Anomalous Traffic Pattern Detection in Large Urban Areas: Tensor-Based Approach with Continuum Modeling of Traffic Flow

Analysis of traffic dynamics in large urban transportation networks is a complicated procedure, yet critical for many areas of transportation research and contemporary intelligent transportation systems. The degree of complexity is increasing, considering the existence of unexpected events such as natural or manmade disasters. The study addresses the needs of detection and description of abnormal traffic patterns formed due to the presence of aforementioned disruptions. In order to take into account complex spatiotemporal structure of traffic dynamics and preserve multi-mode correlations, tensor-based traffic data representation is put forward. Tensor robust principal component analysis is applied for the purpose of discovering distinctive normal and abnormal traffic patterns. For validation purposes, continuum modeling approach is employed to emulate traffic dynamics, with consideration of the effect of disruptions. The results suggested applicability of proposed approach in order to discover abnormal patterns in large urban networks.