Optimization-enhanced neuroadaptive algorithm for motor servo systems with uncertainty compensation.

Optimal control theory has been widely used in theoretical research. However, its application in practical systems is full of challenges due to modeling uncertainties. Consequently, an optimization-enhanced neuroadaptive controller for a class of motor servo systems with modeling uncertainties is developed via the command filtered backstepping framework in this paper. Significantly, a neuroadaptive uncertainty observer is employed to address these uncertainties. Furthermore, the Hamilton-Jacobi-Bellman (HJB) equation is established via a novel error subsystem. Hence, the optimization-enhanced control law can be obtained by solving the HJB equation. Finally, both simulations and experiments reflect a trade-off between tracking performance and control cost under the action of the proposed controller.