Infinite horizon backward stochastic differential equation and exponential convergence index assignment of stochastic control systems (Q1614409): Difference between revisions

A nonlinear stochastic control system \[ dx_t=b(x_t,\nu_t) dt+ \sigma (x_t,\nu_t) dW_t,\quad 0\leq t\leq T,\tag{1} \] and a linear stochastic control system \[ dx_t=(Fx_t+G \nu_t)dt+ (F_1x_t+G_1 \nu_t) dW_t,\tag{2} \] are considered. Here \(b\) and \(\sigma\) are \(\mathbb{R}^n\)-valued functions, defined on \((x,\nu)\in \mathbb{R}^n\times \mathbb{R}^r\), with certain properties; \(F,F_1,G,G_1\) are matrices. The properties of control systems to be exactly terminal-controllable (exactly controllable) are defined. Under a certain condition, the system (2) is equivalent to the system \[ dx_t=(Ax_t+A_1 z_t+Bu_t) dt+z_tdW_t,\tag{3} \] where \(z\) and \(u\) are respectively \(\mathbb{R}^n\)-valued and \(\mathbb{R}^{r-n}\)-valued control processes. Necessary and sufficient conditions for system (2) to be exactly terminal-controllable, and system (3) to be exactly controllable, are established. It is shown that exact controllability of an open-loop stochastic system is equivalent to the possibility of assigning an arbitrary exponential convergence index to the solution of the closed-loop stochastic system, formed by means of suitable feedback of the states. A sufficient and necessary condition for the existence and uniqueness of the solution to a class of infinite horizon forward-backward stochastic differential equations is provided.