Consensus Control Of Multi-agent Systems With Guaranteed Performance

In recent years, with the development of science and technology, multi-agent system(MAS) coordination control technology has been widely applied in military and civilian, such as spying and target attack, intelligent highways, environmental monitoring, and explore the unknown environment. Many scholars devote to researching consensus control which is the most basic and fundamental problem in MAS coordination. Leader-following consensus control has been researched deeply, but most algorithms can only guarantee the steady state performance, cannot guarantee the transient performance. In the actual process, systems require good transient performance, namely fast convergence, small overshoot and small steady-state error. In addition, actuator fault will deteriorate system performance and even make the system crash. In order to solve above problems, the consensus algorithms with guaranteed performance are studied, we put forward three consensus algorithms with guaranteed performance. The main contributions of this paper can be summarized as follows:(1) The leader-following finite-time consensus problem for second-order nonlinear MAS on directed communication topology is considered. Only a subset of the followers has access to the leader system, so a velocity observer for each follower is designed. To deal with the unknown nonlinear function, adaptive law is designed. By using the technique of adding-a-power-integrator, a variable structure control method is proposed. It is proved that the proposed control scheme guarantees that the tracking errors converge to an arbitrary small bound in finite time. In the end, an example is provided to show the effectiveness of the proposed control scheme.(2) The unknown high-order nonlinear MAS in output-feedback form with guaranteed transient performance under a directed graph topology is considered. To ensure the transient performance of the tracking error, a prescribed performance bound is presented. Based on the dynamic surface control and the universal approximation property of neural network, the distributed controllers are derived. It is proved that the cooperative tracking errors are restricted by the prescribed performance bound all the time. A simulation example is provided to illustrate the performance of the proposed control scheme.(3) The MAS consensus control of unknown nonlinear low-triangle dynamics and undetectable actuation failures is considered. Virtual parameter estimation error is introduced to construct Lyapunov function, in order to solve the unknown nonlinear function, time-varying parameters and undetectable actuator failures. Through DSC technique the robust adaptive fault-tolerant control agreement of MAS is presented. In the end, simulation results illustrate and verify the benefits and effectiveness of the proposed scheme.