A mobile robot formation control method, system, device, medium and product

Abstract

The present invention discloses a mobile robot formation control method, system, device, medium and product, belonging to the field of robot control technology, the method includes: obtaining the position, speed and acceleration of each robot in the robot formation; determining the converted speed of each robot considering the position constraint according to the position, speed and acceleration of each robot; determining the speed tracking error of each robot according to the converted speed of each robot; obtaining the tracking error of each robot according to the speed tracking error of the robot and the speed tracking error of its adjacent robot; determining the actual control torque of each robot according to the tracking error of each robot; controlling each robot according to the actual control torque of each robot. The optimal control of the robot formation is realized, solving the technical problem that the position constraint of the robot is not considered when controlling the current robot.

Description

Technical Field

[0001] The present invention relates to the technical field of wheeled mobile robot control, and in particular to a mobile robot formation control method, system, equipment, medium and product. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Wheeled mobile robots are a type of unmanned system with the advantages of low energy consumption, simple structure, and flexible operation. They are widely used in real life. As a typical representative of collaborative multi-robot systems, multi-robot systems have advantages over single robots, such as stable system performance and good self-organization. They can complete complex tasks that single robots cannot complete and improve work efficiency. Therefore, they are one of the objects of extensive research. Since wheeled mobile robot systems belong to a typical under-actuated nonlinear control system, they meet non-holonomic constraints, resulting in under-actuated characteristics, which brings certain difficulties to the control algorithm design and stability analysis of multi-wheeled mobile robots.

[0004] Currently, when optimally controlling a robot, the position error and speed error of the robot are calculated directly based on the position and speed of the robot, and then the tracking error is obtained, and the robot is controlled based on the tracking error; however, this control method does not consider the position constraint of the robot when controlling the robot, which may cause the robot to leave the safe area when it is finally controlled, and the optimal control of the robot cannot be achieved. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a mobile robot formation control method, system, device, medium and product, which realizes optimal control of the robot formation by considering the position constraints of the robots.

[0006] To achieve the above object, the present invention adopts the following technical solution:

[0007] In the first aspect, a mobile robot formation control method is proposed, comprising:

[0008] Get the position, velocity and acceleration of each robot in the robot formation;

[0009] Determine the transformed velocity of each robot considering the position constraint according to the position, velocity and acceleration of each robot;

[0010] Determine the speed tracking error of each robot according to the converted speed of each robot;

[0011] The tracking error of each robot is obtained according to the speed tracking error of the robot and the speed tracking errors of its adjacent robots;

[0012] According to the tracking error of each robot, the actual control torque of each robot is determined;

[0013] Each robot is controlled according to its actual control torque.

[0014] Furthermore, based on the position, velocity and acceleration of each robot, the converted velocity of each robot considering the position constraint is determined using the converted model, wherein the converted model is obtained by introducing variables representing the points at which the robot is most likely to leave the safe area into the robot model containing incomplete constraints.

[0015] Furthermore, the speed tracking error of the robot is equal to the error between the converted speed of the robot and the expected virtual speed of the robot, wherein the expected virtual speed of the robot is calculated according to the expected trajectory of the robot.

[0016] Furthermore, the deviation between the speed tracking error of the robot and the speed tracking error of its adjacent robot is calculated; the deviation is weightedly fused with the speed tracking error of the robot to obtain the tracking error of the robot.

[0017] Further, based on the tracking error of each robot, the optimal control torque of each robot when the robot cooperates with its adjacent robot is determined;

[0018] According to the optimal control torque of each robot, the actual control torque of each robot is determined.

[0019] Furthermore, according to the tracking error of each robot, an optimal performance function with the minimum performance index function value as the goal is determined, wherein when the performance index function value is the minimum, it indicates that the robot achieves coordination with its adjacent robots;

[0020] Solve the optimal performance function to obtain the optimal control torque of the robot;

[0021] According to the optimal control torque, an adaptive fuzzy logic system is used to determine the unknown items in the optimal control torque;

[0022] The unknown term in the optimal control torque and the optimal control torque are used to determine the actual control torque of the robot.

[0023] In the second aspect, a mobile robot formation control system is proposed, including:

[0024] An acquisition module is used to obtain the position, velocity and acceleration of each robot in the robot formation;

[0025] A data conversion module, for determining a converted speed of each robot considering position constraints according to the position, speed and acceleration of each robot;

[0026] The tracking error determination module is used to determine the speed tracking error of each robot according to the converted speed of each robot; and obtain the tracking error of each robot according to the speed tracking error of the robot and the speed tracking error of its adjacent robot;

[0027] A control torque determination module, used for determining the actual control torque of each robot according to the tracking error of each robot;

[0028] The control module is used to control each robot according to the actual control torque of each robot.

[0029] In a third aspect, a computer device is provided, the device comprising:

[0030] a processor adapted to execute a computer program;

[0031] A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a mobile robot formation control method proposed in the first aspect is implemented.

[0032] In a fourth aspect, a computer-readable storage medium is proposed, wherein the computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a mobile robot formation control method proposed in the first aspect.

[0033] In a fifth aspect, a computer program product is proposed, which includes a computer program. When the computer program is executed by a processor, it implements a mobile robot formation control method proposed in the first aspect.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] The present invention provides a mobile robot formation control method, system, equipment, medium and product. When controlling the robot formation, the method first converts the position, speed and acceleration of the robot into the robot's converted speed taking into account the position constraint, so that the robot's tracking error calculated according to the robot's converted speed contains both position information and speed information. When the tracking error is subsequently used to determine the robot's actual control torque and then control the robot, it is ensured that the robot moves within a safe area, thereby achieving optimal control of the robot formation.

[0036] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0038] Figure 1 A flowchart of a mobile robot formation control method disclosed in an embodiment;

[0039] Figure 2 A schematic diagram of a non-holonomic wheeled mobile robot model disclosed in an embodiment;

[0040] Figure 3 A schematic diagram of a motion trajectory simulation result disclosed in an embodiment;

[0041] Figure 4 Schematic diagram of speed error simulation results disclosed in the embodiment, wherein (a) is a schematic diagram of error simulation results between the linear speeds of four mobile robots and the expected linear speeds, and (b) is a schematic diagram of error simulation results between the angular speeds of four mobile robots and the expected angular speeds;

[0042] Figure 5 Schematic diagram of the simulation results of the fuzzy logic system observer parameters disclosed in the embodiment, wherein (a), (b), (c) and (d) are schematic diagrams of the simulation results of the fuzzy logic system observer parameters of four mobile robots respectively;

[0043] Figure 6 Schematic diagram of the simulation results of actor parameters disclosed in the embodiment, wherein (a), (b), (c) and (d) are schematic diagrams of the simulation results of actor parameters of four mobile robots respectively;

[0044] Figure 7 Schematic diagrams of critic parameter simulation results disclosed in the embodiments, wherein (a), (b), (c) and (d) are schematic diagrams of critic parameter simulation results of four mobile robots respectively. DETAILED DESCRIPTION

[0045] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0046] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0047] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0048] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0049] Example 1

[0050] Currently, when optimally controlling a robot, the position error and speed error of the robot are calculated directly based on the position and speed of the robot, and then the tracking error is obtained, and the robot is controlled based on the tracking error; however, this control method does not consider the position constraint of the robot when controlling the robot, which may cause the robot to leave the safe area when it is finally controlled, and the optimal control of the robot cannot be achieved.

[0051] In addition, optimal control is to find the control strategy that makes the robot system achieve the best performance through mathematical modeling and analysis of the robot system. One of the effective methods to obtain the optimal control algorithm is to solve the Hamilton-Jacobi-Bellman equation (HJB equation). However, due to the highly nonlinear and strongly coupled characteristics of this equation, it is almost impossible to find an analytical solution to this equation, especially for the optimal control of non-holonomic mobile robots.

[0052] In order to achieve optimal control of a three-wheeled mobile robot formation with two-wheel differential drive, in this embodiment, a mobile robot formation control method is disclosed, such as Figure 1-Figure 7 As shown, including:

[0053] S1: Get the position, velocity and acceleration of each robot in the formation.

[0054] S2: Determine the transformed velocity of each robot considering the position constraint according to the position, velocity and acceleration of each robot.

[0055] This embodiment determines the converted speed of each robot taking into account the position constraint based on the position, speed and acceleration of each robot using a converted model, wherein the converted model is obtained by introducing variables representing the points at which the robot is most likely to leave the safe area into a robot model containing incomplete constraints.

[0056] Among them, the robot model with non-holonomic constraints is:

[0057]

[0058] in represents the position of the robot, represents the speed of the robot, represents the symmetric positive definite inertia matrix, represents the Coriolis force and the centrifugal force, represents the gravity twisting the vector, is the input torque. The new variable introduced is Y i Indicates the point at which the robot is most likely to leave the safe area, which is determined based on the robot's position. ; By introducing variables Y i , can constrain the robot's moving area so that the robot can only move within the safe area, where x i , y i is the horizontal and vertical coordinates of the robot’s position information, [ l,d ] is the coordinate of point Y relative to the original position. For example, the center of gravity of the robot is the original position, and then moves forward l , move left d , get the new point Y position coordinates.

[0059] Let the first variable x 1i Substitute new variables Y i ,Right now x 1i Represents the position of the robot after conversion, the second variable x 2i Substitute the first-order derivative of the new variable, that is x 2i Represents the speed of the robot after conversion, and builds a robot model of the point that is most likely to leave the safe area. This model is the converted model.

[0060] This embodiment introduces a new variable Y i , and let the first variable x 1i Substitute new variables Y i , the second variable x 2i Substituting the first-order derivative of the new variable, the robot system model is linearized and the converted model is obtained as follows: ,in, is the point at which the robot is most likely to leave the safe area, , , .

[0061] Substitute the acquired position, velocity and acceleration of each robot into the converted model to calculate the converted velocity of the robot x 2i .

[0062] S3: Determine the speed tracking error of each robot according to the converted speed of each robot.

[0063] The speed tracking error of the robot in this embodiment is equal to the error between the converted speed of the robot and the expected virtual speed of the robot, wherein the expected virtual speed of the robot is calculated according to the expected trajectory of the robot.

[0064] This embodiment also determines the continuous function approximated by the converted model according to the universal approximation principle;

[0065] Based on the continuous function, the dynamic observation error of the new variable is determined.

[0066] According to the universal approximation theorem of fuzzy systems, we can get that for any continuous function, it can be expressed as follows:

[0067]

[0068] in, is the ideal FLS parameter vector, is the FLS basis function vector, is the approximation error. is a continuous function Setting an adaptive fuzzy state observer avoids the requirement to design the gain constant so that its characteristic polynomial is Hurwitz. The adaptive fuzzy state observer is set as: The dynamic observation error is: .in, , for The estimate, and is the observer gain constant. The desired virtual speed of this embodiment is determined according to the introduced new variables and the tracking errors of the set position and direction.

[0069] Among them, the tracking error of the set position and direction is: ; Give the desired virtual speed for tracking To assist robots To achieve the desired trajectory tracking: Calculate the speed tracking error between the first-order derivative of the new variable in the converted model and the set desired virtual speed ; .

[0070] S4: Obtain the tracking error of each robot based on the speed tracking error of the robot and the speed tracking errors of its adjacent robots. Specifically:

[0071] The deviation between the robot's speed tracking error and the speed tracking error of its adjacent robots is calculated; the deviation is weightedly fused with the robot's speed tracking error to obtain the robot's tracking error.

[0072] Among them, the robot's speed tracking error for: ;in, Represents a robot With robots Communication information between robots For robots The neighboring robot, Represents a robot Communication information with the leader robot, For robots The velocity tracking error can also be called the synovial variable.

[0073] S5: Determine the actual control torque of each robot based on the tracking error of each robot. The process includes:

[0074] Determine the optimal control torque of each robot when the robot cooperates with its neighboring robots based on the tracking error of each robot;

[0075] According to the optimal control torque of each robot, the actual control torque of each robot is determined.

[0076] Specifically:

[0077] According to the tracking error of each robot, the optimal performance function with the minimum performance index function value as the goal is determined, wherein when the performance index function value is the minimum, it indicates that the robot and its adjacent robots have achieved coordination;

[0078] Solve the optimal performance function to obtain the optimal control torque of the robot;

[0079] According to the optimal control torque, an adaptive fuzzy logic system is used to determine the unknown items in the optimal control torque;

[0080] The unknown term in the optimal control torque and the optimal control torque are used to determine the actual control torque of the robot.

[0081] The performance indicator function constructed in this embodiment is: .

[0082] Construct a performance indicator function when the performance indicator function value is minimum, which is the optimal performance function: .

[0083] The Hamilton-Jacobi-Bellman equation (HJB equation) for determining the optimal performance function:

[0084]

[0085] The estimated form of the sliding mode variable is obtained , and substitute it into the HJB equation, by solving , to obtain the optimal control torque .

[0086] The universal approximation principle of adaptive fuzzy logic system is used to determine the unknown items in the optimal control torque;

[0087] Substitute the determined unknown term into the optimal control torque to obtain the actual control torque.

[0088] Among them, the optimal control torque for: ; ;in, is the fuzzy basis function vector, is the weight of the fuzzy logic system, is the design weight parameter, is an approximate error.

[0089] This embodiment uses actor-critic reinforcement learning (RL algorithm) to determine the unknown terms in the optimal control torque, where the reinforcement learning training rules are as follows: ; ; .in, and are the critic and actor parameter weights, , and are all positive design constants.

[0090] S6: Control each robot according to the actual control torque of each robot.

[0091] By The formation consists of interconnected wheeled robots, each of which has two active wheels and one passive wheel. The dynamic equation of each wheeled robot is:

[0092] (1)

[0093] in, Represents its position status, Represents its speed state, is a symmetric positive definite inertia matrix; represents the centripetal matrix and the Coriolis force matrix, represents the gravity term, Expressed as the input transformation matrix, is represented as a constraint force vector, is the constraint matrix, is the input torque. Among them, the wheeled mobile robot is characterized by controlling the movement of the robot through two independently driven wheels and a universal wheel for steering direction control. Each wheel has an independent motor to control its rotation speed and direction, and steering is usually performed by controlling the speed difference between the two wheels. In this experiment, considering that the wheels do not slip or slide, the movement of the wheeled mobile robot is subject to non-holonomic constraints.

[0094] Nonholonomic motion constraint satisfaction for wheeled mobile robots After a series of calculations, the dynamic equations of the wheeled robot can be transformed into a steering kinematics and additional drive dynamics model, that is, a robot model with non-holonomic constraints:

[0095] (2)

[0096] in, , , , , , Representative The angular velocity and linear velocity of the robot, for Generates a full row rank matrix of the null space.

[0097] Since the constraints in the robot model are non-holonomic constraints, they are difficult to solve. This embodiment linearizes the robot model and uses a new output vector to convert the position constraint into an output constraint. This embodiment limits the robot to move within the safe area. Therefore, according to the expected trajectory and the mechanical structure of the wheeled mobile robot, the point on the mobile platform where the robot is most likely to leave the safe area can be found. ,point It can be expressed as:

[0098] (3)

[0099] in, For point Position in the robot's local coordinate system.

[0100] The time derivative is:

[0101] (4)

[0102] Substitute (3) and its derivative into (2) and premultiply , the robot model can be rewritten as:

[0103] (5)

[0104] in, , , .

[0105] make , , model (2) can be written as:

[0106] (6)

[0107] Among them, through , , the robot model can be further rewritten into the robot converted model as follows:

[0108] (7)

[0109] Here, .

[0110] A fuzzy inference system consists of four main components: a fuzzy parser, a fuzzy rule base, a fuzzy inference engine, and a defuzzifier. Since there are many different choices for each component, many fuzzy logic systems (FLS) can be generated by grouping these choices. The most popular adaptive fuzzy system in the field of adaptive control is composed of a single fuzzifier, a fuzzy rule base with Gaussian membership functions, a product inference engine, and central average fuzzification. They are introduced below.

[0111] The fuzzy rule base is a collection of the following If-Then rules:

[0112]

[0113] in, For FLS output, for The fuzzy sets in Characterization, where represents the number of fuzzy rules, for The fuzzy sets in Characterization, is the fuzzy rule number. It should be mentioned that Gaussian functions are widely used to define these fuzzy membership functions. and .

[0114] Center average deblurring is mapped by the following function:

[0115] (8)

[0116] in, .

[0117] The fuzzy basis function is defined as:

[0118] (9)

[0119] definition and Finally, FLS can be re-described as:

[0120] (10)

[0121] Lemma 1: For a continuous function In tight set There exists a fuzzy logic system that satisfies:

[0122] (11)

[0123] exist There must be an ideal parameter vector , whose expression is:

[0124] (12)

[0125] So the continuous function can be re-expressed as:

[0126] (13)

[0127] in The approximate error is expressed as For the boundary, is an arbitrary design constant.

[0128] In order to ensure the estimation performance of the observer, the fuzzy approximation technology is introduced into the observer design, which does not rely on the system model information:

[0129] (14)

[0130] in , and is the designed observer gain constant.

[0131] Tuning is performed using the following adaptive rules:

[0132] (15)

[0133] in is the design constant. According to equations (7) and (14), the dynamic observation error can be obtained as:

[0134] (16)

[0135] Theorem 1: If the fuzzy state observer (14) has a training law, and the unmeasurable state of the system (7) is estimated using (15) with bounded initial values, then

[0136] 1) Observation error and FLS parameter vector error is semi-globally uniformly bounded;

[0137] 2) By making these observer gain constants large enough, the observation error can converge to the required accuracy.

[0138] Proof: Set the Lyapunov function to verify the stability of the above state:

[0139] (17)

[0140] The time derivative of equations (15) and (16) is:

[0141] (18)

[0142] According to Young's inequality, and using , we can get the following results:

[0143] (19)

[0144] in , and with a constant For the boundary, that is .

[0145] make:

[0146] ,in yes The minimum eigenvalue of , then formula (18) can be expressed as:

[0147] (20)

[0148] Lemma 2: Let is a positive continuous function. If it satisfies ,in , then the following inequality holds:

[0149] (twenty one)

[0150] According to Lemma 2, we can achieve the following result:

[0151] (twenty two)

[0152] The above inequality means that all errors of the observer system are semi-globally uniformly bounded and the observation errors can converge.

[0153] Construction Distributed tracking error of th robot :

[0154] (twenty three)

[0155] in , Represents the trajectory of the reference.

[0156] The desired virtual velocity input for kinematics is constructed as:

[0157] (twenty four)

[0158] In order to achieve multi-agent leader-follower consistency, a speed error is introduced:

[0159] (25)

[0160] Set its sliding mode variable to:

[0161] (26)

[0162] By taking its derivative, we can get:

[0163] (27)

[0164] in Is an agent The neighbor label set of .

[0165] Define the distributed performance function as: (28)

[0166] Cost Function It can be described as:

[0167] .

[0168] in, , .

[0169] The optimal performance function is:

[0170] (29)

[0171] Then the HJB equation related to the above formula can be expressed as:

[0172] (30)

[0173] By solving the equation , we can get :

[0174] (31)

[0175] According to Theorem 1, the derivative of the sliding mode variable can be estimated as ,Will Substituting into the above formula, the estimated optimal control torque can be obtained as:

[0176] (32)

[0177] In order to realize tracking control, the above formula is decomposed into the following style:

[0178] (33)

[0179] in is a continuous function, is the design constant, then we can get:

[0180] (34)

[0181] Due to unknown function is continuous, so it can be used with FLS in a compact set Upper approximation:

[0182] (35)

[0183] (36)

[0184] (37)

[0185] Since the ideal FLS parameter vector is an unknown constant vector. In order to obtain effective optimized distributed control, the RL algorithm is executed by the following Critic and Actor. According to formula (38), the control performance evaluation index is designed as:

[0186] (38)

[0187] Critic FLS parameter vector Trained by the following adaptive rules:

[0188] (39)

[0189] in Design constants for Critic.

[0190] The Actor used to perform control actions is designed according to formula (39):

[0191] (40)

[0192] Actor FLS parameter vector Trained by the following adaptive rules:

[0193] (41)

[0194] in Design constants for Actors.

[0195] Proof: The Lyapunov function can be chosen as:

[0196] (42)

[0197] in , .

[0198] By taking the derivative of the above Lyapunov function, we get the following formula:

[0199] (43)

[0200] According to Young's inequality, etc., through a series of mathematical operations, we can get the following results:

[0201] (44)

[0202] in, , and with a constant For the boundary, .

[0203] make Formula (44) can be expressed as:

[0204] (45)

[0205] According to the above Lemma 2, we get:

[0206] (46)

[0207] The above inequality shows that all error signals are semi-globally uniformly bounded and the sliding mode variable can converge to a small neighborhood around zero by appropriately increasing the value of the gain.

[0208] Select 4 wheeled mobile robots, including 1 virtual leader robot and 4 follower robots. Its parameter matrix and vector are as follows:

[0209]

[0210] , .

[0211] in and Respectively represent the mass and moment of inertia of the wheeled mobile robot, and the remaining parameters are specified as: The reference angular velocity and linear velocity are , , whose reference trajectory is a circle.

[0212] The fuzzy rules of the fuzzy logic system are divided into 8 and 10, and its fuzzy basis function vector is , ,in , exist Evenly distributed in.

[0213] According to the constructed fuzzy state observer, select the design constant , , and , , , the relevant critic and actor networks are designed to contain 10 nodes, each node in arrive Evenly spaced.

[0214] Using the mobile robot formation control method disclosed in this embodiment, four mobile robots are simulated and controlled. The simulation control results are as follows: Figure 3-Figure 7 As shown, Figure 3 The schematic diagram of the motion trajectory simulation results of four mobile robots; Figure 4 (a) is a schematic diagram of the error simulation results between the linear velocity of the four mobile robots and the expected linear velocity. Figure 4 (b) is a schematic diagram of the simulation results of the error between the angular velocity of the four mobile robots and the expected angular velocity; Figure 5 (a), (b), (c) and (d) are schematic diagrams of the simulation results of the observer parameters of the fuzzy logic system of four mobile robots; Figure 6(a), (b), (c) and (d) are schematic diagrams of the simulation results of the Actor parameters of four mobile robots; Figure 7 (a), (b), (c) and (d) are schematic diagrams of the simulation results of the Critic parameters of four mobile robots, respectively, showing the boundedness of these parameter matrices. The simulation control results further prove the accuracy of the mobile robot formation control method disclosed in this embodiment for mobile robot formation control.

[0215] Example 2

[0216] In this embodiment, a mobile robot formation control system is disclosed, comprising:

[0217] An acquisition module is used to obtain the position, velocity and acceleration of each robot in the robot formation;

[0218] A data conversion module, for determining a converted speed of each robot considering position constraints according to the position, speed and acceleration of each robot;

[0219] The tracking error determination module is used to determine the speed tracking error of each robot according to the converted speed of each robot; and obtain the tracking error of each robot according to the speed tracking error of the robot and the speed tracking error of its adjacent robot;

[0220] A control torque determination module, used for determining the actual control torque of each robot according to the tracking error of each robot;

[0221] The control module is used to control each robot according to the actual control torque of each robot.

[0222] The present invention also discloses a computer device, which includes:

[0223] a processor adapted to execute a computer program;

[0224] A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a mobile robot formation control method disclosed in Example 1 is implemented.

[0225] The present invention also discloses a computer-readable storage medium, which stores a computer program. The computer program is suitable for being loaded by a processor and executing a mobile robot formation control method disclosed in Example 1.

[0226] The present invention also discloses a computer program product, which includes a computer program. When the computer program is executed by a processor, a mobile robot formation control method disclosed in Example 1 is implemented.

[0227] The method disclosed in Example 1 can be directly embodied as a hardware processor, or can be executed by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0228] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with this embodiment 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 to be beyond the scope of this application.

[0229] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A mobile robot formation control method, characterized in that: include: Get the position, velocity and acceleration of each robot in the robot formation; Determine the transformed velocity of each robot considering the position constraint according to the position, velocity and acceleration of each robot; Determine the speed tracking error of each robot according to the converted speed of each robot; The tracking error of each robot is obtained according to the speed tracking error of the robot and the speed tracking errors of its adjacent robots; According to the tracking error of each robot, the actual control torque of each robot is determined; Control each robot according to its actual control torque; Determine the converted speed of each robot considering the position constraint by using the converted model according to the position, speed and acceleration of each robot, wherein the converted model is obtained by introducing variables representing the points at which the robot is most likely to leave the safety area into the robot model containing the non-holonomic constraints; Among them, the robot model with non-holonomic constraints is: in represents the position of the robot, represents the speed of the robot, represents the symmetric positive definite inertia matrix, represents the Coriolis force and the centrifugal force, represents the gravity twisting the vector, is the input torque; Introducing variables Y i , let the first variable x 1i Substitute new variables Y i , the second variable x 2i Substituting the first-order derivative of the new variable, the robot system model is linearized to obtain the converted model: ,in, is the point at which the robot is most likely to leave the safe area, , , ; Substitute the acquired position, velocity and acceleration of each robot into the converted model to calculate the converted velocity of the robot x 2i .

2. A mobile robot formation control method as claimed in claim 1, characterized in that: The speed tracking error of the robot is equal to the error between the converted speed of the robot and the expected virtual speed of the robot, wherein the expected virtual speed of the robot is calculated according to the expected trajectory of the robot.

3. A mobile robot formation control method as claimed in claim 1, characterized in that: The deviation between the robot's speed tracking error and the speed tracking error of its adjacent robots is calculated; the deviation is weightedly fused with the robot's speed tracking error to obtain the robot's tracking error.

4. A mobile robot formation control method as claimed in claim 1, characterized in that: Determine the optimal control torque of each robot when the robot cooperates with its neighboring robots based on the tracking error of each robot; According to the optimal control torque of each robot, the actual control torque of each robot is determined.

5. A mobile robot formation control method as claimed in claim 4, characterized in that: According to the tracking error of each robot, the optimal performance function with the minimum performance index function value as the goal is determined, wherein when the performance index function value is the minimum, it indicates that the robot and its adjacent robots have achieved coordination; Solve the optimal performance function to obtain the optimal control torque of the robot; According to the optimal control torque, an adaptive fuzzy logic system is used to determine the unknown items in the optimal control torque; The unknown term in the optimal control torque and the optimal control torque are used to determine the actual control torque of the robot.

6. A mobile robot formation control system, characterized in that: include: An acquisition module is used to obtain the position, velocity and acceleration of each robot in the robot formation; A data conversion module is used to determine the converted speed of each robot considering the position constraint according to the position, speed and acceleration of each robot; wherein the converted model is obtained by introducing a variable representing a point at which the robot is most likely to leave the safe area into the robot model containing the non-holonomic constraints; Among them, the robot model with non-holonomic constraints is: in represents the position of the robot, represents the speed of the robot, represents the symmetric positive definite inertia matrix, represents the Coriolis force and the centrifugal force, represents the gravity twisting the vector, is the input torque; Introducing variables Y i , let the first variable x 1i Substitute new variables Y i , the second variable x 2i Substituting the first-order derivative of the new variable, the robot system model is linearized to obtain the converted model: ,in, is the point at which the robot is most likely to leave the safe area, , , ; Substitute the acquired position, velocity and acceleration of each robot into the converted model to calculate the converted velocity of the robot x 2i The tracking error determination module is used to determine the speed tracking error of each robot according to the converted speed of each robot; and obtain the tracking error of each robot according to the speed tracking error of the robot and the speed tracking error of its adjacent robot; A control torque determination module, used for determining the actual control torque of each robot according to the tracking error of each robot; The control module is used to control each robot according to the actual control torque of each robot.

7. An electronic device, characterized in that: The device comprises: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a mobile robot formation control method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a mobile robot formation control method as described in any one of claims 1-5.

9. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, it implements a mobile robot formation control method as described in any one of claims 1-5.

Images