Multiple Wheeled Mobile Robot Formation Control Method Based On Leader-follower Strategy

Multi-agent systems can accomplish complex tasks that cannot be accomplished by individuals through cooperation.They have many advantages such as high task execution efficiency,low energy consumption,and low risk of individual failure.Therefore,they are widely used in industrial production,transportation,military offensive and defense,aerospace and other fields.Formation control problem is the basis and key problem of multi-agent systems.This paper takes the intelligent wheeled mobile robot(WMR)as the research object,adopts the leader-follower formation strategy,and conducts an in-depth study on the coordination of the formation control system and the stability of the individual robot.Firstly,in order to accurately describe the multi-robot system,the kinematics model of robot is obtained by analyzing the non-holonomic constraints of the robot.At the same time,the dynamics model of WMR was deduced in detail by using of Euler-Lagrange energy method combining with nonholonomic constraint and Routh equation,and the dynamics model is simplified by reducing the nonholonomic constraints.And then,based on the leader-follower control strategy,the formation model including distance,azimuth and attitude angle error information is established.This formation model can simultaneously describe the formation robot position relationship and attitude relationship.Secondly,in order to realize coordinated control of multi-wheeled robot formations,the model predictive control(MPC)and linear quadratic regulator control(LQR)are applied to achieve the coordination control of robot formation.By comparing the control effect of the two methods,the MPC controller has more stable control output and has a smaller overshoot,and the comprehensive performance is better than LQR method.Thirdly,in order to improve the stability of the individual robot and the robustness of the formation system in complex environment,a double closed-loop control strategy is proposed.The outer loop applies MPC controller to realize the trajectory tracking of the leader robot,while the state constraint MPC controller is used for multi-robot formation coordination control.The inner loop take into account the uncertain friction disturbance,and the terminal sliding mode controller are used to achieve dynamic stability control.By designing the adaptive law,the real-time estimation of the friction disturbance is realized to improve the robustness of the dynamic controller.Finally,according to different application scenarios and the prevailing conditions,three leader-followers formation modes are discussed.The formation model discussed are basis SBC mode,obstacle avoidance SDC mode,multi-robots SSC mode,respectively.In view of the proposed formation mode,the switching strategy between SBC and SDC,SBC and SSC are discussed,respectively.The MATLAB/Simulink simulation platform is used to model the multiple wheeled mobile robot leader-follower formation control system,and to verify the effectiveness of the above control methods and control strategies.