Method for detecting traffic anomally of urban road with equidistant spatial-temporal division

摘要

A travel time distribution-based multi-modal traffic anomaly detection method. In the method, spatial positioning information at various points in time is acquired by means of on-board GNSS positioning devices in floating vehicles, and large amounts of tracking information from said floating vehicles are analyzed and mined in order to achieve smart detection of traffic anomalies and incidents on city roads. The detection method uses travel time probability distribution to represent the state of traffic, and uses measurement indicators for differences in probability distribution to reflect differences between states of traffic. Differences between states of traffic under various environmental states are also taken into consideration. The principles of the method are clear, and the method is easy to carry out and yields a high detection rate.