Application of Cloud-Based Real-Time Data Processing in Automated Traffic Management Systems

Abstract

 Autonomous traffic control requires a cloud-based real-time data processing system, and several data sources, including cameras, detectors, linked cars, and signal controllers in cities, continuously produce data. A cloud-edge cooperative processing framework for automated traffic management systems is proposed in this paper. The framework includes low-latency alert, multi-source traffic condition fusion, cloud-based stream analysis, edge processing, and roadside data gathering. Adaptive scheduling will be utilized to dynamically distribute computer resources according to event immediacy, queue length, and traffic stream priority. The suggested approach lowers the average end-to-end latency to 134 ms (from 286 ms) and boosts the processing throughput to 52,000 records/s while also reducing the average intersection delay by 18.6%, according to the experiment results in a simulated urban traffic environment. Based on the study, it is concluded that a scalable system capable of high-quality real-time traffic processing can be constructed by combining edge cooperation and flexible scheduling.