Autonomous navigation obstacle avoidance method and device for robot cluster, medium and product

Abstract

The invention discloses a robot cluster autonomous navigation obstacle avoidance method and device, a medium and a product, and relates to the field of robot cluster control, and the method comprises the steps: carrying out the feature extraction and fusion of environment information and the state quantity of a cluster robot through a KAN network; according to the fusion feature vector, based on a reinforcement learning strategy network of a TD3 algorithm, generating a control action of each robot; the reinforcement learning strategy network based on the TD3 algorithm is trained by using a multi-target collaborative reward and priority experience playback mechanism; the multi-target collaborative reward is determined according to a navigation progress reward, an obstacle avoidance reward, a formation keeping reward and an action smoothing reward; the formation of the cluster robot is adaptively adjusted according to the environment characteristics and the navigation state in the environment information; and performing autonomous navigation obstacle avoidance according to the control action. According to the invention, the environment adaptability and cooperation efficiency of the robot cluster in a dynamic environment can be improved.

Description

Technical Field

[0001] This application relates to the field of robot swarm control, and in particular to a method, device, medium, and product for autonomous navigation and obstacle avoidance of robot swarms. Background Technology

[0002] With the rapid development of robotics technology, multi-robot systems are increasingly widely used in warehousing and logistics, disaster relief, military reconnaissance and other fields. Existing cluster control systems generally have the following technical bottlenecks: (1) poor environmental adaptability: fixed formation strategies are difficult to adapt to complex and ever-changing environments; (2) low obstacle avoidance efficiency: traditional algorithms are slow to respond in dense obstacle environments; (3) unintelligent decision-making: preset rules lack real-time response to dynamic environments; (4) weak collaboration: collaboration between robots lacks global optimization.

[0003] To address the aforementioned issues, existing solutions often employ traditional path planning algorithms or simple reinforcement learning methods. However, these methods struggle to meet the flexibility, efficiency, and robustness requirements of cluster systems in complex environments. Particularly in scenarios involving narrow passages and dense obstacles, fixed formation strategies significantly reduce the cluster's passage efficiency and safety.

[0004] Therefore, there is an urgent need to provide a new method or system for autonomous navigation and obstacle avoidance of robot swarms to solve the problems of poor environmental adaptability, low obstacle avoidance efficiency, and inflexible formation changes in existing technologies. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, medium, and product for autonomous navigation and obstacle avoidance of robot swarms, which can improve the environmental adaptability and collaborative efficiency of robot swarms in dynamic environments.

[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for autonomous navigation and obstacle avoidance of a robot swarm, the method comprising: The system utilizes multimodal sensors to acquire environmental information and state variables of the swarm of robots. These state variables include: the robot's own position, velocity, orientation angle, joint motor actuation, navigation target point and current position information, distance to obstacles, distance to other robots in the swarm, distance traveled towards the target, and maximum travel distance. The KAN network is used to extract and fuse features from environmental information and the state variables of the swarm robot to obtain a fused feature vector. Based on the fused feature vectors, a reinforcement learning policy network based on the TD3 algorithm generates control actions for each robot. The reinforcement learning policy network based on the TD3 algorithm is trained using a multi-objective cooperative reward and a priority experience replay mechanism. The multi-objective cooperative reward is determined based on navigation progress reward, obstacle avoidance reward, formation maintenance reward, and action smoothness reward. The formation of the cluster robots is adaptively adjusted according to the environmental features and navigation status in the environmental information. Autonomous navigation and obstacle avoidance are performed based on the control actions described.

[0007] Optionally, the activation function of the KAN network is: ; in, For the input feature vector, For the input vector of the th One portion, For learnable univariate spline functions, The dimension of the input feature. It is a learnable combinatorial function. The intermediate dimension. This is the index representing the intermediate dimension. The index for the input feature dimension.

[0008] Optionally, the multi-objective collaborative reward for: ; in, and These are all hyperparameters and are used to dynamically adjust the weight of rewards in each domain; As a navigation progress reward, , As an obstacle avoidance reward, , Rewards for collision avoidance between robots. , The collision penalty coefficient for the robot. It is a very small constant. For safe distance threshold, Rewards for preventing collisions between robots and obstacles. , This is the penalty coefficient for colliding with obstacles. To determine the obstacle avoidance safety distance threshold, Rewards for maintaining formation , For the ideal robot orientation, For smooth motion rewards, , and This represents the joint motor actuation values ​​at the preceding and following moments. For the orientation angle, Let be the distance between the i-th robot and the obstacle. Distance between robots in the cluster The distance traveled towards the goal. For the maximum forward distance, This refers to the number of robots.

[0009] Optionally, the adaptive adjustment of the formation of the swarm robots based on environmental features and navigation status in the environmental information specifically includes: Using formula Determine formation ; in, The width at the narrowest point of the passage. For the preset width threshold, The area of ​​the barrier-free zone. To preset the minimum open area area, The radius of curvature of the navigation path. To preset the maximum allowable curvature, Obstacle density is used to characterize the number of obstacles per unit area. To preset the maximum allowable density, the line is in a straight line. It is a triangle, and 1 is a 1-shaped shape. To disperse the formation.

[0010] Optionally, using the formula Determine the straight-line formation ; Using formula Determine the formation of the triangle ; Using formula Determine the 1-shaped formation Using formula Determine the dispersed formation ; in, For the robot's position, For robot serial number, For the ideal spacing of the robot, Let be a set of integers, indicating that the row and column indices are integer values. The robot's distribution radius represents the radius around the center. Let be the phase angle, representing the th The distribution angle of the robots This refers to the number of robots.

[0011] Optionally, the priority experience replay mechanism specifically includes: Using formula Determine the probability of selecting a set of data samples from the experience replay pool. ; Where n represents the sequence number of the extracted empirical data; Indicates the priority of the extracted empirical data; This represents a preset parameter used to adjust the priority of data sample sampling.

[0012] Optionally, the multimodal sensors include: LiDAR and vision sensors.

[0013] Secondly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the aforementioned autonomous navigation and obstacle avoidance method for robot swarms.

[0014] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned autonomous navigation and obstacle avoidance method for robot swarms.

[0015] Fourthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned autonomous navigation and obstacle avoidance method for robot swarms.

[0016] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, device, medium, and product for autonomous navigation and obstacle avoidance of robot swarms. It utilizes a KAN (Kolmogorov-Arnold Network) to extract and fuse environmental features and swarm state, and combines a reinforcement learning policy network based on the TD3 (TwinDelayed Deep Deterministic policy gradient) algorithm to generate control strategies, enabling autonomous navigation, dynamic obstacle avoidance, and formation changes for robot swarms in complex environments. Furthermore, an environment-adaptive formation change mechanism is employed within the TD3 algorithm's reinforcement learning policy network, allowing the swarm to autonomously switch between various formations such as triangles, straight lines, and L-shaped formations based on environmental information extracted by the KAN network and its navigation state. Through multi-objective cooperative reward functions and priority experience replay training, the adaptability and collaborative efficiency of the swarm in dynamic environments are significantly improved. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a robot swarm autonomous navigation and obstacle avoidance method according to one embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] In one exemplary embodiment, such as Figure 1 As shown, a method for autonomous navigation and obstacle avoidance in a robot swarm is provided, comprising the following steps S101 to S104. Wherein: S101 utilizes a multimodal sensor to acquire environmental information. State variables of swarm robots H represents the number of pixels in the vertical direction of the image, W represents the number of pixels in the horizontal direction of the image, and C represents the number of color channels; the state variables include: the robot's own position. ,speed Orientation Angle Joint motor actuation quantity Navigation target point and current location information, distance to obstacles Distance to robots in the cluster Distance to the target Maximum forward distance ;in, The maximum speed within the current cycle, where T is the cycle length, etc.; the robot's initial state variables. ; Specifically, multimodal sensors include: LiDAR and vision sensors.

[0022] S102, using the KAN network to extract and fuse features from environmental information and the state variables of the swarm robots to obtain a fused feature vector; The specific processing procedure of the KAN network is as follows: (1) Environmental characteristics extract: ; (2) Robot state characteristics extract: ; (3) Feature fusion to obtain fused feature vector : ; The activation function of the KAN network is: ; in, For the input feature vector, For the input vector of the th One portion, For learnable univariate spline functions, The dimension of the input feature. It is a learnable combinatorial function. The intermediate dimension. This is the index representing the intermediate dimension. The index for the input feature dimension.

[0023] S103, based on the fused feature vector, a reinforcement learning policy network based on the TD3 algorithm generates control actions for each robot; the reinforcement learning policy network based on the TD3 algorithm is trained using a multi-objective cooperative reward and a priority experience replay mechanism; the multi-objective cooperative reward is determined based on navigation progress reward, obstacle avoidance reward, formation maintenance reward, and action smoothness reward; and the formation of the cluster robots is adaptively adjusted according to the environmental features and navigation status in the environmental information. Specifically, feature vectors will be fused. As input to the reinforcement learning policy network of the TD3 algorithm, the corresponding robot motor control quantity is output. ; The multi-objective collaborative reward for: ; in, and These are all hyperparameters and are used to dynamically adjust the weight of rewards in each domain; This is a navigation progress reward, used to encourage robot formations to perform navigation tasks efficiently. , The obstacle avoidance reward is used to prevent collisions between robots and with obstacles. , This is a collision avoidance reward for robots, used to prevent collisions between them. , The collision penalty coefficient for the robot. To be a very small constant to avoid division by zero, For safe distance threshold, A collision avoidance reward is given to the robot to prevent collisions with obstacles. , This is the penalty coefficient for colliding with obstacles. To determine the obstacle avoidance safety distance threshold, Rewards are given for maintaining formation, which are used to encourage robots to keep their formation. , For the ideal robot orientation, This is a smooth movement reward, used to reduce energy expenditure and maintain motion stability. , and This represents the joint motor actuation values ​​at the preceding and following moments. For the orientation angle, Let be the distance between the i-th robot and the obstacle. Distance between robots in the cluster The distance traveled towards the goal. For the maximum forward distance, This refers to the number of robots.

[0024] Using formula Determine formation ; in, The width at the narrowest point of the passage. For the preset width threshold, The area of ​​the barrier-free zone. To preset the minimum open area area, The radius of curvature of the navigation path. To preset the maximum allowable curvature, Obstacle density is used to characterize the number of obstacles per unit area. To preset the maximum allowable density, the line is in a straight line. It is a triangle, and 1 is a 1-shaped shape. To disperse the formation.

[0025] In narrow passage scenarios, the formation automatically switches to a straight line, which is achieved using a formula. Determine the straight-line formation ; In open areas, the formation automatically switches to a triangular formation, i.e., using a formula. Determine the formation of the triangle ; The turning scene automatically switches to a straight line formation, i.e., using a formula. Determine the 1-shaped formation The obstacle avoidance scenario automatically switches to a dispersed formation, i.e., using a formula. Determine the dispersed formation ; in, For the robot's position, For robot serial number, For the ideal spacing of the robot, Let be a set of integers, indicating that the row and column indices are integer values. The robot's distribution radius represents the radius around the center. Let be the phase angle, representing the th The distribution angle of the robots This refers to the number of robots.

[0026] Storing experience data and prioritizing experience replay sampling, the prioritization of experience replay mechanism specifically includes: Using formula Determine the probability of selecting a set of data samples from the experience replay pool. ; Where n represents the sequence number of the extracted empirical data; Indicates the priority of the extracted empirical data; This represents a preset parameter used to adjust the priority of data sample sampling.

[0027] Continue to obtain the dimensionality-reduced state quantity at the next time step. Dimensionally reduced state quantity at the current moment Motion volume The local reward function at the previous moment As a set of empirical data Store the data in the experience pool, sample it using priority, and update the parameters of the reinforcement learning policy network of the TD3 algorithm. Specifically, the reinforcement learning policy network of the TD3 algorithm includes an Actor network, a Target Actor network, a Critic1 network, a Target Critic1 network, a Critic2 network, and a Target Critic2 network. While the Actor network and the Target Actor network share the same network architecture, their inputs and outputs differ. Similarly, the Critic and Target Critic networks also share the same network architecture, but subtle differences exist between them. Specifically, the input to the Actor network is the perceived state at the current time step. The output is the amount of motion sent to the robot's joint motors at this moment. The input to the Target Actor network is the state variable at the next time step. The output is the motion quantity sent to the robot's joint motors in the next moment. The four Critic-type networks can be divided into two parts: one part consists of Critic1 and Critic2 networks, which are completely identical; the other part consists of Target Critic1 and Target Critic2 networks, which are also completely identical. The main difference between the Critic network and the Target Critic network lies in their inputs and outputs. The input to the Critic network is the current state-action pair (…). The output of the Target Critic network is the state value assessment of the current state-action pair calculated by the neural network; while the input of the Target Critic network is the state-action pair for the next time step. The output is the target value of the state-action pair based on the next time step. The network update is as follows: (1) Parameter update process of Critic1 and Critic2 networks: Calculate the state variables at the next time step using the Target Actor network. The following action ,action The calculation formula is as follows: ; in, For the network parameters of the Target Actor network, Represents the Target Actor network; Then, based on the target policy, smooth regularization is applied to the target action. Add noise : ; ; in, Indicates the current state quantity; Indicates the parameters of the Actor network; Represents an Actor network; Represents random noise parameters; The function represents when hour ,when hour ,besides, ; This indicates that the distribution follows a normal distribution, where -b and b represent fixed parameters, and b>0.

[0028] Next, based on the idea of ​​a dual network, the state-action pair is calculated using the Target Critic network. State value target value State value target value The calculation formula is as follows: ; in, For the reward function, For the discounted rate of return, For the network parameters of Target Critic1 and Target Critic2 networks, This represents the Target Critic1 and Target Critic2 networks.

[0029] Finally, the gradient descent algorithm is used to minimize the error between the evaluated value and the target value. This updates the parameters in the Critic1 and Critic2 networks. ; in, This represents the Target Critic1 and Target Critic2 networks. and This represents the state and action quantities at the current moment. These are the network parameters for Critic1 and Critic2 networks.

[0030] (2) Actor network parameter update process: After updating the Ctitic1 and Critic2 networks by step d, the Actor network update is initiated, and the state is calculated using the Actor network. The following action : ; in, For the network parameters of the Actor network, This represents the Actor network. Note that after calculating the action, no noise needs to be added, because the goal here is for the Actor network to update in the direction of the maximum value, and adding noise is meaningless. Calculate state-action pairs using Critic1 or Critic2 networks State value assessment State value assessment value The calculation formula is as follows, assuming the Critic1 network is used: ; in, These are the network parameters of the Critic1 network. This represents the Critic1 network; Finally, the gradient ascent algorithm is used to maximize the state value assessment value. This updates the parameters in the Actor network. The reason why either Critic1 or Critic2 can be used to calculate the Q-value is mainly because the purpose of the Actor network is to maximize the cumulative expected return, and there is no need to use the minimum value.

[0031] (3) Update the network parameters of Target Actor network and Target Critic network. The TD3 algorithm uses a soft update method, where the parameter update process of the Target Actor network is as follows: ; The update process for the Target Critic1 and Target Critic2 networks is as follows: ; in, The learning rate (momentum). It is usually 0.005.

[0032] Iterative training: until the model converges.

[0033] S104, Perform autonomous navigation and obstacle avoidance according to the control action.

[0034] The method provided in this application has the following specific technical effects: 1. Significantly improved environmental adaptability: The powerful feature extraction capability of the KAN network enables the system to accurately identify environmental features; 2. Optimization of navigation and obstacle avoidance efficiency: Reinforcement learning enables optimal path planning in dynamic environments; 3. Flexible and intelligent formation transformation: The adaptive formation transformation mechanism automatically selects the optimal formation based on environmental characteristics; 4. Enhanced cluster collaboration capabilities: The centralized training and distributed execution architecture balances global optimization and local decision-making. 5. Improved system robustness: The priority experience replay mechanism accelerates convergence and improves stability.

[0035] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for autonomous navigation and obstacle avoidance in a robot swarm.

[0036] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0037] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0038] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0039] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0040] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0041] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0042] In this application, all actions to acquire signals, information, or data are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with the authorization granted by the owner of the relevant device.

[0043] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0044] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for autonomous navigation and obstacle avoidance of a robot swarm, characterized in that, The autonomous navigation and obstacle avoidance method for the robot swarm includes: The system utilizes multimodal sensors to acquire environmental information and state variables of the swarm of robots. These state variables include: the robot's own position, velocity, orientation angle, joint motor actuation, navigation target point and current position information, distance to obstacles, distance to other robots in the swarm, distance traveled towards the target, and maximum travel distance. The KAN network is used to extract and fuse features from environmental information and the state variables of the swarm robot to obtain a fused feature vector. Based on the fused feature vectors, a reinforcement learning policy network based on the TD3 algorithm generates control actions for each robot. The reinforcement learning policy network based on the TD3 algorithm is trained using a multi-objective cooperative reward and a priority experience replay mechanism. The multi-objective cooperative reward is determined based on navigation progress reward, obstacle avoidance reward, formation maintenance reward, and action smoothness reward. The formation of the cluster robots is adaptively adjusted according to the environmental features and navigation status in the environmental information. Autonomous navigation and obstacle avoidance are performed based on the control actions described.

2. The method of claim 1, wherein, The activation function of the KAN network is: ； in, For the input feature vector, For the input vector of the th One portion, For learnable univariate spline functions, The dimension of the input feature. It is a learnable combinatorial function. The intermediate dimension. This is the index representing the intermediate dimension. The index for the input feature dimension.

3. The robot swarm autonomous navigation and obstacle avoidance method according to claim 1, characterized in that, The multi-objective cooperative reward Is: ； wherein, and are hyperparameters and used to dynamically adjust the weights of rewards in each domain; is a navigation progress reward, , is an obstacle avoidance reward, , is a robot-robot collision avoidance reward, , is a collision robot penalty coefficient, is a minimum constant, is a safety distance threshold, is a robot-obstacle collision avoidance reward, , is a collision obstacle penalty coefficient, is an obstacle avoidance safety distance threshold, is a formation keeping reward, , is an ideal robot orientation, is an action smoothness reward, , and are joint motor execution amounts at previous and current time, is an orientation angle, is a distance between the ith robot and an obstacle, is a distance between robots in the swarm, is a distance towards the target, is a maximum distance towards the target, is a number of robots.

4. The robot swarm autonomous navigation and obstacle avoidance method according to claim 1, characterized in that, The adaptive adjustment of the formation of the swarm robots based on environmental features and navigation status in the environmental information specifically includes: Using formula Determine formation ; in, The width at the narrowest point of the passage. For the preset width threshold, The area of ​​the barrier-free zone. To preset the minimum open area area, The radius of curvature of the navigation path. To preset the maximum allowable curvature, Obstacle density is used to characterize the number of obstacles per unit area. To preset the maximum allowable density, the line is in a straight line. It is a triangle, and 1 is a 1-shaped shape. To disperse the formation.

5. The robot swarm autonomous navigation and obstacle avoidance method according to claim 4, characterized in that, Using formula Determine the straight-line formation ; Using formula Determine the formation of the triangle ; Using formula Determine the 1-shaped formation Using formula Determine the dispersed formation ; in, For the robot's position, For robot serial number, For the ideal spacing of the robot, Let be a set of integers, indicating that the row and column indices are integer values. The robot's distribution radius represents the radius around the center. Let be the phase angle, representing the th The distribution angle of the robots This refers to the number of robots.

6. The robot swarm autonomous navigation and obstacle avoidance method according to claim 1, characterized in that, The priority experience replay mechanism specifically includes: Using formula Determine the probability of selecting a set of data samples from the experience replay pool. ; Where n represents the sequence number of the extracted empirical data; Indicates the priority of the extracted empirical data; This represents a preset parameter used to adjust the priority of data sample sampling.

7. The robot swarm autonomous navigation and obstacle avoidance method according to claim 1, characterized in that, Multimodal sensors include: LiDAR and vision sensors.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the autonomous navigation and obstacle avoidance method for a robot swarm according to any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the autonomous navigation and obstacle avoidance method for robot swarms as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the autonomous navigation and obstacle avoidance method for robot swarms as described in any one of claims 1-7.

Images