Research On Distributed Coordinated Tracking Control For Multi-Manipulator Systems Under Intermittent Communication

With the development of artificial intelligence and computer technology,the research and application of robot technology have been widely extended from industrial fields to aerospace,national defense security,national security,medical rehabilitation,social services and other fields.The State Council issued “Made in China 2025” in 2015,and the core of the plan is to make China move from “Made in China” to “Intelligent Manufacturing in China” and eventually achieve “Created in China”,which will promote Chinese’s robot industry to move towards medium and high-end in the future.With the rapid development of automation technology,a single robotic manipulator possesses certain limitations in terms of information acquisition and processing,control and operation capabilities,especially in the face of huge assembly work,handling heavy or soft objects,installation or repair of load parts and other complex tasks and changing work environments.Therefore,a multi-manipulator system composed of two or even more manipulators is needed to accomplish complex tasks that cannot be done by a single robotic manipulator through information interaction,resource and task allocation,and coordinaiton cooperation.Compared with a single robotic manipulator,in multi-manipulator systems,not only a single robotic manipulator has certain abilities of perception,communication,calculation and execution,but also complex work tasks and goals can be realized through the coordination and cooperation among neighboring manipulators.The research of distributed control for multi-manipulator systems is more difficult and challenging due to the disturbance of communication network,the change of topology and so on.In this Ph D dissertation,based on the multi-agent theory,we study the distributed coordinated tracking control problem for multi-manipulator systems under intermittent communication by fully considering the uncertainty of parameters and whether the structures are the same under complex network conditions,and the problem belongs to the technical field of intelligent electromechanical equipment and control.The main research contents include:1.The dynamics modeling of the multiple single-link and multiple double-link manipulator systems using multi-agent theory and Lagrange method is introduced.At the same time,the mathematical tools,theoretical knowledge of algebraic graph theory,matrix theory and Lyapunov stability theory are described,which provide the theoretical basis for further research contents.2.For the linearized multiple single-link manipulator systems,a new distributed switching controller is proposed,the optimal multi-step design algorithm is constructed to design the state feedback gain matrix,coupling gain and communication rate of the controller,and the stability of the controller is demonstrated by using the stability theory of the switching system.Based on these,a new distributed switched controller is proposed by considering the co-existence of intermittent communication and nonlinear dynamics in multiple single-link manipulator systems.The novel multi-step design algorithm is proposed to calculate the parameters and determine the communication rate by processing the nonlinear term through Lipschitz condition.Furthermore,it is proved that the system is stable when the communication rate is greater than a certain threshold using switching system theory.3.The distributed coordinated tracking problem in multiple single-link manipulator systems oriented to intermittent communication and information constraints is researched with considering dynamically switching communication network topologies.First,a state observer is designed to estimate the actual motor angular velocity and linkage angular velocity of the following robotic manipulator;Then,adjustable heterogeneous coupling weighting parameters are introduced based on the dynamically switching communication network topologies,and the distributed coordinated tracking control law based on intermittent communication and state observer is designed,and its stability is analyzed;Based on this,the above proposed distributed coordinated tracking control method is improved and extended to a more extensive case,i.e.,the intermittent communication method is extended from two interval methods of normal communication and communication interruption to three interval methods of normal communication,communication interruption between the leader and following manipulators,and communication interruption of all manipulators.Finally,simulation experiments are conducted for the multiple single-link manipulator systems to verify that the distributed tracking control law can not only achieve the tracking control objective,but also solve the constraints of intermittent communication and unmeasured state information.4.The distributed output consensus tracking control problem is solved for the heterogeneous multi-manipulator systems with no information constraint and subject to information constraint under intermittent communication is considered.Firstly,a distributed coordinated tracking controller based on intermittent communication is proposed without considering information constraint,and a multi-step design algorithm is constructed to solve the controller parameters;And a virtual system is established to reconstruct the leader’s state;Then,the effectiveness of the control method is demonstrated using Lyapunov stability theory.Secondly,to address the combined constraints of intermittent communication and state unpredictability,a virtual system based on the optimal control method is first established to realize the estimation of the leader state matrix and reconstruct the leader state information;Then a state observer is constructed to estimate the actual state of the following robotic manipulator;And a new multi-step design algorithm is proposed to design the parameters of the switched distributed output tracking controller;Finally,the stability analysis proves that the controller designed in this Ph D dissertation can effectively solve the communication constraint and information constraint problems of heterogeneous multi-manipulator systems and achieve the output consensus tracking goal.Moreover,this method can be also extended to heterogeneous multi-agent systems with different parameters,structures and system matrix dimensions.5.The distributed coordinated tracking control problem is solved for multiple double-link manipulator systems respectively with a stationary leader and a dynamic leader under intermittent communication.First,a distributed coordinated tracking controller is proposed for the multi-manipulator systems with a stationary leader;Then,a multi-step algorithm is constructed to design the controller parameters and prove the stability of the system by Lyapunov stability.For the multi-manipulator systems with a dynamic leader,firstly,the distributed estimator is designed to estimate the unknown velocity information of the robotic manipulator;Secondly,the distributed coordinated tracking controller based on the state estimator is designed,and a new multi-step design algorithm is proposed to construct the controller parameters;Finally,the correctness of the controller is proved through simulation experiments and analysis.Moreover,this method is applied to the double-link manipulator platform developed in the laboratory,and the experimental results show that the distributed tracking controller designed in this Ph D dissertation can effectively solve the distributed coordinated tracking problem of multiple double-link manipulator systems under intermittent communication.