robot to compliantly complete various preset motion tasks, such as balanced driving motion, center of gravity follow-up steering motion and height adjustment motion.

[0046] The technical solutions of the application will be described below through specific embodiments.

[0047] In a first aspect, as shown in the drawings, the embodiment provides a motion state control method applied to a wheel-leg robot. The control method comprises the following steps. Figure 2

[0048] S100, obtaining wheel total torque, linear velocity parameter, pitch angle parameter, yaw angle parameter, weight coefficient, cooperative state variable, intermediate parameter, characteristic parameter, preset proportional gain coefficient, preset leg height, left-right leg target height difference, wheel angular velocity estimation value, discrete time period and cost function.

[0049] ​In one embodiment, the linear velocity parameters include a target tracking linear velocity, a tracking linear velocity, the pitch angle parameters include a target pitch angle, a pitch angle, a pitch angle velocity, a pitch angle velocity, and the yaw angle parameters include a target yaw angle velocity, a yaw angle velocity.

[0050] In one embodiment, the wheel torque includes a left wheel torque and a right wheel torque, and the total wheel torque is a sum of the left wheel torque and the right wheel torque.

[0051] In one embodiment, various motion parameters of the wheel-legged robot are obtained through various sensors, preset or real-time data processing, facilitating control of the motion state of the wheel-legged robot.

[0052] In one embodiment, as shown in Figure 3 the wheel angular velocity estimation value is obtained, including:

[0053] S110, obtaining a wheel angular velocity at a previous time, a wheel torque at a current time, a wheel pitch moment of inertia, process noise, measurement noise, a wheel angular velocity prior estimation value, and a gain parameter.

[0054] In one embodiment, the wheel angular velocity includes a left wheel angular velocity and a right wheel angular velocity.

[0055] S111, based on the wheel angular velocity at the previous time, a discrete time period, the wheel torque at the current time, the wheel pitch moment of inertia, the process noise, and the measurement noise, a first wheel angular velocity system equation at the current time is constructed.

[0056] In one embodiment, the first wheel angular velocity system equation is:

[0057]

[0058]

[0059] wherein, is a wheel angular velocity intermediate variable at the current time k, k is a positive integer;

[0060] is a wheel angular velocity at a previous time k-1; is a wheel angular velocity at the current time k;

[0061] Δt is a discrete time period; T w,k is a wheel torque at the current time k;

[0062] J w,pch is a wheel pitch moment of inertia; d k is process noise; e k is measurement noise.

[0063] It should be noted that the first wheel angular velocity system equation represents the angular velocity of the left or right wheel of the wheel-legged robot. If it represents the left wheel angular velocity, the intermediate variables of the wheel angular velocity in the first wheel angular velocity system equation are the left wheel angular velocity variable, the wheel angular velocity at the previous time k-1 is the left wheel angular velocity at the previous time k-1, the wheel angular velocity at the current time k is the left wheel angular velocity at the current time k, the wheel torque at the current time k is the left wheel torque at the current time k, and the wheel pitch inertia is the left wheel pitch inertia. Correspondingly, if it represents the right wheel angular velocity, the relevant variables are all parameters corresponding to the right wheel.

[0064] S112, process the first wheel angular velocity system equation to obtain the second wheel angular velocity system equation.

[0065] In one embodiment, the system equation for the first wheel angular velocity is replaced to obtain the system equation for the second wheel angular velocity, which simplifies the system equation and reduces its complexity. Specifically, the replacement and rearrangement method involves using x... k replace Use y k replace Replace with coefficient σ2 Increase coefficients σ1 and σ3.

[0066] In one embodiment, the system equation for the second wheel angular velocity is:

[0067] x k =σ1×x k-1 +σ2×T w,k +d k

[0068] y k =σ3×x k +e k

[0069] Where, x k Let k be the intermediate variable representing the wheel angular velocity at the current time k, where k is a positive integer;

[0070] x k-1 y is the wheel angular velocity at the previous moment k-1; k Let k be the wheel angular velocity at the current time.

[0071] T w,k σ1 is the wheel torque; σ2 is the first coefficient; σ2 is the second coefficient.

[0072] σ3 is the third coefficient.

[0073] In one embodiment, σ1 = 1, σ3 = 1.

[0074] It should be noted that the second wheel angular velocity system equation represents the left wheel angular velocity or the right wheel angular velocity of the wheel-legged robot. If the left wheel angular velocity is represented, the related variables are the parameters corresponding to the left wheel. If the right wheel angular velocity is represented, the related variables are the parameters corresponding to the right wheel.

[0075] In S113, the wheel angular velocity is obtained based on the second wheel angular velocity system equation.

[0076] In S114, the wheel angular velocity estimate is obtained based on the wheel angular velocity, the wheel angular velocity prior estimate, and the gain parameter.

[0077] In one embodiment, the wheel angular velocity estimate includes a left wheel angular velocity estimate and a right wheel angular velocity estimate.

[0078] In one embodiment, the wheel angular velocity prior estimate includes a left wheel angular velocity prior estimate and a right wheel angular velocity prior estimate.

[0079] In one embodiment, a Kalman filtering method is used to establish a filtering equation of the wheel angular velocity estimate, which is used to smooth the noise of the wheel encoder angular velocity and obtain a more accurate wheel angular velocity estimate.

[0080] In one embodiment, the filtering equation of the wheel angular velocity estimate is as follows:

[0081]

[0082] wherein, is the wheel angular velocity estimate at the current time k, and k is a positive integer;

[0083] is the wheel angular velocity prior estimate at the current time k;

[0084] K k is the gain parameter at the current time k; y k is the wheel angular velocity at the current time k;

[0085] σ3 is a third coefficient.

[0086] In one embodiment, the calculation formula of the wheel angular velocity prior estimate is as follows:

[0087]

[0088] wherein, is the wheel angular velocity prior estimate at the current time k;

[0089] σ1 is a first coefficient; σ2 is a second coefficient,

[0090] is the wheel angular velocity estimate value at the previous time k-1; T w,k is the wheel torque at the current time k;

[0091] In one embodiment, the gain parameter is calculated by the following formula:

[0092]

[0093] wherein K k is the gain parameter at the current time k; and σ3 is the third coefficient;

[0094] Var(e k ) is the variance of the measurement noise e k . is the variance of the wheel angular velocity prior estimate value.

[0095] In one embodiment, the variance of the wheel angular velocity prior estimate value is calculated by the following formula:

[0096]

[0097] wherein, is the variance of the wheel angular velocity prior estimate value;

[0098] σ1 is the first coefficient;

[0099] is the variance of the wheel angular velocity prior estimate value at the previous time k-1;

[0100] Var(d k ) is the variance of the process noise d k .

[0101] In one embodiment, the conversion calculation formula of the variance of the wheel angular velocity estimate value at the current time k and the variance of the wheel angular velocity prior estimate value is as follows:

[0102]

[0103] wherein, is the variance of the wheel angular velocity estimate value;

[0104] is the variance of the wheel angular velocity prior estimate value;

[0105] K k is the gain parameter at the current time k.

[0106] It should be noted that the filtering equation of the wheel angular velocity estimation value, the calculation formula of the wheel angular velocity priori estimation value, the calculation formula of the gain parameter, the variance calculation formula of the wheel angular velocity priori estimation value, and the conversion calculation formula are used to obtain the left wheel angular velocity estimation value or the right wheel angular velocity estimation value, if each calculation formula is used to obtain the left wheel angular velocity estimation value, each variable related to the calculation formula is the parameter corresponding to the left wheel, and if each calculation formula is used to obtain the right wheel angular velocity estimation value, each variable related to the calculation formula is the parameter corresponding to the right wheel.

[0107] In one embodiment, based on the above calculation formula and given input parameters, the left wheel angular velocity estimation value and the right wheel angular velocity estimation value

[0108] In one embodiment, the characteristic parameters further include a wheel spacing s w , and a body mass M b .

[0109] In one embodiment, as shown in Figure 4 , the left and right leg target height difference is obtained, including:

[0110] S120, based on the yaw angular velocity, the left wheel angular velocity estimation value, the right wheel angular velocity estimation value, the wheel spacing, the gravitational acceleration, and the body mass, obtaining the body roll angle.

[0111] In one embodiment, based on the yaw angular velocity, the left wheel angular velocity estimation value, the right wheel angular velocity estimation value, the wheel spacing, the gravitational acceleration, and the body mass, the body roll angle is obtained by the body roll angle calculation formula, which is beneficial to control the complex rolling motion or the balance motion of the complex terrain.

[0112] In one embodiment, the body roll angle calculation formula is:

[0113]

[0114] wherein, θ roll is the body roll angle; is the yaw angular velocity;

[0115] s w is the wheel spacing, i.e. the spacing between the left wheel and the right wheel; is the left wheel angular velocity estimation value;

[0116] is the right wheel angular velocity estimation value; M b is the body mass; and g is the gravitational acceleration.

[0117] In one embodiment, the yaw rate is obtained by using a yaw rate calculation formula based on the wheel angular velocity estimate, wheel radius, and wheel spacing.

[0118] In one embodiment, the formula for calculating yaw rate is:

[0119]

[0120] in, Yaw angular velocity; This is an estimated value for the angular velocity of the left wheel;

[0121] This is the estimated angular velocity of the right wheel; r w s is the radius of the wheel; w This refers to the wheel spacing.

[0122] S121, based on the aircraft's roll angle and wheel spacing, obtains the height difference between the left and right leg targets.

[0123] In one embodiment, the target height difference between the left and right legs is obtained by calculating the target height difference between the left and right legs based on the body roll angle and wheel spacing. This is beneficial for controlling the balance of the wheeled robot during movement based on the target height difference between the left and right legs.

[0124] In one embodiment, the formula for calculating the height difference between the left and right legs is:

[0125]

[0126] in, The target height for the left leg; The target height for the right leg;

[0127] θ roll s is the roll angle of the aircraft. w This refers to the wheel spacing.

[0128] In one embodiment, the tracking linear velocity is obtained by using a linear velocity calculation formula based on the wheel angular velocity estimate, wheel radius, pitch angle, pitch angular velocity, and straight distance.

[0129] In one embodiment, the linear velocity is calculated as follows:

[0130]

[0131] Among them, v c To track linear velocity; This is an estimated value for the angular velocity of the left wheel;

[0132] This is the estimated angular velocity of the right wheel; r w The radius of the wheel;

[0133] θpch The pitch angle; It is the pitch angular velocity;

[0134] l represents the straight-line distance, which is half the distance between the aircraft's center of gravity and the center of the wheel axle.

[0135] In one embodiment, the joint angle is obtained through experimental calibration. Target pitch angle relative to equilibrium position The functional relationship, and then based on the joint angle The target pitch angle at the equilibrium position is obtained by looking up the table.

[0136] In one embodiment, setting Adjust the target pitch angle at the equilibrium position The value is until the actual tracking linear velocity v. c =0, at this point the equilibrium pitch angle is the theoretical equilibrium pitch angle under the current joint angle. (The joint angle is then set to...) The workspace is discretized into several working points. A balancing experiment is performed at each working point to obtain the equilibrium position pitch angle. The functional relationship between the joint angle and the equilibrium position pitch angle is obtained through polynomial fitting.

[0137] S200, based on total wheel torque, tracking linear velocity, pitch angle, pitch angular velocity, characteristic parameters, gravitational acceleration and intermediate parameters, constructs a dynamic model.

[0138] In one embodiment, the dynamic model characterizes the dynamic state of the wheel-legged robot in balancing motion, and the intermediate parameters characterize the intermediate motion state of the wheel-legged robot based on the characteristic parameters.

[0139] In one embodiment, a dynamic model is constructed based on the total wheel torque, tracking linear velocity, pitch angle, pitch angular velocity, characteristic parameters, gravitational acceleration, and intermediate parameters, with the target angular acceleration of the wheel as the control variable. This is beneficial for controlling the balance and movement of the wheel-legged robot when performing complex actions through the dynamic model.

[0140] In one embodiment, the intermediate parameters include a first intermediate parameter, a second intermediate parameter, a third intermediate parameter, and a fourth intermediate parameter; the dynamic model includes a first dynamic model and a second dynamic model; and the characteristic parameter includes the wheel radius.

[0141] In one embodiment, such as Figure 5 As shown, a dynamic model is constructed based on the total wheel torque, tracking linear velocity, pitch angle, pitch angular velocity, characteristic parameters, gravitational acceleration, and intermediate parameters, including:

[0142] S210, obtain a state matrix based on the tracking linear velocity, the pitch angle, and the pitch angle velocity.

[0143] In one embodiment, the state matrix is The state matrix is a transpose matrix of a one-dimensional matrix including three variables of tracking linear velocity v c , pitch angle θ pch , and pitch angle velocity The state matrix represents the dynamic state of the wheel-legged robot.

[0144] S220, construct a first dynamic model based on the state matrix, the total wheel torque, the first intermediate parameter, the second intermediate parameter, the third intermediate parameter, the fourth intermediate parameter, the gravitational acceleration, and the wheel radius.

[0145] In one embodiment, the first dynamic model represents the dynamic state of the wheel-legged robot in balanced motion controlled by the wheel torque.

[0146] In one embodiment, the first dynamic model is:

[0147]

[0148]

[0149]

[0150] wherein, is a first-order derivative of , and is a first variable in the state matrix , i.e.

[0151] is a first-order derivative of , and is a second variable in the state matrix , i.e.

[0152] is a first-order derivative of , and is a third variable in the state matrix , i.e.

[0153] a1 is the first intermediate parameter; a2 is the second intermediate parameter; a3 is the third intermediate parameter;

[0154] a4 is the fourth intermediate parameter; g is the gravitational acceleration;

[0155] u is the total wheel torque, u=Tlw +T rw ;r w The radius is the wheel radius.

[0156] In one embodiment, the formula for calculating the first intermediate parameter is:

[0157]

[0158] Where a1 is the first intermediate parameter; M w Mass of a single wheel;

[0159] M p For the mass of one leg; M b For the body mass;

[0160] J w,pch r is the pitch inertia of the wheel. w The radius is the wheel radius.

[0161] In one embodiment, the second intermediate parameter is calculated as follows:

[0162] a2=2(M p + b )×l×cosθ pch

[0163] Where a2 is the second intermediate parameter; M p For the mass of one leg;

[0164] M b θ is the mass of the machine body; l is the straight-line distance; θ pch The pitch angle.

[0165] In one embodiment, the third intermediate parameter is calculated as follows:

[0166] a3=2(M p + b )×l×sinθ pch

[0167] Where a3 is the third intermediate parameter; M p For the mass of one leg;

[0168] M b θ is the mass of the machine body; l is the straight-line distance; θ pch The pitch angle.

[0169] In one embodiment, the fourth intermediate parameter is calculated as follows:

[0170] a4=2×M p ×l 2 +4×M b ×l 2 +2×Jp,pch + b,pch

[0171] wherein a4 is a fourth intermediate parameter; M p is a single-sided leg mass;

[0172] M b is a body mass; and l is a straight-line distance;

[0173] J p,pch is a leg pitch rotational inertia; J b,pch is a body pitch rotational inertia.

[0174] S230, transforming the first dynamic model to obtain a second dynamic model.

[0175] In one embodiment, the second dynamic model characterizes a dynamic state of the wheel-legged robot in balanced motion controlled by a wheel target angular acceleration.

[0176] In one embodiment, the first dynamic model is feedback linearized to obtain the second dynamic model, which facilitates optimizing the balanced motion of the wheel-legged robot controlled by a wheel torque into the balanced motion of the wheel-legged robot controlled by a wheel target angular acceleration. Since controlling the wheel target angular acceleration is more effective than controlling the wheel torque, the control effect of the balanced motion of the wheel-legged robot controlled by the wheel target angular acceleration is better, and the control accuracy of the balanced motion of the wheel-legged robot is improved.

[0177] In one embodiment, the wheel target angular acceleration and value is set as The transformation calculation formula of the wheel target angular acceleration and value is:

[0178]

[0179] The second dynamic model is obtained by transforming the first dynamic model, and the second dynamic model is:

[0180]

[0181]

[0182]

[0183] wherein, is a first-order derivative of , a first-order derivative of is a state matrix the first variable in the state matrix tracks a linear velocity, i.e.

[0184] is a first-order derivative of , a first-order derivative of is a state matrix is a third variable pitch angle velocity, i.e.

[0185] is a first-order derivative of is a state matrix is a third variable pitch angle velocity, i.e.

[0186] a1 is a first intermediate parameter; a2 is a second intermediate parameter;

[0187] a3 is a third intermediate parameter; a4 is a fourth intermediate parameter;

[0188] is a wheel target angular acceleration and value, is a left wheel target angular acceleration, is a right wheel target angular acceleration;

[0189] g is a gravity acceleration; r w is a wheel radius.

[0190] S300, based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the co-state variable, the cost function and the dynamics model, obtaining the wheel target angular acceleration and value.

[0191] In one embodiment, the weight coefficient includes a first weight coefficient, a second weight coefficient and a third weight coefficient, and the co-state variable includes a first co-state variable, a second co-state variable and a third co-state variable.

[0192] In one embodiment, based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the co-state variable, the cost function and the dynamics model, obtaining the wheel target angular acceleration and value, is conducive to controlling the wheel-legged robot through the wheel target angular acceleration and value.

[0193] In one embodiment, as shown in Figure 6 based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the cost function and the dynamics model, obtaining the wheel target angular acceleration and value, includes:

[0194] S310, based on the first weight coefficient, the second weight coefficient, the third weight coefficient, the target tracking linear velocity, the tracking linear velocity, the target pitch angle, the pitch angle, the pitch angle velocity and the cost function, constructing an optimization problem description equation with the dynamics model as a constraint condition.

[0195] In one embodiment, constructing an optimization problem description equation with the dynamics model as a constraint condition is to convert the control method of the wheel-legged robot into an optimization method, which is conducive to reducing the complexity of control. ​

[0196] In one embodiment, the optimization problem description equation is as follows:

[0197]

[0198] The constraint condition is a second dynamic model, and

[0199] Wherein, O is a cost function; is a wheel target angular acceleration and value at time t;

[0200] τ1 is a first weight coefficient; τ2 is a second weight coefficient; τ3 is a third weight coefficient;

[0201] is a target tracking linear velocity; v c is a tracking linear velocity; θ pch is a pitch angle;

[0202] is a target pitch angle; is a pitch angular velocity; t0 is a prediction time domain initial time;

[0203] t f is a prediction time domain termination time; is a state matrix vector at a prediction time domain initial time t0;

[0204] is an initial vector of a state matrix ; U is an allowable set of wheel target angular acceleration and value.

[0205] It should be noted that the target tracking linear velocity is pre-set, and the target tracking linear velocity is set with a specific value according to different application scene environment requirements.

[0206] S320, based on the first costate variable, the second costate variable, the third costate variable, and the optimization problem description equation, a Hamilton function equation is constructed.

[0207] In one embodiment, the Hamilton function equation H is constructed by using the Pontryagin extremum principle, and is used for solving the optimization problem description equation.

[0208] In one embodiment, the calculation formula of the Hamilton function equation is as follows:

[0209] H = L + λ1 × f1 + 2 × f2 + 3 × f3

[0210] Wherein, H is a Hamilton function value;

[0211] L is a first reference coefficient,

[0212] f1 is a second reference coefficient,

[0213] f2 is a second reference coefficient,

[0214] f3 is a second reference coefficient, and a calculation formula of f3 is:

[0215]

[0216] λ1 is a first costate variable; λ2 is a second costate variable; and λ3 is a third costate variable.

[0217] S330, obtaining an optimal control law equation of a wheel target angular acceleration and value based on a Hamilton function equation.

[0218] In one embodiment, the optimal control law equation of the wheel target angular acceleration and value is obtained based on the Hamilton function equation, so as to obtain an optimized wheel target angular acceleration and value, thereby improving the control effect of the wheel-legged robot.

[0219] In one embodiment, a calculation formula of the optimal control law equation is:

[0220]

[0221] wherein, is an optimal wheel target angular acceleration and value; and H() is a Hamilton function;

[0222] is an optimal state matrix vector at time t λ * is an optimal costate variable at time t;

[0223] is a wheel target angular acceleration and value at time t.

[0224] In one embodiment, a constraint condition of the optimal control law equation is:

[0225]

[0226]

[0227] and λ(t f ) = 0

[0228] wherein, is a first-order derivative of a state matrix vector at time t;

[0229] is a first-order derivative of a costate variable at time t;

[0230] H m is the Hamiltonian function H for the state matrix vector ;

[0231] is the state matrix vector at time t is the wheel target angular acceleration and value at time t.

[0232] S340, based on the optimal control law equation, iterative calculation is performed until the preset condition is met, and the wheel target angular acceleration and value is obtained.

[0233] In one embodiment, based on the optimal control law equation, iterative calculation is performed until the preset condition is met, and the optimized wheel target angular acceleration and value is obtained The preset condition is that the iteration number n is less than or equal to N p -1, wherein N p is the number of discrete time periods from the initial time t0 of the prediction horizon to the terminal time t f of the prediction horizon.

[0234] In one embodiment, a mapping relationship of the co-state variable from the initial value λ(t0) at t0 to the terminal value λ(t f ) at t f is established along the optimal control law equation, and starting from , along the optimal control law trajectory, the terminal value satisfies the transversality condition λ(t f ) = 0, and each optimization variable corresponding Hamiltonian function value H is calculated, and the wheel angular acceleration is selected in the allowable set according to the optimal control law equation to make the Hamiltonian function value obtain the minimum value H * corresponding to the optimal wheel target angular acceleration and value , so as to obtain the wheel target angular acceleration and value

[0235] S400, based on the preset proportional gain coefficient and the yaw angle parameter, the wheel target angular acceleration difference value is obtained.

[0236] In one embodiment, based on the preset proportional gain coefficient and the yaw angle parameter, the wheel target angular acceleration difference value is obtained, including:

[0237] Based on the preset proportional gain coefficient and the yaw angle parameter, the wheel target angular acceleration difference value is obtained through a wheel target angular acceleration difference value calculation formula.

[0238] In one embodiment, the wheel target angular acceleration difference value calculation formula is:

[0239]

[0240] wherein, is a left wheel target angular acceleration; is a right wheel target angular acceleration;

[0241] K p is a preset proportional gain coefficient; is a target yaw angular velocity; is a yaw angular velocity.

[0242] It should be noted that the target yaw angular velocity is set in advance, and a specific value of the target yaw angular velocity is set according to different application scene environment requirements.

[0243] S500, obtaining a left wheel target angular velocity and a right wheel target angular velocity based on the wheel target angular acceleration and value, the wheel target angular acceleration difference value, the discrete time period, and the wheel angular velocity estimation value.

[0244] In one embodiment, the wheel angular velocity estimation value includes a left wheel angular velocity estimation value and a right wheel angular velocity estimation value.

[0245] In one embodiment, the left wheel target angular velocity and the right wheel target angular velocity are obtained based on the wheel target angular acceleration and value, the wheel target angular acceleration difference value, the discrete time period, and the wheel angular velocity estimation value, so as to control a motion state of the wheel-legged robot by using the left wheel target angular velocity and the right wheel target angular velocity.

[0246] In one embodiment, as shown in Figure 7 obtaining the left wheel target angular velocity and the right wheel target angular velocity based on the wheel target angular acceleration and value, the wheel target angular acceleration difference value, the discrete time period, and the wheel angular velocity estimation value includes:

[0247] S510, obtaining a left wheel target angular acceleration and a right wheel target angular acceleration based on the wheel target angular acceleration and value and the wheel target angular acceleration difference value.

[0248] S520, obtaining a left wheel target angular velocity based on the left wheel target angular acceleration, the discrete time period, and a left wheel angular velocity estimation value by using a target angular velocity calculation formula.

[0249] In one embodiment, the target angular velocity calculation formula includes a first target angular velocity calculation formula and a second target angular velocity calculation formula.

[0250] In one embodiment, the first target angular velocity calculation formula is:

[0251]

[0252] wherein, is the left wheel target angular velocity; is the left wheel angular velocity estimation value;

[0253] Δt is a discrete time interval; is a left wheel target angular acceleration.

[0254] S530, based on the right wheel target angular acceleration, the discrete time interval and the right wheel angular velocity estimate, the right wheel target angular velocity is obtained through a target angular velocity calculation formula.

[0255] In one embodiment, the second target angular velocity calculation formula is:

[0256]

[0257] wherein, is a right wheel target angular velocity; is a right wheel angular velocity estimate;

[0258] Δt is a discrete time interval; is a right wheel target angular acceleration.

[0259] S600, based on the preset leg height and the left-right leg target height difference, a target joint angle is obtained.

[0260] In one embodiment, based on the preset leg height and the left-right leg target height difference, the target joint angle is obtained, which is beneficial to control the center of gravity of the wheel-legged robot through the target joint angle to maintain balance, and the input of the target joint angle also takes into account the compliant control of the wheel-legged robot.

[0261] In one embodiment, the target joint angle includes a left leg target joint angle and a right leg target joint angle.

[0262] In one embodiment, as shown in Figure 8 based on the preset leg height and the left-right leg target height difference, the target joint angle is obtained, including:

[0263] S610, based on the preset leg height and the left-right leg target height difference, a left leg target height and a right leg target height are obtained through a target leg height calculation formula respectively.

[0264] In one embodiment, the target leg height calculation formula is:

[0265]

[0266]

[0267] wherein, is a left leg target height; is a right leg target height;

[0268] h ref is a preset leg height; is a left-right leg target height difference.

[0269] S620, performing leg inverse dynamics processing based on the left leg target height and the right leg target height to obtain a target joint angle.

[0270] In an embodiment, the target joint angle includes a hip joint target joint angle and a knee joint target joint angle.

[0271] S700, controlling the wheel-legged robot to complete a preset motion task based on the target joint angle, the left wheel target angular velocity, and the right wheel target angular velocity.

[0272] In an embodiment, the preset motion task includes a balance driving motion, a center of gravity follow-up steering motion, and a height adjustment motion.

[0273] In an embodiment, the wheel-legged robot is controlled to complete the preset motion task based on the target joint angle, the left wheel target angular velocity, and the right wheel target angular velocity, the left wheel target angular velocity, the right wheel target angular velocity, and the target joint angle are used as inputs of a wheel-legged robot controller, the accuracy of center of gravity balance control of the wheel-legged robot in a complex motion state is improved, and the input of the target joint angle takes into account compliant control of the wheel-legged robot, so that the wheel-legged robot can be controlled to compliantly complete various preset motion tasks.

[0274] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0275] Compared with the prior art, the embodiment has the following beneficial effects:

[0276] The motion state control method of the embodiment is applied to a wheel-legged robot. The wheel total torque, the linear velocity parameter, the pitch angle parameter, the yaw angle parameter, the weight coefficient, the coordination variable, the intermediate parameter, the characteristic parameter, the preset proportional gain coefficient, the preset leg height, the left-right leg target height difference, the wheel angular velocity estimation value, the discrete time period and the cost function are obtained. The linear velocity parameter includes the target tracking linear velocity and the tracking linear velocity. The pitch angle parameter includes the target pitch angle, the pitch angle and the pitch angle velocity. The yaw angle parameter includes the target yaw angle velocity and the yaw angle velocity. A dynamics model is constructed based on the wheel total torque, the tracking linear velocity, the pitch angle, the pitch angle velocity, the characteristic parameter, the gravity acceleration and the intermediate parameter. The dynamics model represents the dynamics state of the wheel-legged robot in balanced motion. The intermediate parameter represents the intermediate motion state of the wheel-legged robot based on the characteristic parameter. The wheel target angular acceleration and value is obtained based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the coordination variable, the cost function and the dynamics model. The wheel target angular acceleration difference is obtained based on the preset proportional gain coefficient and the yaw angle parameter. The left wheel target angular velocity and the right wheel target angular velocity are obtained based on the wheel target angular acceleration and value, the wheel target angular acceleration difference, the discrete time period and the wheel angular velocity estimation value. The target joint angle is obtained based on the preset leg height and the left-right leg target height difference. The wheel-legged robot is controlled to complete a preset motion task based on the target joint angle, the left wheel target angular velocity and the right wheel target angular velocity. The accuracy of the center of gravity balance control of the wheel-legged robot in a complex motion state is improved. The input of the target joint angle takes into account the compliant control of the wheel-legged robot, and the wheel-legged robot can compliantly complete various preset motion tasks, such as balanced driving motion, center of gravity follow-up steering motion and height adjustment motion.

[0277] In a second aspect, as shown in the figure, the embodiment of the present application provides a motion state control device, which comprises: Figure 9

[0278] The acquisition module 100 is configured to acquire the wheel total torque, the linear velocity parameter, the pitch angle parameter, the yaw angle parameter, the weight coefficient, the coordination variable, the intermediate parameter, the characteristic parameter, the preset proportional gain coefficient, the preset leg height, the left-right leg target height difference, the wheel angular velocity estimation value, the discrete time period and the cost function. The linear velocity parameter includes the target tracking linear velocity and the tracking linear velocity. The pitch angle parameter includes the target pitch angle, the pitch angle and the pitch angle velocity. The yaw angle parameter includes the target yaw angle velocity and the yaw angle velocity.

[0279] ​The construction module 200 is configured to construct a dynamic model based on the total torque of the wheels, the tracking linear velocity, the pitch angle, the pitch angle velocity, the characteristic parameter, the gravity acceleration, and an intermediate parameter, wherein the dynamic model represents a dynamic state of the balance movement of the wheel-legged robot, and the intermediate parameter represents an intermediate movement state of the wheel-legged robot based on the characteristic parameter.

[0280] The first obtaining module 300 is configured to obtain the wheel target angular acceleration sum value based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the co-state variable, the cost function, and the dynamic model.

[0281] The second obtaining module 400 is configured to obtain the wheel target angular acceleration difference value based on the preset proportional gain coefficient and the yaw angle parameter.

[0282] The third obtaining module 500 is configured to obtain the left wheel target angular velocity and the right wheel target angular velocity based on the wheel target angular acceleration sum value, the wheel target angular acceleration difference value, a discrete time period, and a wheel angular velocity estimation value.

[0283] The fourth obtaining module 600 is configured to obtain the target joint angle based on a preset leg height and a left-right leg target height difference value.

[0284] The control module 700 is configured to control the wheel-legged robot to complete the preset movement task based on the target joint angle, the left wheel target angular velocity, and the right wheel target angular velocity.

[0285] It should be noted that the information interaction and execution process between the above apparatus / module are based on the same concept as the method embodiments, and the specific functions and technical effects thereof can be referred to the method embodiments part, which will not be repeated here.

[0286] In a third aspect, the present embodiment provides a terminal device, comprising:

[0287] The memory, the processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the control method according to any one of the above first aspect.

[0288] In a fourth aspect, the present embodiment further provides a computer readable storage medium, comprising:

[0289] The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the control method according to any one of the above first aspect.

[0290] In a fifth aspect, the present embodiment provides a computer program product, which, when executed on a terminal device, causes the terminal device to perform the control method according to any one of the above first aspect.

[0291] It can be understood that the beneficial effects of the above-mentioned second aspect to the fifth aspect can be referred to the relevant description in the first aspect, and will not be repeated here.

[0292] The computer readable medium can at least include any entity or device capable of carrying the computer program code to the photographing device / terminal equipment, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium. For example, U disk, mobile hard disk, magnetic disk or optical disk and the like. In some jurisdictions, according to legislation and patent practice, the computer readable medium can not be an electrical carrier signal and a telecommunication signal.

[0293] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in a certain embodiment can be referred to the relevant description of other embodiments.

[0294] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A control method of a motion state, characterized by, The control method is applied to a wheel-legged robot including a body, a leg and a wheel, and comprises the following steps: obtaining a total torque of the wheel, a linear velocity parameter, a pitch angle parameter, a yaw angle parameter, a weight coefficient, a co-state variable, an intermediate parameter, a characteristic parameter, a preset proportional gain coefficient, a preset leg height, a left-right leg target height difference, a wheel angular velocity estimation value, a discrete time period and a cost function, wherein the linear velocity parameter comprises a target tracking linear velocity and a tracking linear velocity, the pitch angle parameter comprises a target pitch angle, a pitch angle and a pitch angle velocity, and the yaw angle parameter comprises a target yaw angle velocity and a yaw angle velocity; the characteristic parameter comprises a wheel spacing, a body mass and a wheel radius; constructing a dynamic model based on the total torque of the wheel, the tracking linear velocity, the pitch angle, the pitch angle velocity, the characteristic parameter, a gravitational acceleration and the intermediate parameter, wherein the dynamic model represents a dynamic state of the wheel-legged robot in a balanced motion, and the intermediate parameter represents an intermediate motion state of the wheel-legged robot based on the characteristic parameter; obtaining a wheel target angular acceleration sum value based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the co-state variable, the cost function and the dynamic model; obtaining a wheel target angular acceleration difference value based on the preset proportional gain coefficient and the yaw angle parameter; obtaining a left wheel target angular velocity and a right wheel target angular velocity based on the wheel target angular acceleration sum value, the wheel target angular acceleration difference value, the discrete time period and the wheel angular velocity estimation value; obtaining a target joint angle based on the preset leg height and the left-right leg target height difference; controlling the wheel-legged robot to complete a preset motion task based on the target joint angle, the left wheel target angular velocity and the right wheel target angular velocity; the weight coefficient comprises a first weight coefficient, a second weight coefficient and a third weight coefficient, and the co-state variable comprises a first co-state variable, a second co-state variable and a third co-state variable; the obtaining of the wheel target angular acceleration sum value based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the cost function and the dynamic model comprises: constructing an optimization problem description equation with the dynamic model as a constraint condition based on the first weight coefficient, the second weight coefficient, the third weight coefficient, the target tracking linear velocity, the tracking linear velocity, the target pitch angle, the pitch angle, the pitch angle velocity and the cost function; constructing a Hamilton function equation based on the first co-state variable, the second co-state variable, the third co-state variable and the optimization problem description equation; obtaining an optimal control law equation of the wheel target angular acceleration sum value based on the Hamilton function equation; iteratively calculating based on the optimal control law equation until a preset condition is met to obtain the wheel target angular acceleration sum value.

2. The method of claim 1, wherein the obtaining of the wheel angular velocity estimation value comprises: ​ obtaining a wheel angular velocity at a previous time, a wheel torque at a current time, a wheel pitch moment of inertia, a process noise, a measurement noise, a wheel angular velocity prior estimation value and a gain parameter; constructing a first wheel angular velocity system equation at the current time based on the wheel angular velocity at the previous time, the discrete time period, the wheel torque at the current time, the wheel pitch moment of inertia, the process noise and the measurement noise; processing the first wheel angular velocity system equation to obtain a second wheel angular velocity system equation; obtaining a wheel angular velocity based on the second wheel angular velocity system equation; obtaining the wheel angular velocity estimation value based on the wheel angular velocity, the wheel angular velocity prior estimation value and the gain parameter.

3. The method of claim 2, wherein, The wheel angular velocity estimation value includes a left wheel angular velocity estimation value and a right wheel angular velocity estimation value. The left wheel target angular velocity and the right wheel target angular velocity are obtained based on the wheel target angular acceleration sum value, the wheel target angular acceleration difference value, a discrete time period and the wheel angular velocity estimation value, including: The left wheel target angular velocity and the right wheel target angular velocity are obtained based on the wheel target angular acceleration sum value and the wheel target angular acceleration difference value. The left wheel target angular velocity is obtained by a target angular velocity calculation formula based on the left wheel target angular acceleration, the discrete time period and the left wheel angular velocity estimation value. The right wheel target angular velocity is obtained by a target angular velocity calculation formula based on the right wheel target angular acceleration, the discrete time period and the right wheel angular velocity estimation value.

4. The method of claim 1, wherein, The wheel angular velocity estimation value includes a left wheel angular velocity estimation value and a right wheel angular velocity estimation value. The left and right leg target height difference is obtained, including: The body roll angle is obtained based on the yaw angular velocity, the left wheel angular velocity estimation value, the right wheel angular velocity estimation value, the wheel distance, the gravity acceleration and the body mass. The left and right leg target height difference is obtained based on the body roll angle and the wheel distance.

5. The method of claim 1, wherein, The intermediate parameters include a first intermediate parameter, a second intermediate parameter, a third intermediate parameter and a fourth intermediate parameter, and the dynamic model includes a first dynamic model and a second dynamic model. The dynamic model is constructed based on the wheel total torque, the tracking linear velocity, the pitch angle, the pitch angular velocity, the characteristic parameter, the gravity acceleration and the intermediate parameters, including: A state matrix is obtained based on the tracking linear velocity, the pitch angle and the pitch angular velocity, and the state matrix represents a dynamic state of the wheel-legged robot; The first dynamic model is constructed based on the state matrix, the wheel total torque, the first intermediate parameter, the second intermediate parameter, the third intermediate parameter, the fourth intermediate parameter, the gravity acceleration and the wheel radius, and the first dynamic model represents a dynamic state of a balanced motion of the wheel-legged robot controlled by the wheel torque; The second dynamic model is obtained by transformation processing of the first dynamic model, and the second dynamic model represents a dynamic state of a balanced motion of the wheel-legged robot controlled by the wheel target angular acceleration.

6. The method of claim 1, wherein, The wheel target angular acceleration difference value is obtained based on the preset proportional gain coefficient and the yaw angle parameter, and the wheel target angular acceleration difference value is obtained by a wheel target angular acceleration difference value calculation formula. The control device comprises:

7. A motion state control device characterized by comprising: The acquisition module is configured to acquire a total wheel torque, a linear velocity parameter, a pitch angle parameter, a yaw angle parameter, a weight coefficient, a co-state variable, an intermediate parameter, a characteristic parameter, a preset proportional gain coefficient, a preset leg height, a left-right leg target height difference, a wheel angular velocity estimation value, a discrete time period, and a cost function, wherein the linear velocity parameter comprises a target tracking linear velocity and a tracking linear velocity, the pitch angle parameter comprises a target pitch angle, a pitch angle, and a pitch angle velocity, and the yaw angle parameter comprises a target yaw angle velocity and a yaw angle velocity; the characteristic parameter comprises a wheel spacing, a body mass, and a wheel radius; and the construction module is configured to construct a dynamics model based on the total wheel torque, the tracking linear velocity, the pitch angle, the pitch angle velocity, the characteristic parameter, a gravitational acceleration, and the intermediate parameter, wherein the dynamics model represents a dynamics state of a balance motion of the wheel-legged robot, and the intermediate parameter represents an intermediate motion state of the wheel-legged robot based on the characteristic parameter. The first obtaining module is configured to obtain a wheel target angular acceleration sum value based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the co-state variable, the cost function, and the dynamics model. The second obtaining module is configured to obtain a wheel target angular acceleration difference value based on the preset proportional gain coefficient and the yaw angle parameter. The third obtaining module is configured to obtain a left wheel target angular velocity and a right wheel target angular velocity based on the wheel target angular acceleration sum value, the wheel target angular acceleration difference value, the discrete time period, and the wheel angular velocity estimation value. The fourth obtaining module is configured to obtain a target joint angle based on the preset leg height and the left-right leg target height difference. The control module is configured to control the wheel-legged robot to complete a preset motion task based on the target joint angle, the left wheel target angular velocity, and the right wheel target angular velocity. The weight coefficient comprises a first weight coefficient, a second weight coefficient, and a third weight coefficient, and the co-state variable comprises a first co-state variable, a second co-state variable, and a third co-state variable. The wheel target angular acceleration sum value is obtained based on the linear velocity parameter, the pitch angle parameter, the weight coefficient, the cost function, and the dynamics model, and the wheel target angular acceleration sum value is obtained by constructing an optimization problem description equation with the dynamics model as a constraint condition based on the first weight coefficient, the second weight coefficient, the third weight coefficient, the target tracking linear velocity, the tracking linear velocity, the target pitch angle, the pitch angle, the pitch angle velocity, and the cost function. The Hamilton function equation is constructed based on the first co-state variable, the second co-state variable, the third co-state variable, and the optimization problem description equation. ​ ​ An optimal control law equation of the wheel target angular acceleration and value is obtained based on the Hamilton function equation; An iterative calculation is performed based on the optimal control law equation until a preset condition is met, and the wheel target angular acceleration and value is obtained.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor implements the method of any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: The computer program is executed by the processor to implement the method of any one of claims 1 to 6.

Images