Impulsive fault-tolerant adaptive control on second-order nonlinear multiagent systems by extended observer approach

This paper investigates the consensus of multiagent systems based on impulsive control with actuator faults. A typical fault that the actuator has partial loss of effectiveness (PLOE) with multiple modes is taken into consideration. Two main adaptive control schemes are designed to solve the tracking consensus problem with second-order nonlinear agent model. On the one hand, to estimate the PLOE mode, an asynchronous impulsive adaptive scheme using full system information is presented and the consensus proves to be achieved under sufficient conditions. On the other hand, by considering each agent as an input–output system, an impulsive adaptive scheme with extended observers is introduced with only output information available. The consensus first proves to be achieved by a method using time-varying linear matrix inequalities (LMIs) theoretically. However, it is difficult to obtain certain solutions of the time-varying LMIs for condition checking. Then, the feasibility analysis is made and an alternative scheme is proposed to deal with this issue instead. Numerical simulations are presented to support the theoretical results.