The Research On Muti-robot Formation Control For Swarm Systems With Overlapping Information Structure Constraints

Multi-robot formation control system is a kind of complex swarm system, the system not only has the characteristics of interconnected systems, but also subsystems overlap under information structure constraints.In this paper, starting with overlapping information structural constraint for multi-robot formation control ,research decentralized coordination control of such complex system. Focusing on how to design decentralized optimal controller under overlapping structure about swarm system of multi-robot formation control, thus ensuring that the stability of entire swarm formation system;mainly focused on the problem of keeping the connective stability of formation control system when the swarm model structure changed in the system,that is for one robot, because of its failure or communications on emergency situations such as missing the system or restoring robot formation control systemFirstly,This paper based on inclusion principle and the decomposition method of multi-overlapping information structure, the swarm system were decomposed two-two-subsystems which are composed by two robots,then designing a degree of stability LQR optimal decentralized controller for the robot subsystem,and in order to improve the dynamic performance, double particle swarms PSO algorithm identify the optimal system's value of the weighted matrix.Secondly, in this paper ,an improved method is present to ensure the connective stability of the swarm system during the movement of the robots when missing robot, or re-back increasing robot.The improved approach not only can be simply operated, but also guarantee the overall system's stability under optimal subsystem stability,and ensure the the entire system's optimal connective stability.This paper based on the Leader-follower coordinated approach to address different formations structures, take different coordinated approachs.For the problem of connective stability, coordinated approach which based on the role conversion and information exchange is present.In the end ,this improved method of connective stability is applied with three different structures of robot formations,through simulation results of this method show that the whole swarm system's stability doesn't be affected under model structural change, and it is simple and feasible, has better control performance.