Hybrid Control for Robot Swarms to Detect Critical Nodes in Heterogeneous Sensor Networks

Robot swarms have recently shown great promise in monitoring the operation of heterogeneous sensor networks (HSNs). However, current methods suffer from prohibitive search costs and inaccurate detection of critical nodes that require maintenance. In this study, we develop a hybrid control approach that integrates two control stages to efficiently identify maintenance-requiring nodes within a widely deployed HSN using robot swarms, providing reliable support for subsequent maintenance tasks. First, we introduce a coverage control strategy based on the Voronoi tessellation of the task environment, rapidly identifying areas with high signal intensity (SI) in the HSN. Next, we define the Node Influence metric to quantitatively measure the urgency of maintenance for each node based on SI and the robustness of the connections between nodes. Then, we switch to another control profile to identify nodes that potentially require maintenance. We also prove the convergence and correctness of the method. Finally, results from numerical simulation and experiments are provided to validate the efficiency, scalability, and transferability of our approach.