Optimal Coverage Control and Stochastic Multi-Target Tracking *

We address the problem of simultaneous coverage control and stochastic, multi-target tracking with a single pursuer. We presume linear dynamics for the pursuer and linear stochastic dynamics for the targets. The pursuer is equipped with two sensors of varying fidelity: broad-range and narrow-range. We seek the optimal trajectory for the pursuer, as well as optimal sensor selection, over a finite time horizon. We formulate the problem as a mixed-integer program with quadratic constraints, and exploit a convex relaxation method to enable fast solution of local minima. We demonstrate our approach on several simulated scenarios.