Formation Control Of Wheeled Mobile Robots On Cooperative-Antagonistic Networks

In recent years,with the rapid advancement of robotics technology and the development of cooperative control theory,distributed cooperative control of multi-robot systems has become a research hotspot with considerable potential.As an essential branch of the distributed cooperative control of robots,the distributed cooperative formation control of robots is dedicated to designing appropriate control laws to make the robots form the desired formation to complete the specific tasks,only by using local information interaction.Compared with single-robot control and traditional multi-robot centralized control strategies,multi-robot distributed cooperative formation control has excellent advantages,such as stronger computing ability,lower communication pressure,and better fault tolerance.Thanks to these advantages,the research on multi-robot distributed cooperative formation control is widely used in many scenarios,such as ground reconnaissance,naval escort,robot inspection,etc.,and hence is quite significant and valuable.The existing research on distributed formation control of multi-robots under cooperative-antagonistic networks mainly focuses on linear and general nonlinear robots,less on non-holonomic wheeled mobile robot systems,which is still in the infancy stage.Through literature review,it is found that the existing researches on wheeled mobile robot systems all relied on input-output feedback linearization methods to directly apply linear or nonlinear cooperative control theory to the external hand-position subsystem of the robot,which has the following shortcomings:(1)The stability of the internal subsystem(attitude angle subsystem)is not considered.The attitude angle may not converge.That causes the robot to keep rotating,which will destroy the stability and hardware equipment;(2)All following robots are required to know the information of the leader,resulting in increased communication burden;(3)The attitude angle subsystem has multiple equilibrium points,so the control strategy can only achieve local stability,which is not conducive to practical applications.In order to solve the above problems,this thesis takes the wheeled mobile robots under the cooperative-antagonistic network as the research object,based on the matrix theory,graph theory,input-output feedback linearization,Lyapunov method,sliding mode control technology,distributed cooperative control,and other methods,starting from the robot mechanism model and the communication network to carry out distributed formation tracking control research.The main contents are as follows:1)The tracking control problem of a wheeled mobile robot with leader-follow mode under a cooperation-antagonistic communication network is studied in this paper.First,based on the input-output feedback linearization method,construct the hand-position output for all following robots,and construct the external hand-position subsystem and the internal attitude angle subsystem.Then the zero dynamic equation and equilibrium point of the attitude angle subsystem are computed,based on it and Lyapunov method.The convergence conditions on controller parameters,leader’s velocities,and an attraction zone are derived to guarantee the angle error is bounded and convergent.Finally,use coordinated control and the Lyapunov method to design a distributed control law that the global index of position formation error converges to an arbitrary small neighborhood of the origin.Stability analysis shows that under the condition of the communication graph is structurally balanced,the proposed control scheme can simultaneously ensure the stability of the external and internal robot systems and solves the existing research that does not consider the convergence of internal attitude angles.2)Solve the tracking and controlling the problem of the wheeled mobile robot over the cooperative-antagonistic network,on the condition of only part of the following robots can access the status information from the leader.Based on the control law proposed in the first research content,a new control law is designed by introducing a control parameter matrix and a new symbolic function.Based on the global reversible model transformation,a suitable Lyapunov function is constructed,and the stability of the system is analyzed through rigorous mathematical calculations.It is proved that the designed control law can ensure that the formation position tracking error globally converges to an arbitrary small neighborhood of the origin.Finally,further analyzing the zero-dynamic equation of the attitude angle to obtain the local convergence of the attitude angle.Under the condition of structural balance signed graph,this control law breaks through the restriction condition that all following robots need to know the information of leader,and it is completely distributed,which significantly reduces the communication burden.3)Research on the global asymptotic bipartite formation tracking control problem of the wheeled mobile robot to the pilot robot under the condition of the cooperativeantagonistic communication network.First,use variable substitution to transform the robot model into a chain system composed of attitude angle linear subsystem and nonlinear position subsystem.Then,for the posture angle segment,base on the above research content,design an angular velocity control law that can globally converge the angle tracking error to zero.After that,the sliding mode surface is constructed for the positioning subsystem,and the linear velocity control law is designed using the Lyapunov method to ensure that the formation position tracking error globally asymptotically converges to zero.The stability analysis shows that under the conditions of continuous excitation,only part of the following robots can obtain the leader’s state information,and the structural balance symbol diagram,the designed control law can realize the robot’s global asymptotic bipartite formation tracking control,which makes the stability and robustness are greatly improved.The above control laws are given the corresponding experimental simulation cases,and the effectiveness of the control strategy is verified.The research results formed have high theoretical value and practical engineering significance.