Modular formation collision avoidance control system and method for multi-wheel mobile robot

Abstract

The application discloses a kind of multi-wheel mobile robot modularization formation collision avoidance control system and method, including online regulator module and trajectory tracking controller module.First, a plurality of virtual intelligent agents are constructed in online regulator module, and formation collision avoidance reference trajectory satisfying the requirements of wheeled mobile robot system is generated;Then, the trajectory tracking controller module of wheeled mobile robot is designed, so that it accurately tracks formation collision avoidance reference trajectory, to realize formation collision avoidance control.The system and method of the application can perform formation collision avoidance task without changing the controller of wheeled mobile robot, enhance the generality and convenience of design, is the expansion of prior art method, and is also a kind of prediction to possible application direction of future multi-wheel mobile robot system.

Description

Technical Field

[0001] This invention belongs to the field of robot control technology, and mainly relates to a modular formation collision avoidance control system and method for multi-wheeled mobile robots. Background Technology

[0002] In recent decades, the widespread application of formation control for wheeled mobile robots (WMRs) in logistics, resource exploration, and environmental monitoring has attracted increasing research from scientists. Compared to a single WMR, a group of WMRs can maintain a pre-designed formation shape, efficiently complete more complex collaborative tasks, and exhibit higher reliability, stronger robustness, and better economy. Many control methods have been proposed for the formation control of multiple WMRs, such as virtual structure methods, behavior-based methods, artificial potential field methods, and leader-follower methods. However, these methods largely follow the conventional formation control design approach, directly incorporating coordination errors into the formation controller design. In practical applications, some wheeled mobile robot (WMR) products are already equipped with pre-defined controllers. According to traditional design logic, to complete different formation control tasks, it is often necessary to redesign and replace the WMR controller, which creates many obstacles for practical applications. This demonstrates the limitations of traditional formation control design concepts.

[0003] In addition, avoiding collisions is a significant challenge in collaborative tasks involving multiple wheeled mobile robots, as these robots need to work closely together. In tasks such as encirclement, search and rescue, and autonomous driving, multiple robots often share the same workspace, making collision avoidance a critical technical issue. Traditional formation collision avoidance control methods rely heavily on centralized control or simple distance sensing, but these methods generally incur excessive computational overhead in large-scale robotic systems and lack pre-processing capabilities, failing to guarantee safe and efficient collaborative operation between wheeled mobile robots. Therefore, a more efficient and flexible formation collision avoidance control method for multi-wheeled mobile robots is urgently needed. Summary of the Invention

[0004] This invention addresses the shortcomings of conventional formation control designs for wheeled mobile robots and the collision problems that easily occur in collaborative tasks involving multiple wheeled mobile robots. It provides a modular formation collision avoidance control system and method for multi-wheeled mobile robots. The modular design decomposes the formation control system into an online regulator module and a trajectory tracking controller module. Multiple virtual agents are constructed in the online regulator module to generate formation collision avoidance reference trajectories that meet the requirements of the wheeled mobile robot system. The trajectory tracking controller module for the wheeled mobile robot is then designed to accurately track the formation reference trajectory, thereby achieving formation control. The system of this invention can perform different formation tasks without changing the wheeled mobile robot controller, solving the formation collision problem. It enhances the versatility and convenience of the design, expands upon existing technologies, and foreshadows potential applications for future multi-wheeled mobile robot systems. The success of this research will contribute to the practical application of multi-wheeled mobile robot systems and provide a more effective tool for solving formation collision avoidance problems.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A formation collision avoidance control system for a multi-wheeled mobile robot system includes at least an online regulator module and a trajectory tracking controller module.

[0006] The online regulator module: constructs multiple virtual agents for the wheeled mobile robot. The virtual agents obtain the state information of their neighboring virtual agents through their communication topology and make a safe distance judgment; introduces obstacle function gain to generate a safe reference trajectory that meets the collision avoidance requirements of the wheeled mobile robot formation;

[0007] The trajectory tracking controller module is used to adjust the linear velocity and angular velocity of the wheeled mobile robot, introduces a different obstacle function gain than that in the online regulator module, and iteratively corrects motion deviations through an error feedback mechanism to achieve trajectory tracking and formation collision avoidance.

[0008] As an improvement to the present invention, the online regulator module is designed as a second-order integrator:

[0009]

[0010] in and Virtual intelligent agents The position and velocity of the generated safety reference trajectory;

[0011] Input of the online regulator module Specifically:

[0012]

[0013] in This indicates the control input of the online regulator module; Represents virtual intelligent agents and Communication weight value; virtual intelligent agent and The gain of the barrier function between them is The sum of the obstacle function gains for each virtual agent is ; Represents virtual intelligent agents Formation configuration, Represents virtual intelligent agents The second derivative of the formation configuration and These represent the position and velocity errors of the safety reference trajectory and the formation reference trajectory generated by the virtual intelligent agent, respectively. and These represent the pre-defined formation reference trajectory and its derivative, respectively. This represents the second derivative of the pre-defined formation reference trajectory; Indicates and The relevant communication weight values ​​for virtual intelligent agents If it can obtain the formation reference trajectory state, then ,on the contrary ; This indicates the control gain.

[0014] As another improvement of the present invention, the trajectory tracking controller module is configured as follows:

[0015]

[0016] in, For robots The distance from the center on its axis is The point at that location is in a Cartesian coordinate system. The location of the calibration point in the middle, For the control input of the trajectory tracking controller module, For the control input of the online regulator module, Indicates control gain. for With reference signal Tracking error, This is the obstacle function gain in the trajectory tracking controller module.

[0017] As another improvement of the present invention, the obstacle function gain in the trajectory tracking controller module is specifically as follows:

[0018]

[0019] in, For robots actual location With reference signal Tracking error, For tracking error The upper boundary.

[0020] To achieve the above objectives, the present invention also adopts the following technical solution: a modular formation collision avoidance control method for multi-wheeled mobile robots, comprising at least the following steps:

[0021] S1: Among multiple wheeled mobile robots, for each robot Construct a virtual intelligent agent system consistent with its communication topology, and virtual intelligent agents within the virtual intelligent agent system. It interacts with neighboring virtual intelligent agents through communication topology. Status information;

[0022] S2: Collect initial data, which includes at least the virtual agent at the initial moment. Location and speed Formation reference trajectory Formation configuration and virtual intelligent agents and Communication weight value S3: Based on the data collected in step S2, construct an online regulator module. The input of the online regulator module... Specifically:

[0023]

[0024] in, This indicates the control input of the online regulator module; Represents virtual intelligent agents and The communication weight value; For virtual intelligent agents and The gain of the barrier function between, and Representing virtual intelligent agents The position and velocity errors between the generated safety reference trajectory and the formation reference trajectory, where This indicates the pre-set formation reference trajectory. This represents the derivative of a pre-defined formation reference trajectory. The second derivative of the pre-defined formation reference trajectory; Represents virtual intelligent agents Formation configuration, Represents virtual intelligent agents The second derivative of the formation configuration; Indicates and The relevant communication weight values ​​for virtual intelligent agents If it can obtain the formation reference trajectory status, ,on the contrary ; Indicates control gain;

[0025] Barrier function gain Specifically:

[0026]

[0027] in, , For virtual intelligent agents and The Euclidean distance between them For virtual intelligent agents and The sum of the safety radii between them The collision trigger radius, , These are the parameters of the barrier function;

[0028] S4: Establish the kinematic model of the multi-wheeled mobile robot, based on linear velocity. and angular velocity Realize the wheeled mobile robot The center point position of the wheeled mobile robot is obtained through control. And based on the position of the center point Construct a kinematic model of the calibration point of a multi-wheeled mobile robot; based on the controller at the calibration point of the wheeled mobile robot... The linear velocity at the center point of the wheeled mobile robot is obtained through mathematical mapping. and angular velocity The center point position of the wheeled mobile robot is obtained. This controls the movement of the wheeled mobile robot; the mathematical mapping is:

[0029]

[0030] in, Indicates the direction of movement of the center point and The included angle of the axis, It is the set of real numbers; and These represent the calibration points of the wheeled mobile robot along... shaft and Linear acceleration of the axis; This indicates the distance from the center point of the wheeled mobile robot to the calibration point.

[0031] S5: Build a controller that includes all calibration points of the wheeled mobile robot. The trajectory tracking controller module for wheeled mobile robots Perform the following controls:

[0032]

[0033] in, For the control input of the online regulator module, Indicates control gain. The obstacle function gain in the trajectory tracking controller module. For robots actual location With reference signal Tracking error;

[0034] S6: Based on the mathematical mapping in step S4, the controller at the marked point on the wheeled mobile robot obtained in step S5 is... Substitute the values ​​to obtain the linear velocity for each wheeled mobile robot. and angular velocity The system drives multi-wheeled mobile robots to move in formation to avoid collisions; the principle of collision avoidance is that the distance between a robot and its neighboring robots is Euclidean distance. ,in and Robots and robots Location, Representing vectors The Euclidean norm; the sum of the safe radii of the robot and its neighboring robots. ,in and Robots and robots The safety radius; when A collision will not occur.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] (1) This invention discloses a modular formation collision avoidance control system and method for multi-wheeled mobile robots, which solves the problem that traditional methods cannot accurately form formations for collision avoidance in complex environments. Through modular design, each virtual agent can autonomously perceive the position and velocity information of neighboring virtual agents and perform collision detection locally, significantly improving the robustness and reliability of the system. Modular design simplifies the control process, making the method have good scalability and flexibility in large-scale robot systems.

[0037] (2) This invention introduces obstacle function gain to generate a safe trajectory, enabling wheeled mobile robots to achieve formation collision avoidance in complex environments. The online regulator module generates a safe reference trajectory by introducing obstacle function gain, ensuring that the movement trajectory of the wheeled mobile robot is always within the safe area. At the same time, different obstacle function gains are introduced in the trajectory tracking controller module for secondary collision avoidance, enabling it to accurately track the safe reference trajectory, thereby achieving autonomous formation collision avoidance.

[0038] (3) The formation collision avoidance control method of the present invention has low design complexity. It controls the multi-wheeled mobile robot system through modular design and enables the system to pre-process collisions through a secondary collision avoidance algorithm. It can respond to environmental changes in real time and is suitable for large-scale real-time control in various scenarios. This formation collision avoidance control method can play an advantage in tasks such as multi-wheeled mobile robot encirclement, drone formation, and autonomous vehicle fleet, significantly improving the efficiency and safety of collaborative tasks. Attached Figure Description

[0039] Figure 1 This is a flowchart illustrating the steps of a modular formation collision avoidance control method for a multi-wheeled mobile robot according to the present invention.

[0040] Figure 2 This is a schematic diagram of the simulation verification environment in the test examples of this invention;

[0041] Figure 3 This is a time-sharing diagram of the formation collision avoidance verification simulation in the test examples of this invention;

[0042] Figure 4 This is a diagram showing the relative distance record of the robot in the simulation verification of the test example of this invention. Detailed Implementation

[0043] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0044] Example 1

[0045] This paper addresses a collision avoidance control system for multi-wheeled mobile robot formations, achieving effective collision avoidance between the robots and ensuring the safe and efficient operation of the formation. The system employs a modular design, comprising an online regulator module and a trajectory tracking controller module. First, multiple virtual agents are constructed in the online regulator module, and obstacle function gains are introduced to generate safe reference trajectories that meet the collision avoidance requirements of the wheeled mobile robot formation. Second, trajectory tracking controllers for the wheeled mobile robots are designed in the trajectory tracking controller module, and different obstacle function gains are introduced for secondary collision avoidance, ensuring accurate tracking of the safe reference trajectory and thus preventing collisions during the formation process.

[0046] A modular formation collision avoidance control method for multi-wheeled mobile robots, such as Figure 1 As shown, the specific steps include the following:

[0047] Step S1: First, define the common... A wheeled mobile robot, for wheeled mobile robots , , , Let be the set of positive integers. To simulate the formation collision avoidance behavior of a multi-wheeled mobile robot, a... A virtual intelligent agent, for the virtual intelligent agent and wheeled mobile robots correspond, , Then, a communication topology for multi-wheeled mobile robots is constructed for state information exchange between the wheeled robots, thereby achieving formation collision avoidance.

[0048] The communication topology of a multi-wheeled mobile robot is defined as an undirected graph. , For point set, For wheeled mobile robots, the edge set is used. It can be achieved through communication topology To obtain neighbor's wheeled mobile robot Status information, , , Represents wheeled mobile robots The set of neighbors. After this, the communication topology between virtual agents is constructed in the same way as that of the wheeled mobile robot, using an undirected graph. For virtual intelligent agents It can be achieved through communication topology To obtain neighbor virtual intelligent agents Status information, , , Represents wheeled mobile robots A set of neighbors. Defining a virtual intelligent agent. and The communication weight value is .

[0049] Step S2: Collect the data required by the online regulator module, including the initial position of the virtual agent. and speed Pre-set formation reference trajectory Virtual intelligent agent Formation configuration And the virtual agent from step S1 and Communication weight value .

[0050] Step S3: Receive the data collected in Step S2 and construct an online regulator module to allow multiple virtual agents to pre-simulate formation collision avoidance tasks instead of a multi-wheeled mobile robot. The online regulator module is designed as a second-order integrator.

[0051]

[0052] in and Virtual intelligent agents The position and velocity of the generated safety reference trajectory;

[0053] Input of the online regulator module Specifically:

[0054]

[0055] in, This indicates the control input of the online regulator module; Represents virtual intelligent agents and The communication weight value; For virtual intelligent agents and The barrier function gain between them, specifically

[0056]

[0057] in, For virtual intelligent agents and The Euclidean distance between them For about The barrier function gain, , For virtual intelligent agents and The sum of the safety radii between them The collision trigger radius, , These are the obstacle function parameters; the sum of the obstacle function gains for each virtual agent is... ; and Representing virtual intelligent agents The position and velocity errors between the generated safety reference trajectory and the formation reference trajectory, where The derivative of the formation reference trajectory, The second derivative of the formation reference trajectory; Represents virtual intelligent agents Formation configuration, Represents virtual intelligent agents The second derivative of the formation configuration; Indicates and The relevant communication weight values, This represents a pre-defined formation reference trajectory; for virtual intelligent agents If it can obtain the formation reference trajectory status, ,on the contrary ; This indicates the control gain.

[0058] Step S4: Establish the kinematic model of the multi-wheeled mobile robot. For wheeled mobile robots... Its kinematic model can be expressed as

[0059] ,

[0060] in, , For the set of real numbers, These represent the x-coordinate and y-coordinate of the center point of the wheeled mobile robot, respectively. This represents the linear velocity at the center point; Indicates the direction of movement of the center point and The included angle of the axis; Let the angular velocity at the center point be... , If we consider a two-dimensional Euclidean space, then This can represent the center point of a wheeled mobile robot in a Cartesian coordinate system. The position in the middle.

[0061] For wheeled mobile robots The linear velocity needs to be given. and angular velocity To control it, thereby obtaining the center point position of the wheeled mobile robot. Based on this, a kinematic model of the calibration point for a multi-wheeled mobile robot is constructed. (This is followed by a seemingly unrelated sentence about wheeled mobile robots.) If we take any point on the central axis as the calibration point, the kinematic model of the calibration point is as follows:

[0062]

[0063] in, This represents the distance from the center point of the wheeled mobile robot to the calibration point. The x-coordinate of the robot calibration point. Represents the ordinate of the robot's calibration point; Indicates the direction of movement of the robot's calibration point and The included angle of the axis.

[0064] set up ,but This can represent the calibration points of a wheeled mobile robot in a Cartesian coordinate system. The location within. For the calibration point of a wheeled mobile robot. In other words, it can be viewed as a second-order integrator system, specifically as follows:

[0065]

[0066] in, This refers to the controller at the calibration point of the wheeled mobile robot. and These represent the calibration points of the wheeled mobile robot along... shaft and Linear acceleration of the axis.

[0067] Based on mathematical mapping, the following relationship can be obtained.

[0068]

[0069] Therefore, by designing a controller at the calibration point of the wheeled mobile robot... The linear velocity at the center point of the wheeled mobile robot is obtained through mathematical mapping. and angular velocity The center point of the wheeled mobile robot can then be obtained. This allows for the control of the wheeled mobile robot's movement.

[0070] Step S5: Design a trajectory tracking controller module, which includes controllers for all calibration points of the wheeled mobile robot. The control input of the online regulator module obtained in step S3. Virtual intelligent agent Location of generated safety reference trajectory and speed For wheeled mobile robots The corresponding controller forms are as follows:

[0071]

[0072] in, For robots The distance from the center on its axis is The point at that location is in a Cartesian coordinate system. The position in the middle, For the control input of the trajectory tracking controller module, Indicates control gain. The obstacle function gain in the trajectory tracking controller module is specifically...

[0073]

[0074] in, For robots actual location With reference signal Tracking error, For tracking error The upper boundary.

[0075] Step S6: The controller at the marked point on the wheeled mobile robot... Reverse substitution step S4 In this process, the linear velocity corresponding to each wheeled mobile robot is obtained. and angular velocity This ultimately drives the movement of multi-wheeled mobile robots to achieve formation collision avoidance. During execution, the principle of collision avoidance for the wheeled mobile robots is: the distance between the robot and its neighboring robots is Euclidean distance, specifically:

[0076]

[0077] in and Robots and robots Location, Representing vectors The Euclidean norm; the sum of the safety radii of the robot and its neighboring robots is:

[0078]

[0079] in and Robots and robots The safety radius;

[0080] when A collision will not occur.

[0081] In summary, the proposed system and method for formation collision avoidance control of multi-wheeled mobile robot systems adopts a modular design. Based on an online regulator and trajectory tracking controller, it can complete the formation of multi-wheeled mobile robots while ensuring collision avoidance. It has high robustness and low algorithm design complexity, and is suitable for real-time control of large-scale multi-wheeled mobile robot systems.

[0082] Test case

[0083] like Figure 2 As shown, MATLAB simulation was used to verify the escort scenario of a multi-wheeled mobile robot. Five WMRs tracked three other WMRs and formed a formation to escort them, with all WMRs maintaining collision avoidance throughout the process. The method used is as follows: Figure 1 The modular formation collision avoidance control method for multi-wheeled mobile robots based on an online regulator, as shown, constructs multiple virtual agents in the online regulator module and introduces obstacle function gain to generate a safe reference trajectory that meets the formation requirements of the wheeled mobile robots. Then, the trajectory tracking controller module of the wheeled mobile robot accurately tracks the safe reference trajectory, thereby achieving formation control. The specific process is as follows:

[0084] 1. Verification scenario and parameter settings

[0085] Number and Roles of Robots: The verification involved a total of 8 WMRs. Among them, the 3 robots numbered R1, R2, and R3 acted as escort targets, moving along a preset straight path; the 5 robots numbered F1 to F5 acted as escort robots, which needed to form and maintain a rectangular formation to accompany and escort the targets.

[0086] Communication topology: The five escort robots use an undirected communication topology, and each robot can exchange status information with the two adjacent robots. Escort robots F1 to F3 can communicate with the escorted target robots R1 to R3 respectively to obtain the formation reference trajectory.

[0087] Control parameters: Online regulator module controls the gain. , The trajectory tracking controller module controls the gain. , ; Barrier function gain parameter , safe distance .

[0088] 2. Verification process and key steps

[0089] The simulation lasts for a total of 40 seconds and is divided into four stages according to the task progress:

[0090] Initial Phase (0s): All robots disperse in the initial area. Escort targets R1~R3 ​​begin moving along a preset straight trajectory. Escort robots F1~F5 utilize obstacle function gain. Combined with control gain Through the online regulator module , Generate a formation collision avoidance reference trajectory.

[0091] Chase Phase (0-15s): Escort robots converge towards the target formation position. Obstacle function gain is utilized. Through the trajectory tracking controller Ensure that the escort robots track the formation's collision avoidance reference trajectory in real time and avoid collisions.

[0092] Formation formation and stabilization phase (approximately 15 seconds): At approximately 15 seconds, the escort robots successfully form the predetermined formation (rectangle).

[0093] Escort phase (15~40s): After the formation is formed, all robots move in coordination and always maintain the formation.

[0094] 3. Verify data collection and analysis

[0095] Key collision avoidance performance data were collected during the simulation to quantitatively analyze the effectiveness of the method: the relative distances between all pairs of robots were recorded throughout the simulation. ,like Figure 3 As shown in the data, throughout the entire verification process (0-40 seconds), the minimum distance between any two robots never fell below the safe distance. No collision occurred. The highest risk of collision occurred during... The relative distance between escort robots F2 and F3 reaches a minimum of 0.257m, at which point the obstacle function gain increases significantly, ensuring that no collision occurs between the robots. Therefore, Figure 3 This demonstrates that, based on the proposed method, multi-wheeled mobile robots can consistently avoid collisions.

[0096] like Figure 4 The images show time-series images of the multi-wheeled mobile robot system's formation collision avoidance verification during the initial phase (0s), pursuit phase (0~15s), successful pursuit (15s), and escort phase (15s~40s). In the initial phase, escort robots F1-F3 were not yet in formation. During the pursuit phase, escort robots F1-F3 began performing formation tasks and gradually formed a formation. Upon successful pursuit, escort robots F1-F3 had achieved a square formation to escort the target. During the escort phase, escort robots F1-F3 maintained the square formation while providing escort. Figure 4The proposed system and method demonstrate that multi-wheeled mobile robots can form and maintain a specified formation.

[0097] In conclusion, Figure 3 and Figure 4 This demonstrates that, under the proposed system and method, multi-wheeled mobile robots can achieve formation collision avoidance.

[0098] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.

Claims

1. A modular formation collision avoidance control system for a multi-wheeled mobile robot, characterized in that... It includes an online regulator module and a trajectory tracking controller module. The online regulator module: constructs multiple virtual agents for the multi-wheeled mobile robot. The virtual agents obtain the status information of neighboring virtual agents through their communication topology and make a safe distance judgment; introduces obstacle function gain to generate a safe reference trajectory that meets the collision avoidance requirements of the wheeled mobile robot formation; The trajectory tracking controller module includes at least a controller, which achieves formation collision avoidance by introducing a different obstacle function gain than that in the online regulator module and by using the tracking error input into the controller.

2. The modular formation collision avoidance control system for a multi-wheeled mobile robot as described in claim 1, characterized in that: The online regulator module is designed as a second-order integrator. ； in and Virtual intelligent agents The position and velocity of the generated safety reference trajectory; Input of the online regulator module Specifically: ； in, This indicates the control input of the online regulator module; Represents virtual intelligent agents and The communication weight value; For virtual intelligent agents and The barrier function gain between them; the sum of the barrier function gains of each virtual agent is ; and Representing virtual intelligent agents The position and velocity errors between the generated safety reference trajectory and the formation reference trajectory are as follows: This indicates the pre-set formation reference trajectory. The derivative of the formation reference trajectory, The second derivative of the formation reference trajectory; Represents virtual intelligent agents Formation configuration, Represents virtual intelligent agents The second derivative of the formation configuration; Indicates and The relevant communication weight values ​​for virtual intelligent agents If it can obtain the formation reference trajectory status, ,on the contrary ; This indicates the control gain.

3. The modular formation collision avoidance control system for a multi-wheeled mobile robot as described in claim 1, characterized in that: The trajectory tracking controller module is configured as follows: ； in, For robots The distance from the center on its axis is The point at that location is in a Cartesian coordinate system. The position in the middle, For the control input of the trajectory tracking controller module, For the control input of the online regulator module, Indicates control gain. for With reference signal Tracking error, This is the obstacle function gain in the trajectory tracking controller module.

4. The modular formation collision avoidance control system for a multi-wheeled mobile robot as described in claim 3, characterized in that: The obstacle function gain in the trajectory tracking controller module is: ； in For robots actual location With reference signal Tracking error, For tracking error The upper boundary.

5. A modular formation collision avoidance control method for a multi-wheeled mobile robot, using the system as described in claim 1, characterized in that, It should include at least the following steps: S1: Among multiple wheeled mobile robots, for each robot Construct a virtual intelligent agent system consistent with its communication topology, and virtual intelligent agents within the virtual intelligent agent system. It interacts with neighboring virtual intelligent agents through communication topology. Status information; S2: Collect initial data, which includes at least the virtual agent at the initial moment. Location and speed Formation reference trajectory The derivative of the formation reference trajectory Formation configuration and virtual intelligent agents and Communication weight value ; S3: Based on the data collected in step S2, construct an online regulator module. The input of the online regulator module... Specifically: ； in, This indicates the control input of the online regulator module; Represents virtual intelligent agents and The communication weight value; For virtual intelligent agents and The gain of the barrier function between them and Representing virtual intelligent agents The position and velocity errors between the generated safety reference trajectory and the formation reference trajectory, where This indicates the pre-set formation reference trajectory. The derivative of the pre-defined formation reference trajectory; The second derivative of the formation reference trajectory; Represents virtual intelligent agents Formation configuration, Represents virtual intelligent agents The second derivative of the formation configuration; Indicates and The relevant communication weight values ​​for virtual intelligent agents If it can obtain the formation reference trajectory status, ,on the contrary ; Indicates control gain; Barrier function gain Specifically: ； in, , For virtual intelligent agents and The Euclidean distance between them For virtual intelligent agents and The sum of the safety radii between them The collision trigger radius, , These are the parameters of the barrier function; S4: Establish the kinematic model of the multi-wheeled mobile robot, based on linear velocity. and angular velocity Realize the wheeled mobile robot The center point position of the wheeled mobile robot is obtained through control. And based on the position of the center point Construct a kinematic model of the calibration point of a multi-wheeled mobile robot; based on the controller at the calibration point of the wheeled mobile robot... The linear velocity at the center point of the wheeled mobile robot is obtained through mathematical mapping. and angular velocity The center point position of the wheeled mobile robot is obtained. This controls the movement of the wheeled mobile robot; the mathematical mapping is: ； in, Indicates the direction of movement of the center point and The included angle of the axis, It is the set of real numbers; and These represent the calibration points of the wheeled mobile robot along... shaft and Linear acceleration of the axis; This indicates the distance from the center point of the wheeled mobile robot to the calibration point. S5: Build a controller that includes all calibration points of the wheeled mobile robot. The trajectory tracking controller module for wheeled mobile robots Perform the following controls: ； in, For the control input of the online regulator module, Indicates control gain. The obstacle function gain in the trajectory tracking controller module. For robots actual location With reference signal Tracking error; S6: Based on the mathematical mapping in step S4, the controller at the marked point on the wheeled mobile robot obtained in step S5 is... Substitute the values ​​to obtain the linear velocity for each wheeled mobile robot. and angular velocity The system drives multi-wheeled mobile robots to move in formation to avoid collisions; the principle of collision avoidance is that the distance between a robot and its neighboring robots is Euclidean distance. ,in and Robots and robots Location, Representing vectors The Euclidean norm; the sum of the safe radii of the robot and its neighboring robots. ,in and Robots and robots The safety radius; when A collision will not occur.

6. The modular formation collision avoidance control method for a multi-wheeled mobile robot as described in claim 5, characterized in that: In step S4, the kinematic model of the multi-wheeled mobile robot is specifically as follows: ， ； in, , These represent the x-coordinate and y-coordinate of the center point of the wheeled mobile robot, respectively. The kinematic model of the calibration point of the multi-wheeled mobile robot is as follows: ； in This represents the distance from the center point of the wheeled mobile robot to the calibration point. The x-coordinate of the robot calibration point. Represents the ordinate of the robot's calibration point; Indicates the direction of movement of the robot's calibration point and The included angle of the axis.

Images