Decentralized output-feedback adaptive control for a class of interconnected nonlinear systems with unknown actuator failures

In this paper, a decentralized output-feedback adaptive backstepping control scheme is proposed for a class of interconnected nonlinear systems with unknown actuator failures. By introducing a kind of high gain K-filters, a bound estimation approach and some smooth functions, the effect of actuator failures and interactions among subsystems is successfully compensated for and the actuators are allowed to change among the normal operation case and different failure cases infinitely many times. The proposed scheme is able to guarantee the global stability of the overall closed-loop system, regardless of the possibly infinite number of unknown actuator failures. An initialization technique is also introduced so that the L-infinity performance of tracking errors can be adjusted no matter if there exist unknown actuator failures. Simulation results performed on double inverted pendulums are presented to illustrate the effectiveness of the proposed scheme. (C) 2016 Elsevier Ltd. All rights reserved.