Robust Adaptive Optimal Tracking Switching Control of Nonlinear Systems Using Embedding Transformation and Approximate Dynamic Programming

For unknown continuous-time nonlinear systems with multiple equilibrium points, this article proposes a robust adaptive optimal tracking switching controller. The controller integrates multiple nonlinear approximate optimal tracking controllers, an optimal switching sequence, and a robust compensator. First, multiple nonlinear neural network models are established to construct a controller design model, whose unknown parameters are online estimated by resorting to Lyapunov stability theory. Then, at every instant, the optimal switching sequence and approximate optimal tracking controller are obtained by using embedding transformation and approximate dynamic programming technology. After that, according to the error between the nonlinear system and the given input, a robust compensator is designed. Finally, stability and convergence of the control system are verified, and the simulations are conducted to validate the effectiveness and superiority.