Leader-Follower Formation Control Of Mobile Robots With Prescribed Performance Guarantees

Multiple mobile robot systems have received considerable attention from control engineering communities due to its significant applications in many aspects,such as cooperative coverage,target searching,cooperative monitoring and exploring,robotic networks,and automated highway systems.Compared with the use of a single robot,multi-mobile robot systems have distribution characteristics,which take great advantages in task applicability,robustness and scalability.Formation control is a typical issue of multi-robot system cooperation.In order to ensure the safety and efficiency of cooperative formation,it is usually necessary to consider the transient performance of the formation control system.Among various formation control strategies,leader-follower formation architecture is preferred owing to its simplicity and scalability.Therefore,the study on the leader-follower formation control of multiple mobile robots with prescribed performance has important theoretical value and practical significance.The main contents of the thesis are outlined as follows.For multi-robot systems with uncertain parameters,Chapter 2 studies the problem of leader-follower formation control of multiple mobile robots with communication range constraints.Each mobile robot in the formation group acquires the state information with respect to its leader through communication.In practical applications,the communication range of on-board sensors is usually limited.If the relative distance between multiple robots is too far,their communication may fail.Additionally,if a collision occurs between multiple robots,it usually affects the accuracy of task execution and leads to an inefficient work.The connectivity maintenance and collision avoidance problems between neighboring robots could be described by the relative distance and bearing angle constraints: the relative distance and bearing angle between neighboring robots are required to be always within an allowed maximal range during the entire formation movement such that the reliable communication and collision avoidance are guaranteed.For the multi-mobile robot systems with uncertain model parameters,the tantype barrier Lyapunov synthesis method with symmetric boundaries and adaptive backstepping control design are applying in the design of formation controllers such that asymptotic stability of the closed-loop systems is achieved,and the relative distance and bearing angle between neighboring robots are always within a predetermined range during the formation movement.The boundaries of the predefined feasible regions are enforced by designer-specified performance functions of time to provide the preselected specifications on the transient and steady-state performances of formation errors.Finally,simulation results show the effectiveness of the formation controllers.For multi-robot systems with uncertain dynamics,Chapter 3 studies the problem of leaderfollower formation control of multiple mobile robots using visual feedback.Each mobile robot in the formation equips with on-board visual sensors to measure the relative distance and bearing angle with respect to neighboring robots without knowing the pose of a global coordinate frame,the velocity and acceleration information with respect to the leader.Firstly,parameter estimators is employed to estimate the upper bound of the leader's linear velocity,and the estimated parameters are applied in the virtual controller design.Secondly,dynamic surface control technique is employed to avoid the used of the derivative of the virtual controller.Then,adaptive neural network control technique is applied to approximate the dynamic uncertainty of the robot systems.With the help of tan-type barrier Lyapunov synthesis and adaptive backstepping method,adaptive neural network formation controller is designed such that the relative distance and bearing angle between the leader and follower are always within predetermined ranges,and the formation tracking errors satisfy the prescribed steady-state and transient performances.Simulation results show the effectiveness of the designed formation controllers.