ABSTRACT This work presents a robust adaptive tracking control strategy for a category of nonlinear systems characterized by multiple uncertainties, encompassing sensor/actuator faults, external disturbances, and fully unknown nonlinearities. The challenges lie in avoiding the circular reasoning in traditional neural network algorithms and circumventing the escalating complexity by repeating differentiation of virtual control signals. To the escalating complexity arising from repeated differentiation of virtual control signals, we introduce a series of finite‐time command filters. By means of dynamic signals with fast convergence for compensating tracking/errors and an event‐triggered scheme, the constructed controller guarantees that all closed‐loop signals are uniformly bounded, while the tracking error converges to a small neighborhood of the origin within a finite‐time instant, and the Zeno phenomenon is avoided. Lastly, simulation results for a robotic manipulator system substantiate the efficacy of the developed scheme across a spectrum of fault scenarios and initial conditions.
Yin et al. (Tue,) studied this question.