Distributed Control Of Networked Robotic Systems With Input Saturation

The problem of distributed control of networked multi-robotic systems can be regarded as the consensus of a collection of micro-robots to accomplish the overall missions via sharing information with each other. More precisely, each networked robot can be regarded as an agent, and then the multi-agent networked system will be established.In such a networked multi-agent system, coordination and cooperation can be achieved by exchanging the neighbor information of each agent. The whole system may fulfill a complex task as expected. In the existing literature, the distributed control problem of multi-agent systems has widely investigated in the field of distributed sensor networks,formation control, unmanned aerial vehicles and other aerospace and industrial production processes, etc. Due to the saturation constraint exists widely in any actual control systems,and the saturation constraint problems for a control system have been widely concerned in recent decades. Therefore, it is also very important in both theoretical significance and practical applications to study the stability of multi-agent systems with input saturation.On the basis of the existing research, this thesis studies the consensus problem for multi-agent systems with input saturation, the main contributions of this paper can be summarized as follows:(1) For continuous-time multi-agent linear systems with input saturation, we assume that each agent system is controllable. Then the consensus problem for such systems can be solved by using an algebra Lyapunov equation method. The low feedback gain matrix of the distributed control protocol can be obtained by means of the positive definite solution of the algebra Lyapunov equation. Based on the low gain feedback strategy and observer-based design approach, we design the state feedback and output feedback distributed control algorithms for multi-agent systems, respectively. It has been shown that under some mild assumptions, the whole multi-agent system with input saturation can research semi-global consensus.(2) For discrete-time multi-agent systems with input saturation, we consider the general linear discrete-time case with a leader. Under the assumption that the each agentsystem is asymptotically null controllable with bounded controls(ANCBC), we design the both state feedback and output feedback control algorithms. It is shown that under ANCBC of each agent system, the whole multi-agent system can achieve global consensus;In addition, based on the Matlab platform, the numerical simulation results are given for the proposed consensus control algorithms. For continuous-time multi-agent systems,by means of choosing different low gain parameters, the simulation results demonstrate the effectiveness of the designed low gain feedback strategy. For discrete-time multi-agent systems, the simulation results also verify the validity of the theoretical analysis.