Multi-agent Formation Control And Obstacle Avoidance Research

The formation control of multi-agent systems has been widely favored by scholars all over the world because of its wide range of fields and powerful practical application capabilities.As the basic problem of multi-agent collaborative control,which has attracted more and more attention.The main problem at present is how to divide,reconstruct,maintain the formation and avoid obstacles in the complex environment of static or dynamic obstacles,and then safely move to the target location to complete complex tasks.With the continuous improvement of theoretical research on multi-agent formation control and the increasing application technology,the formation control and obstacle avoidance problems have become the hot trend of multi-agent formation research.The algorithms involved in multi-agent formation control and obstacle avoidance also have their own advantages and disadvantages,which require constant optimization and improvement.The main work is summarized as follows:(1)Aiming at the obstacle avoidance problem of multi-agent formation in dynamic environment,a method of formation obstacle avoidance based on artificial potential field method and cuckoo search algorithm is proposed.In the heterogeneous mode of the dynamic formation transformation strategy,the artificial potential field method is used to plan obstacle avoidance for each agent in the multi-agent formation;In view of the limitation of the attractive increment coefficient and the repulsion increment coefficient set by the artificial potential field method,the adaptation environment is randomly searched by the idea of L?vy flights in the cuckoo search.Simulation experiments show that the proposed method can effectively solve the multi-agent formation obstacle avoidance problem in complex environment,and use the efficiency function to evaluate and analyze the experimental data to verify the rationality and effectiveness of the optimization method.(2)Aiming at the problem of multi-agent formation control in unknown environment,a formation control method combining fuzzy controller and leader-follower method is proposed.Firstly,using the leader-follower method to control the shape of the entire multi-agent formation,the artificial potential field method navigates the leader agent in the formation toward the target position;secondly,using the designed fuzzy controller to control the follower agents to track the leader agent.At the same time,the relative distance between the follower agents and the leader agent is kept.When the unknown obstacles are encountered,the shaped of the multi-agent formations is maintained and obstacle-avoided in time.Finally,in view of the lack of feedback on the leader-follower method,the loss of leader agent leads to the formation control failed.The virtual leader is added to the formation.The position of the virtual leader is consistent with the leader agent,so that the entire multi-agent formation remains in shape.The simulation results of MATLAB show that the proposed method can effectively maintain the formation and avoid obstacles in the unknown environment,and verify the rationality and effectiveness of this method.(3)In the unknown environment,the multi-agent formation obstacle avoidance problem,for the limitation of the attractive increment coefficient and the repulsion increment coefficient setting of the artificial potential field method,the fuzzy controller with the adjustment coefficient is designed to give the gain coefficient of the adaptive environment,and the performance of artificial potential is enhanced.The field method adaptability also prevents the shortcomings of the local minimum value.The simulation results show that the optimization method is reasonable and effective.