Improved PSO-Based Task Offloading Model for Internet of Vehicles Edge Computing

Engineering notes

Key topics: autonomous driving, deployment. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

The demand for computer resources for internet of vehicles services like autonomous driving, real-time navigation, and in-vehicle entertainment has grown rapidly due to the widespread deployment of intelligent transportation systems and the ongoing advancement of information and communication technologies.Therefore, a novel task offloading optimization allocation model for internet of vehicles edge computing is proposed.The model is based on mobile edge computing architecture.Through clustering algorithm, it intelligently clusters all nodes in the static parked vehicles edge computing architecture.Moreover, the PSO algorithm is coded and optimized, which improves the efficiency and resource utilization of internet of vehicles task offloading.The experimental results indicated that the model was able to realize obvious intercluster separation under 2 min, 10 min, 50 min, and 100 min time nodes.The vehicles inside the clusters were also more closely distributed, resulting in good internal consistency and external separation.When the number of tasks was increased to 60, the corresponding total system cost of the research model was only 198.When the task computation volume was 120 GHZ, the total system cost of the research model was only 214.In addition, the research model still maintained a high offloading success rate of 97.5%, 94.6%, and 92.8 in low-density, medium-density, and high-density environments.In summary, the research model not only can effectively improve the vehicle task processing efficiency and reduce the system overhead, but also shows strong adaptability and robustness, which has good prospects for practical applications.

Links and sources

• PDF from original source

Comments