Traffic congestion is a significant issue faced by the country and leads to heavy financial losses every year. This issue is related to the lack of infrastructure, bad weather, work zones, and major and minor accidents. Researchers at the University of South Florida have devised a method and an algorithm for congestion detection and mitigation. The algorithm uses data-driven and signal-processing techniques with real-time data of the factors that can cause congestion. Furthermore, this algorithm can distinguish and identify four different congestion levels. An advantageous feature of this method is that it can identify minute factors related to the travel time of vehicles that can cause congestion, so actions could be taken to mitigate the jam in advance. The method uses state-of-the-art techniques with a novel application of Butterworth filter for congestion prediction, which is unique to traffic detection research. The congestion level thresholds related to the jam can be continuously updated with a feedback loop relating to previous databases for the roadway and geographical traffic data, which helps in providing accurate predictions of future traffic congestion.
Finalized workflow of the congestion detection algorithm