Multi-UAV path planning for connectivity-based sweep coverage

Abstract

Quality of service in wireless surveillance networks relies on two key factors: area coverage and connectivity. In this work, we investigate the spatial coverage and connectivity of a wireless surveillance network comprising of unmanned aerial vehicles (UAVs) used for surveying an area of interest. In many such applications, the data collected by the UAVs must be relayed, perhaps through multiple hops, to the base station in real-time. The goal of the network, in such applications, is to improve area coverage of the network while ensuring the UAVs are connected to the base station/sink throughout their flight. To achieve this, we propose two mixed-integer linear programming (MILP)-based formulations to plan the path of a set of UAVs. In the first formulation, we maximize the area covered by the UAVs while ensuring that they maintain their connectivity to the base station. In the second formulation, we minimize the number of movements required by the UAVs to achieve a desired coverage level while maintaining connectivity to the base station. We present extensive performance evaluation of the proposed algorithms and their comparison with path-planning approaches that do not consider connectivity constraints. We explore some of the trade-offs involved in meeting the connectivity requirement.