Joint Computation Offloading and Routing Optimization for UAV-Edge-Cloud Computing Environments

Abstract

Computation offloading is significant to the UAV swarms by migrating computational tasks from the UAV swarms to the edge or cloud computing infrastructure. However, when studying the computation offloading problem, existing approaches do not fully consider the characteristics of UAV swarms and do not differentiate the characteristics of cloud computing and edge computing well. In this paper, we study a joint computation offloading and routing optimization problem for UAV swarms under an UAV-Edge-Cloud computing architecture. Our contributions can be summarized as the following three aspects. First, to fully optimize the characteristics of UAV swarms, we jointly consider the computation offloading and routing for UAV swarms. Then, to highlight the characteristics of cloud computing and edge computing, we propose a novel three-layer computing model joint computation offloading and routing problem. Finally, we design a polynomial near-optimal approximation algorithm to solve the joint optimization problem using the Markov approximation technique. Finally, our simulation results demonstrate the high efficiency of our proposed algorithm.