Neuroadaptive control of robotic systems with intermittent state feedback and triggering adaptation

This paper focuses on the tracking control of nonlinear uncertain robotic systems under a double event-triggering mechanism, where intermittent state feedback and triggering adaptation are simultaneously considered in the design of the control scheme, thereby significantly saving communication resources from sensors and parameter estimators to the controller and reducing computational burden. Since the triggered states are discontinuous, the standard backstepping design is no longer applicable because the partial derivatives of the virtual controllers do not exist. To avoid such technical difficulties, we first propose a conventional neuroadaptive control scheme for robotic systems, then replace the original variables with triggered variables based on this control structure, and thus a feasible neuroadaptive control with double event-triggering mechanism is developed by combining an important lemma established in this paper. In the control synthesis, we not only improved the setting of the triggering mechanism for vector variables to facilitate the implementation of the controller, but also provided a new formulaic proof when proving that the proposed method does not involve Zeno behavior, and also eliminated the potential singular value problems in the stability analysis. Numerical simulation verified the effectiveness of the developed algorithm.