Towards Intelligent Computation Offloading in Dynamic Vehicular Networks: A Scalable Multilayer Pipeline

2026-04-29

AI Summary

This study addresses the degradation of real-time performance in advanced driver-assistance systems (ADAS) caused by limited onboard computational resources in software-defined vehicles. To mitigate this issue, the authors propose a four-tier computation offloading architecture that integrates Kubernetes-based cloud infrastructure with a GPU/CPU cooperative scheduling mechanism to dynamically offload tasks to edge or cloud nodes. The core innovation lies in an enhanced particle swarm optimization algorithm incorporating distance- and direction-based penalty terms, which effectively tackles the challenge of edge server selection in high-mobility scenarios while satisfying task latency constraints and improving scalability. Experimental results demonstrate that, in a 10-server setup, the average CPU execution time is 26 ms, and in a 15-server scenario with 1,000 tasks, GPU response time drops to 550 ms, significantly reducing latency while maintaining high task success rates.

Abstract

Software Defined Vehicles face an increasing computational gap as advanced algorithms and frequent software updates demand more processing power while onboard hardware remains static throughout a vehicle's 10+ year lifespan. This mismatch threatens the performance of safety-critical functions including advanced driver-assistance systems and real-time perception tasks. We propose a novel four-layer computation offloading pipeline that dynamically distributes vehicular functions to cloud and edge resources while meeting strict Round Trip Time constraints. Our key contribution is an enhanced Particle Swarm Optimization algorithm that integrates distance- and direction-based penalties with functional requirements to optimize edge server selection for mobile vehicles. Evaluation using a Kubernetes-based cloud infrastructure with realistic vehicular mobility patterns demonstrates that our approach reduces average response time compared to conventional Brute-Force methods while maintaining the success rate for latency-critical tasks. The modified Particle Swarm Optimization algorithm achieves an average execution time of 26 ms across ten servers and tasks on Central Processing Unit, and 550ms across 15 servers with 1000 tasks on Graphics Processing Unit. These results confirm the pipeline's effectiveness in bridging the computational gap for next-generation Software Defined Vehicles (SDV).

Problem

• Software Defined Vehicles
• Computation Offloading
• Vehicular Networks
• Computational Gap
• Latency-Critical Tasks

Innovation

• Computation Offloading
• Particle Swarm Optimization
• Software Defined Vehicles
• Edge Computing
• Dynamic Vehicular Networks