Simulation-Based Optimal Sensor Scheduling with Application to Observer Trajectory Planning

Abstract — Sensor scheduling has been a topic of interest to the target tracking community for some years now. Recently, research into it has enjoyed fresh impetus with the current importance and popularity of applications in Sensor Networks and Robotics. The sensor scheduling problem can be formulated as a controlled Hidden Markov Model. In this paper, we address precisely this problem and consider the case in which the state, observation and action spaces are continuous valued vectors. This general case is important as it is the natural framework for many applications. In sensor scheduling, our aim is to minimise the variance of the estimation error of the hidden state with respect to the action sequence. We present a novel simulation-based method that uses a stochastic gradient algorithm to find optimal actions. This is in contrast to existing works in the literature that only solve approximations to the original problem