Risk-sensitive control of stochastic hybrid systems on infinite time horizon (Q1570013)

This paper deals with the infinite horizon risk sensitive control problem for a hybrid system of diffusion process that depends on a mode variable modeled by a Markov chain. For this control problem, the author obtained the following results: 1. Optimality conditions and the existence of an optimal policy. 2. The relationship between the control problem and stochastic differential games. 3. Characterization of the effects of the noise input and the mode variable in the cost functional. Moreover, he showed that the equivalence relationship that holds between risk-sensitive and \(H_\infty\) control in the nonhybrid case does not hold for stochastic hybrid systems.