Approximate Safety Verification and Control of Partially Observable Stochastic Hybrid Systems

TL;DR

This paper develops approximate methods for safety verification and control of partially observable stochastic hybrid systems, using finite state and Gaussian mixture models to handle noisy observations and ensure safety with error bounds.

Contribution

It introduces two novel approximation techniques for probabilistic safety verification in hybrid systems with partial observability, including error analysis and convergence guarantees.

Findings

Finite state Markov decision process approximation performs well on examples.

Gaussian mixture approach effectively represents complex information states.

Both methods provide safety probability bounds with convergence guarantees.

Abstract

Assuring safety in discrete time stochastic hybrid systems is particularly difficult when only noisy or incomplete observations of the state are available. We first review a formulation of the probabilistic safety problem under noisy hybrid observations as a dynamic program over an equivalent information state. Two methods for approximately solving the dynamic program are presented. The first method approximates the hybrid system as an equivalent finite state Markov decision process, so that the information state is a probability mass function. The second approach approximates an indicator function over the safe region using radial basis functions, to represent the information state as a Gaussian mixture. In both cases, we discretize the hybrid observation process and generate a sampled set of information states, then use point-based value iteration to under-approximate the safety probability and synthesize a suboptimal control policy. We obtain error bounds and convergence results in both cases, assuming switched affine dynamics and additive Gaussian noise on the continuous states and observations. We compare the performance of the finite state and Gaussian mixture approaches on a simple numerical example.