一种基于分布式边缘智能的车联网计算卸载方法及系统
By employing a distributed edge intelligence multi-agent deep Q-network method in the vehicle-to-everything (V2X) environment, vehicles act as independent agents to make distributed decisions, thus solving the stability and efficiency problems of offloading decisions in the V2X edge computing environment and minimizing the average offloading latency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING UNIV OF SCI & TECH
- Filing Date
- 2023-10-12
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies cannot effectively solve the problem of computation offloading under dynamic and resource-constrained conditions in the edge computing environment of vehicle-to-everything (V2X) networks. This leads to an exponential increase in the dimensions of the state space and action space, affecting network training efficiency and the stability of offloading decisions.
The method of multi-agent independent deep Q network (IDQN) based on distributed edge intelligence is adopted. Each vehicle is an independent intelligent agent and makes distributed intelligent decisions using local environmental information. The computational offloading decision is optimized through training and execution framework to build a vehicle network computational offloading model to minimize the average offloading latency.
It achieves long-term stable calculation of vehicle unloading decisions in dynamic environments, reduces average unloading latency, and improves the efficiency and stability of the unloading system, showing significant advantages over centralized solutions.
Smart Images
Figure CN117369899B_ABST