5G MEC-Based Intelligent Computation Offloading in Power Robotic Inspection

Abstract

Power robotic inspection plays a critical role in the realization of real-time visualization and perception of substation in power grid. 5G mobile edge computing (MEC) has emerged as a promising solution to provide the large bandwidth, wide connectivity, and proximate computing capabilities for the computation offloading of power robotic inspection with stringent delay requirements. This article proposes a 5G MEC-based intelligent computation offloading framework in power robotic inspection to cope with multi-dimension entity heterogeneity, environment dynamics, and inspection delay guarantee. Specifically, the proposed framework and the implementation procedures of computation offloading are firstly elaborated, and the research challenges are outlined. Then, we propose an artificial intelligence (AI)-enabled multi-dimension collaborative optimization algorithm of route planning and task offloading to address the low-latency computation offloading problem under queue stability constraint. A case study is provided to verify the superiority of delay and queue backlog performance through simulation results.