A solvable stochastic control problem in hyperbolic three space

Smoothness and regularity of solutions to PDEs (35B65) Dynamic programming in optimal control and differential games (49L20) Diffusion processes and stochastic analysis on manifolds (58J65) Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Optimal stochastic control (93E20)

Review text

The author considers a stochastic optimal control problem in hyperbolic three space, and explicitly solves it, by finding a smooth solution of the dynamic programming equation. More precisely, the stochastic system in \(H^ 3\) is described by the following stochastic differential equations in the geodesic polar coordinates \[ dX^ 1_ t=\frac{\cosh X^ 1_ t}{\sinh X^ 1_ t}dt+\frac{U_ t}{\sinh X^ 1_ t}dt+dW^ 1_ t, \] \[ dX^ 2_ t=\frac{1}{\sin X^ 3_ t\quad \sinh X^ 1_ t}dW^ 2_ t,\quad dX^ 3_ t=\frac{\cos X^ 3_ t}{2\quad \sin X^ 3_ t\quad \sinh X^ 1_ t}+\frac{1}{\sinh X^ 1_ t}dW^ 3_ t \] where \(t\in [0,T]\), \(X(0)=(X^ 1_ 0,X^ 2_ 0,X^ 3_ 0)=\alpha\), \(X^ 1_ 0\neq 0\), and \((W^ 1_ t,W^ 2_ t,W^ 3_ t)\), is a standard three-dimensional Brownian motion and the cost function is: \[ J(U)=E_{\alpha}\int^{T}_{0}(a\quad \cosh | X(t)| +\frac{| U(t)|^ 2}{\cosh | X(t)|})dt. \] Then there exists an optimal stochastic control \(u^*\) that in geodesic polar coordinates is: \(u^*(s,x)=- g(s)\) cosh\(| x|\) where \(s\in [0,T]\), \(x\in H^ 3\) and g is the unique positive solution of the Riccati equation: \(g'+3/2 g-1/4 g^ 2+a=0\), \(g(T)=0\).

Cites work

Cited in

(3)

This page was built for publication: A solvable stochastic control problem in hyperbolic three space

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q579213)