Internet of Vehicles calculation unloading method and system based on multi-objective reinforcement learning

Abstract

The invention discloses an Internet of Vehicles calculation unloading method and system based on multi-objective reinforcement learning, and the method comprises the steps: carrying out the multi-objective optimization of a cost model according to the cost of a vehicle calculation unloading process through employing an RMDDQN-Learning method, and obtaining a Pareto optimal solution of the multi-objective optimization; optimizing a plurality of optimization targets involved in the calculation unloading process of the vehicle at the same time, and learning the decision value of each target based on the learning method of the RBF neural network, so that the weight of each target is better and dynamically adjusted. The purpose of multi-objective optimization is to realize energy consumption, task delay, RSU load balance and privacy security of an unloading task by jointly considering an unloading decision and allocation of computing resources, and by optimizing and computing a plurality of indexes of unloading, the unloading time delay and energy consumption of vehicles in the Internet of Vehicles are kept at a relatively low basic level, the resource-limited equipment can unload the computation-intensive task to the edge equipment, so that a unique delay-limited service quality guarantee is provided for wide businesses.

Description

technical field [0001] The present invention relates to the field of computing offloading and resource allocation in the Internet of Vehicles driven by a 5G network, and specifically relates to a method and system for computing offloading in the Internet of Vehicles based on multi-objective reinforcement learning. Background technique [0002] As one of the most promising technologies in the 5G era, the traditional ad hoc network of vehicles is rapidly developing towards the Internet of Vehicles (IoV). In-vehicle applications and services are becoming more and more abundant (such as autonomous driving, video-assisted real-time navigation, augmented reality, etc.), and these applications are often computationally intensive, energy-intensive, and low-latency applications. However, the limited computing power of the on-board computing unit has become the bottleneck of these applications, making it difficult to meet the low-latency real-time requirements. Mobile edge computing ...

AI Technical Summary

Problems solved by technology

To make matters worse, user needs may change over time, for example, some in-vehicle applications require less latency when emergency events are detected and longer battery life when no events are detected, Unable to effectively improve the computing offloading efficiency of the Internet of Vehicles

Images