Research On Multi-robot Formation And Obstacle Avoidance Control Strategy In Dynamic Environment

With the continuous development of science and technology,the mobile robot has been applicated in different and wider areas,while the single robot is harder to meet diverse requirements.Compared with the single robot,the multiple robot system shows great advantages in formation patrol,environment exploration,assembly line production and other specific applications.In applications such as environment exploration,transportation of large objects and round-up of targets,multiple robots need to maintain a stable formation and avoid obstacles in the environment.In this dissertation,the researcher focused on the research of four-wheel differential mobile robot and the obstacle avoidance control of multiple robot.The main research contents include:(1)In this paper,the kinematics of the four-wheel differential mobile chassis has been analyzed,which lead to result of control formula for the mobile chassis.For the problem of environment perception,lidar has been used to detect the information of obstacles in the environment.Then,the nearest neighbor matching algorithm is adopted to these obstacles information clustering to get the distribution of obstacles.At the same time,the association algorithm is used to analyze the association of the vector model of obstacles obtained from the clustering of the two frames of data.Therefore,the motion information of dynamic obstacles is obtained.(2)For the obstacle avoidance problem of multiple robots,the obstacle avoidance has been researched based on single robot.Also the artificial potential field method has been applied as the obstacle avoidance algorithm.In the static environment,there are some problems about path oscillation and local minimum point in the artificial potential field method.In this dissertation,the problem of path oscillation is improved by adding Angle filter and calculating gravity with different coefficients,and the problem of local minimum point is solved by adding virtual target point.In dynamic environment,the artificial potential field based on relative time and the variable dynamic obstacle scope based on relative speed are presented in this dissertation with the conclusion of obstacle avoidance simulation for obstacles under the various moving speed conditions.(3)For the problem of multi robot formation control,this dissertation combines the leader-following method and the cluster consistency theory to design the pose ring-speed ring-cascade PID controller,and designs the master controller for the leader robot and the following robot respectively.The obstacle avoidance strategy of multiple robots is designed based on the formation transformation factor and the environment fitness function.The static obstacle avoidance and dynamic obstacle avoidance are achieved based on the artificial potential field method.(4)The robot experimental platform is introduced and divided into upper computer system and lower computer system.The lower computer system is responsible for the control of the robot and the processing of positioning information,while the upper computer system is responsible for the operation environment modeling algorithm,formation control algorithm and obstacle avoidance control algorithm.Lastly,variety experiments are designed to verify that the formation control and multiple robot obstacle avoidance has been achieved by the formation control algorithm.