Event-triggered Consensus Control Of Multi-agent Systems And Distributed Cooperation Of Multiple Lagrangian Systems

This thesis focuses on the consensus problem of linear/nonlinear multi-agent systems(MAS’s). Some research on consensus control algorithms is carried out based on event-trigger and distributed cooperative control.In order to solve the continuous communication problem existing in consensus control of linear continuous-time homogeneous MAS’s, the distributed cooperative control algorithms are studied based on event-triggered control strategy. An event-triggered prediction state feedback consensus controller(ETPSFCC) is proposed to guarantee the states of agents to achieve consensus, which only relies on communication on some discrete time instants. The closed-loop system does not exhibit Zeno behavior. To remove the limit on continuous state-monitoring of each agent, a self-triggered prediction state feedback consensus controller(STPSFCC) is proposed. An event-triggered prediction output feedback consensus controller(ETPOFCC) is further proposed to solve the consensus problem in the case with output feedback.An ETPSFCC is proposed for consensus of linear discrete-time homogeneous MAS’s solving the periodic communication problem. The event-triggered prediction state feedback consensus control algorithms are studied to drive the states of agents to reach consensus with no Zeno-like behavior. Under the proposed controller, the periodic communication is replaced with the communication on some sampled time instants.The STPSFCC and ETPOFCC are also proposed for discrete-time MAS’s to remove the periodic monitoring on the agents’ states and to deal with the output feedback consensus problem.The event-triggered consensus for linear heterogeneous MAS’s is also considered.An event-triggered prediction feedback consensus controller(ETPFCC) is proposed for agents without disturbances based on internal reference model and state observer to make the outputs of agents reach consensus. To deal with the case with external matchable disturbances, a switching ETPFCC is studied based on state feedback to achieve output consensus with no Zeno behavior exhibited.The event-triggered consensus of multiple Lagrangian systems is investigated. Based on auxiliary variable measurement error and triggering condition, an event-triggered consensus controller(ETCC) is proposed for multiple Lagrangian systems to achieve the consensus of generalized position vector and the asymptotical stability of generalized velocity vector with no Zeno behavior exhibited. The ETCC could save the communication resource to a great extent. Considering unknown constant parameters,the ETCC for multiple robot systems is further proposed, which guarantees the consensus of position and the asymptotical stability of velocity by communication on ?nite discrete instants.The distributed attitude tracking of multiple spacecraft based on Lagrangian model is considered. The research begins with a typical system with Lagrangian dynamicsmultiple spacecraft. A distributed discontinuous adaptive controller is investigated for each follower spacecraft to guarantee the attitude tracking error to converge to zero asymptotically, where the leader’s input is nonzero, bounded, and unknown to the followers. To tackle the e?ect of the discontinuous controller, a distributed continuous adaptive controller is further proposed based on σ modi?cation. A robust attitude tracking controller is studied to deal with the case with external disturbances and unmodelled dynamics. After converting the spacecraft attitude model to generalized Lagrangian model, the adaptive tracking problem of multiple Lagrangian systems is solved.The distributed containment control of multiple Lagrangian systems is investigated. Based on distributed sliding mode estimators, a non-fully and a fully distributed adaptive sliding mode control algorithm are proposed, where the latter removed the requirement of followers’ knowing about the upper bound of the leader and other global information. The proposed controller can solve the distributed containment control problem of multiple Lagrangian systems with parameter uncertainties and external disturbances under undirected graphs.