Session 1: Multi-Agent Systems

Patrolling strategies for surveillance environments must often accommodate prioritization of certain critical locations. This work presents a comprehensive exploration of the Priority Patrol Problem, advancing both single-agent and multi-agent paradigms. In the single-agent setting, we characterize the theoretical bounds of feasibility by deriving a tight lower bound on the minimum allowable Time Period—the maximum permissible interval between successive visits to priority nodes. We further prove the existence of patrol strategies with recurring circuits and develop a 2α-approximate algorithm that constructs feasible patrol schedules while guaranteeing Time Period constraints. Expanding this framework, we introduce the Time Period Based Patrolling (TPBP) algorithm for multi-agent systems, which dynamically assigns Valid Walks to each agent using a centralized decision-making protocol. This method accommodates real-time traffic conditions and minimizes the average graph idleness while respecting priority node constraints. Both approaches are validated through extensive simulations, including grid-based graphs and SUMO-based urban models, demonstrating superior performance compared to traditional strategies in maintaining surveillance efficacy under resource constraints.

Geofencing control plays a critical role in robotics applications where the motion of robots must be constrained within predefined spatial boundaries. This talk will focus on the problem of stabilizing a unicycle robot around a desired circular orbit while confining its motion within nonconcentric circular geofences. Our solution approach relies on the Mobius transformation that, under certain practical conditions, maps two nonconcentric circles to a pair of concentric circles, and hence, results in uniform spatial motion constraints. Depending upon the choice of such a Mobius transformation, we show that the problem can be formulated either as a trajectory-constraining problem or obstacle-avoidance problem in the transformed plane. Exploiting the idea of the barrier Lyapunov function, we propose a unique control law that solves both these contrasting problems in the transformed plane and renders a solution to the original problem in the actual plane. Simulation and experimental results demonstrate the effectiveness of the proposed method in safety-critical multi-robot settings.