Energy-Aware Usv-Uav Cooperative Task Offloading Optimization in Water Monitoring System

Traditional underwater monitoring systems often suffer from high latency and excessive energy consumption when transmitting sensor data to the cloud for processing. To address these limitations, we propose an intelligent water monitoring framework that integrates Mobile Crowd Sensing (MCS) and Mobile Edge Computing (MEC). Unmanned Surface Vehicles (USVs) in this paradigm collect data from submerged sensors while navigating across water surfaces, and Unmanned Aerial Vehicles (UAVs) equipped with edge computing servers in the MEC architecture act as airborne computing nodes, facilitating real-time task offloading and data analysis. The objective of this paper is to minimize overall system delay and energy consumption through joint optimization of USV path planning, UAV-assisted task offloading, and computational resource allocation. To reduce the complexity, we decompose the problem into two interdependent subproblems: a USV path planning subproblem solved using an Ant Colony Optimization (ACO)-based approach, and a joint task offloading and resource allocation subproblem for UAVs addressed via a hybrid strategy combining deep reinforcement learning (DRL) and Sequential Least Squares Programming (SLSQP). The simulation results show that our approach outperforms existing methods in terms of both feasibility and effectiveness, achieving significantly lower latency and energy usage in smart water monitoring applications.