Optimal control of piecewise deterministic markov process

From MaRDI portal





A piecewise deterministic Markov process which is characterized by its behaviour between the jumps, \(x_ t=f(x_ t),\) and by the jump behaviour given by the jump rate \(\lambda\), the jump height q in the interior and the jump height p on the boundary of the state space, is controlled by strategies depending on the initial state of the deterministic motion and on the time spent since the starting point. The aim is to minimize continuous, jump and terminal costs. An optimality condition is given in the form of a limiting Hamilton-Jacobi-Bellman equation, the existence of optimal controls is derived, and by considering the original problem as a convex mathematical programming problem, the value function is characterized as supremum of smooth subsolutions.




Cited in
(71)








This page was built for publication: Optimal control of piecewise deterministic markov process

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3682371)