Model Free Adaptive Control Based Robot Formation Methods

With the rapid development of robotics,multi-robot systems have received attention in the fields of production line,formation patrol and military counter-terrorism.Multi-robot system is a complex MIMO nonlinear discrete-time system.Traditional formation control methods include model predictive control,feedback linearization control,robust control and PID control.The first three of these methods are model-based control schemes and mostly rely on accurate mathematical models of the controlled system.Therefore,it is difficult to achieve good results in the actual control system.PID control algorithm performs poorly in dealing with nonlinear systems.The Model-free adaptive control(MFAC)method,which has emerged in recent years,requires only the input and output data of the controlled system for controller design and analysis.It is free from the dependence of traditional control methods on accurate mathematical models.Therefore,the MFAC is of great significance in solving the formation control problem of multi-robot system.The main work of this thesis includes:(1)The leader-follower,which has a simple structure and is easy to implement,is chosen as the research method for formation control.Based on this,a model free adaptive formation control(MFAFC)scheme is designed,including multi-robot formation control algorithm,parameter matrix estimation algorithm and parameter matrix reset algorithm.A comparative study of robot formation simulation using the MFAFC scheme and the PID formation control scheme was conducted in the Gazebo simulation environment.The error curves and performance indicators show that the MFAFC scheme has higher accuracy and faster response.(2)To further improve the accuracy,a model free adaptive predictive formation control(MFAPFC)scheme is designed,including multi-robot formation control algorithm,parameter matrix estimation algorithm,parameter matrix prediction algorithm and parameter matrix reset algorithm.A comparative simulation study was conducted using MFAPFC scheme,MFAFC scheme and PID scheme in Gazebo simulation environment.Error curves and performance indicators show that MFAPC scheme has higher accuracy than MFAFC scheme and PID formation control scheme.(3)An experimental platform for robot formation control based on ROS development was built.The MFAFC scheme and MFAPFC scheme were applied to a multi-robot system.The experimental results show that the two formation schemes are feasible in practical systems.