New Edge Computing Framework Boosts Real-Time Vehicle Tracking in Large Sensor Networks

A new edge computing framework has been introduced by researchers, offering enhanced real-time vehicle tracking capabilities across vast roadside sensor networks. The framework aims to improve the precision and responsiveness of intelligent transportation systems by leveraging multi-access edge computing (MEC) to process data closer to the source. This approach mitigates issues associated with transmitting large volumes of data to centralized cloud systems, which can lead to latency and affect tracking accuracy. The developed system has demonstrated its ability to continuously monitor tens of thousands of vehicles on highways with high accuracy. It achieved an average longitudinal error of 2.14 meters, a lateral error of 0.84 meters, and a speed error of 1.91 kilometers per hour. The framework was rigorously evaluated on a large-scale road network encompassing 1,777 sensors and covering 157 kilometers of highway. Crucially, the processing latency remained below 340 milliseconds, indicating its potential for near-real-time traffic perception at scale. This distributed framework integrates several stages, including data preprocessing, calibration, multi-sensor trajectory matching, and trajectory prediction. It incorporates machine learning methods, such as LSTM-based prediction, to enhance performance in complex driving conditions. The research highlights the increasing importance of high-precision vehicle positioning and trajectory tracking for connected transportation systems where vehicles, roadside infrastructure, and cloud platforms must interact in real time. If validated in broader traffic scenarios, this framework could significantly improve road safety, traffic efficiency, and the overall responsiveness of connected transportation systems.