Sensor Scheduling Under Action Dependent Decision-Making Epochs

Abstract

In this paper, we consider the problem of optimally allocating sensing resources for maximizing information gained in a multi-target tracking scenario. In particular, we examine the optimal allocation of a single ground-based sensor to multiple space-based targets to maximize the information gained in the space situational awareness problem. The optimization problem is solved in a receding horizon fashion at action dependent decision epochs that are not assumed to occur at regular intervals. We use a parallel Markov Chain Monte Carlo algorithm to compute the optimal target assignment sequence under constraints posed by the dynamics of the sensor. Information gain is quantified in terms of the differential entropy of the state probability density function. The effectiveness of the approach is demonstrated through a simulation study.