Integrated planning of constraint sensor management and patrolling

This paper proposes a multi-purpose planning scheme that effectively solves patrolling and constrained sensor planning problems for large-scale multi-agent systems. This situation occurs when some mobile-robots are performing a patrolling mission, and, at the same time, they gather visual data from predefined objects in the map. The proposed technique does not directly tackle the merged problem with the complicated cost function. Instead, it suggests a sequential scheme which allows compromises between trajectory length and observation quality. To handle large-scale systems, it offers a polynomial time approximation scheme to solve the corresponding constraint visual sensor management to achieve near-optimal planning. The simulation confirms the effectiveness of the proposed multi-objective multi-agent planning scheme on realistic size systems.