Optimal Control Performance Monitoring under Uncertainty

LUIS AVILA; E. C. MARTÍNEZ

Buenos Aires

Congreso; 24º Congreso Argentino de Control Automático; 2014

AADECA

The increasing trend towards delegating complex tasks to autonomous artificial agents in safety-critical socio-technical systems makes its behavior monitoring of paramount importance. In this work, a probabilistic approach for on-line monitoring using optimal action selection and twin Gaussian processes (TGP) is proposed. A Kullback-Leibler (KL) based metric is used to characterize the deviation of an optimal controller (modeled as a stochastic process) from its specification. The control system specification is obtained using Linearly Solvable Markov Decision Processes (LSMDP) whereby the Bellman equation is made linear through an exponential transformation such that the optimal control policy is obtained in an explicit form.