Greedy iterative DHP algorithm-based near-optimal control for a class of nonlinear descriptor systems with actuator saturating

The near-optimal control problem for nonlinear descriptor systems is solved by greedy iterative dual heuristic dynamic programming (DHP) algorithm. A nonquadratic performance index is developed in order to deal with the actuator saturation problem. In this way, the greedy iterative DHP algorithm can be properly introduced to solve the optimal control problem of the original descriptor system. For facilitating the implementation of the iterative algorithm, two neural networks are utilized to approximate the costate vector and compute the optimal control policy respectively. An example is given to show the effectiveness of the proposed optimal control scheme.