An Iterative Algorithm for the Optimal Control of Itô Stochastic Systems with Markovian Jump
Jiayue Tian, Xueyan Zhao · 2021
This paper investigates the optimal control problem of Itô stochastic systems with Markovian jump. By using Newton's method and values of the current step with Kronecker product, a new iterative algorithm is proposed for solving the corresponding stochastic coupled Riccati matrix equations (SCAREs). Based on some simple conditions, a monotonic convergence theorem is established, and an initialization method is proposed. Furthermore, a simulation example is presented for demonstrating the superiority of our algorithm by comparing to the existing iterative algorithms.