Adaptive NN Output Feedback Control of Switched Nonlinear Systems via Multiple Event-Triggering Communications

In this article, the problem of multiple event-triggering communications-based adaptive neural network (NN) output-feedback control is investigated for a class of switched uncertain nonlinear systems. In particular, the NNs with self-growing/pruning neurons are utilized to handle unknown nonlinearities of system. By developing backstepping, an event-triggered switched NN observer, event-triggered adaptive NN controllers of subsystems and three novel switching dynamic event-triggering mechanisms (ETMs) are constructed. Multiple event-triggering communications from sensor to controller and observer to controller and controller to actuator are thus achieved under arbitrary switchings. Naturally, more communication burdens can be reduced compared with those existing single or dual event-triggering communications methods for non-switched and switched systems. Note that one difficulty caused by dual asynchronous switchings among candidate subsystems and candidate controllers and candidate observers is overcome. Also, other difficulties caused by finding an adjustable positive lower bound of interexecution times for each ETM and the errors between continuous-time and sampled-data-based basis function vectors of NNs are overcome. A switched one-link robotic manipulator system is employed to illustrate the validity of the scheme developed.