A Constrained POMDP Formulation and Algorithmic Solution for Radar Resource Management in Multi-Target Tracking
paper Menu
The radar resource management problem in a multitarget tracking scenario is considered. The problem is solved using a dynamic budget balancing algorithm. It models the different sensor tasks as partially observable Markov decision processes and solves them by applying a combination of Lagrangian relaxation and policy rollout. The algorithm has a generic architecture and can be applied to different radar or sensor systems and cost functions. This is shown through simulations of two-dimensional tracking scenarios. Moreover, it is demonstrated how the algorithm allocates the sensor time budgets dynamically to a changing environment in a nonmyopic fashion. Its performance is com-pared with different resource allocation techniques and its computational load is investigated with respect to several input parameters.