Adaptive optimal control for linear stochastic systems with additive noise

In this paper, an optimal control design scheme is proposed for continuous-time linear stochastic systems with unknown dynamics. Both signal-dependent noise and additive noise are considered. A non-model based optimal control design methodology is employed to iteratively update the control policy online by using the system state and input information. A new adaptive dynamic programming algorithm is developed, and the convergence result for the proposed methods is presented. The effectiveness of the obtained method is also illustrated by a practical simulation example of 2-DOF vehicle suspension control system.