Computation offloading algorithm for cloud robot based on improved game theory

Abstract

How to use the resources of the edge cloud more reasonably, reduce the energy consumption of machine equipment and ensure the shortest time for task completion are the challenges faced in cloud robot computation offloading research. In this paper, multiple heterogeneous cloud robot computing offloading problems are converted into game-type problems, and the computation-intensive tasks are divided to achieve partial offloading of tasks. An improved distributed game theory algorithm is designed to make each cloud robot's computation offloading strategy reaches the Nash equilibrium state, which maximizes the benefits of multiple participants, reduces the network load pressure of the central cloud, and reduces the transmission delay of computation offload. Simulation results show that the improved distributed game computation offload algorithm proposed enables cloud robots to reduce local computing energy consumption and shorten the average task completion time, greatly improving the edge cloud service quality.