An Energy Sensitive Computation Offloading Strategy in Cloud Robotic Network Based on GA

Abstract

Cloud robotic network (CRN) normally contains multiple mobile robots and a cloud computing center providing feasible solutions for many multiagent applications. One of the most critical issues in CRN and its application is how to effectively assign/offload computational tasks. This paper presents a novel energy sensitive task offloading strategy to answer the question particularly for CRN. First, we propose a novel strategy to offload tasks to cloud center, as well as other robots to greatly improve the computing ability and execution efficiency. Second, an energy sensitive model is developed to balance the energy level of the robots and eventually prolong the lifetime of the robot network. A modified genetic algorithm (GA), named energy sensitive GA, is finally developed and integrated into the strategy to get the optimized task offloading result as soon as possible, which is critical to most CRN applications. The correctness, efficiency, and scalability of the proposed strategy are proved with both theoretical analysis and experimental simulations. The evaluation results show that the proposed method can effectively assign tasks and prolong the lifetime of the network to a certain extent.