Distributed Optimal Consensus Control Of Multi-agent Systems With Actuator Saturation

In the recent years,with the development of science and technology,the research on modern control technology gains a lot of attention and it has many practical applications in engineering field.However,with the scale of engineering application growing and complex,the deficiencies of traditional centralized control in these large projects have been gradually emerged.At the same time,the advantages of distributed control is that it could solve these problems with the agents collaborating with each other.Gradually,the distributed multi-agents control has been a hot spot in control field.As the optimal consensus problem is a basic problem of distributed control,there are still some problems need to be considered if these theoretical researches hope to be applied in engineering projects.The actuator saturation is one of the problem that need to be considered so that this paper mainly considers the distributed optimal consensus problem of multi-agent systems with actuator saturation.The main contributions of this paper can be summarized as follows:This paper considers the distributed optimal consensus problem of multi-agent systems with actuator saturation and uses the event triggered stategy to reduce the unnecessary communica-tions among the agents.With the event triggered strategy being considered,the frequency of the communications among agents can be reduced and the requirement on communication band-width can also be reduced.According to the structural characteristics of the system,the event triggered conditions can be set.Once the conditions are met,the agents are allowed to send their status information to other agents.Accroding to the event triggered conditions,the Zeno behav-ior is proved to be avoided.With the Lyapunov stability theory,it is proved that the distributed systems can be converge to the optimal consensus point.At last,the proposed control algorithm is proved to be effectiveness with the simulation results in MATLAB.When considering the distributed optimal consensus preblem of multi-agent systems,there is a problem that it is difficult to measure the system status in the actual physical environment.In order to solve this problem,this paper utilizes the high-gian observer to estimate the system status.Compared with other kinds observers,the design of the high-gain observer does not need to compute differential equations.The high-gain observer just need to choose an appropriate gain value,then an output feedback controller with similar performance of state feedback can be constructed.Through the selected high-gain observer parameters,it can be analyzed that the system under the control algorithm finally reach the steady state when the observation error converges to 0,which means the multi-agent systems could converge to optimal consensus point.The results are verified by simulations in Matlab.