A Study on Lane Congestion Recognition Mechanism for Highways Based on Multi-Source Data

This paper mainly studies the recognition mechanism of traffic congestion on the highway based on multi-source data. To form an accurate and good means for recognizing lane congestion by putting together various data sources such as traffic flow, speed, density and video surveillance data. We propose the use of the combination of machine learning algorithms and traditional traffic theory for its data fusion model. Realworld highway data is used for experiments to prove this method. The results show that the proposed mechanism performs better than traditional single-source data-based approach w.r.t accuracy and robustness.