Multi-Agent Systems Coverage Control in Mixed-Dimensional and Hybrid Environments ?

摘要

In this paper, we address the problem of deploying a team of agents over a given environment. The environment we consider here is different from most of the existing works and includes mixed-dimensional and hybrid cases. To deploy the agents, we first find the geodesic Voronoi partitions from the agents’ current locations. Then, we let each agent move to the “center” of its partition; this would minimize the worst-case response time for any agent to arrive at an arising event inside the workspace. Due to the non-convexity of the environment under study, it is natural to use geodesic distance instead of the generally used Euclidean distance as the distance measurement metric to keep the trajectory of each agent inside its admissible space. We present algorithms for agents deployment and show that the proposed approaches ensure that each agent converges to the optimal position quantified by a cost function in terms of the geodesic distance. Finally, the results of the simulation study are presented to demonstrate the effectiveness of the distributed deployment algorithms.