Cooperative UAV-USV MEC Platform for Wireless Inland Waterway Communications

Unmanned surface vehicles (USVs) play a key role in inland waterway transport through intelligent IoT sensing and analytics. The embedded USV platforms, however, suffer limited computation resources. This could be tackled by offloading computation tasks from USVs to unmanned aerial vehicles (UAVs), which perform as relays and forward data to multi-access edge computing (MEC) servers. A major challenge for UAVs is that the process consumes a great deal of battery power, limiting their operational time. This paper proposes a novel cooperative UAV-USV MEC platform, where UAVs are deployed to provide communication and computation backhaul for USVs. A UAV energy consumption minimization problem is formulated by jointly considering USVs task execution mode, UAVs trajectories, UAVs arrival times and UAVs hovering coordinates. This problem is divided into three subproblems: 1) optimizing UAVs trajectories using enhanced ant colony optimization; 2) identifying the optimal USVs task execution mode selection indicators based on a surrogate LR algorithm; and 3) reducing the time between UAV arrivals and hovering coordinates by solving the joint ADMM and successive convex approximation subproblems. The results show that the proposed solution significantly reduces the energy consumption of UAVs by up to 55% compared to some baseline energy-saving algorithms.