Joint Computation Offloading and Resource Management for USVs Cluster of Fog-Cloud Computing Architecture
Kuntao Cui, Wenli Sun, Wenqiang Sun · 2019
In recent years, unmanned surface vehicles (USVs) have made important advances in civil, maritime, and military applications. With the continuous improvement of autonomy, the increasing complexity of tasks, and the emergence of various types of advanced sensors, higher requirements are imposed on the computing performance of USVs clusters, especially for latency sensitive tasks. Therefore, this paper proposes a fog-cloud computing architecture for USVs clusters, which dynamically allocates USVs cluster computing resources and cloud computing resources to achieve overall optimized computing performance. However, due to the mobility of USVs cluster nodes, the network topology, the wireless channel states and the available computing resources are changing rapidly and difficult to predict. In this work, we develop a learning-based task offloading framework using the multi-armed bandit (MAB) theory, which enables vehicles to learn the potential task offloading performance of its neighboring team nodes with excessive computing resources, and minimizes the average offloading delay. We propose an adaptive upper confidence bound (AUCB) algorithm and augment it with load-awareness and occurrence-awareness, by redesigning the utility function of the classic MAB algorithms. The proposed AUCB algorithm can effectively adapt to the dynamic marine fog computing environment, balance the tradeoff between exploration and exploitation in the learning process, and converge fast to the optimal computing nodes with theoretical performance guarantee. Simulations under synthetic scenario are carried out, showing that the proposed algorithm achieves close-to-optimal delay performance under both heavy and light input data load.