Research On Formation Control For Multi-robot With Wireless Communication

This thesis investigates wireless communication based multi-robot formation control problem. Novel modeling and control algorithms are proposed to achieve multi-robot formation considering the unknown pose, packet loss and other practical constraints in implementing the system. We also estabilish a novel experimental platform and conduct extensive experiments to evaluate the effectiveness of our proposed methods. This study has some theoretical significance and practical value in the related areas. The main contributions of this thesis include:(1) This thesis proposes a novel distributed formation control algorithm considering the relationship between the orientation of the robot and its line speed based on the virtual structure, and theoretical analysis of the stability is given. Meanwhile, the simulation and experimental results further validate the effectiveness of this algorithm.(2) This thesis studies the pose estimation based formation control problem. The extended Kalman filter based on the position of the robot and the relative distance is used to estimate the robot’s pose without orientation measurement, and theoretical analysis of pose estimation error is given. Then combining the virtual structure control law, the parameter condition guaranteeing the stability of formation is given. Simulation and experiment further verify the validity of the estimation method.(3) This thesis studies the distributed formation control problem with packet loss. Firstly, the discrete control problem of the single robot with packet loss and without packet loss is studied. Then, the distributed multi-robot formation control algorithm with packet loss is proposed. Theoretical analysis shows that this algorithm can guarantee the stability of the formation. Simulation and experiment further verify the effectiveness of this algorithm.(4) This thesis constructs a novel and pratical experimental platform to evaluate the effectiveness of the designed control law and algorithms.