Fusing Moving Average Model and Stationary Wavelet Decomposition for Automatic Incident Detection: Case Study of Tokyo Expressway

Traffic congestion is a growing problem in urban areas all over the world. The transport sector has been in full swing event study on intelligent transportation system for automatic detection. The functionality of automatic incident detection on expressways is a primary objective of advanced traffic management system. In order to save lives and prevent secondary incidents, accurate and prompt incident detection is necessary. This paper presents a methodology that integrates moving average (MA) model with stationary wavelet decomposition for automatic incident detection, in which parameters of layer coefficient are extracted from the difference between the upstream and downstream occupancy. Unlike other wavelet-based method presented before, firstly it smooths the raw data with MA model. Then it uses stationary wavelet to decompose, which can achieve accurate reconstruction of the signal, and does not shift the signal transfer coefficients. Thus, it can detect the incidents more accurately. The threshold to trigger incident alarm is also adjusted according to normal traffic condition with congestion. The methodology is validated with real data from Tokyo Expressway ultrasonic sensors. Experimental results show that it is accurate and effective, and that it can differentiate traffic accident from other condition such as recurring traffic congestion.

[1]  Samir A. Ahmed,et al.  Discrete dynamic models for freeway incident detection systems , 1982 .

[2]  Edward Chung,et al.  Effective incident detection and management on freeways , 1999 .

[3]  Lisa Aultman-Hall,et al.  A Catastrophe Theory Approach to Freeway Incident Detection , 1991 .

[4]  Fred L. Hall,et al.  Catastrophe theory and patterns in 30 second freeway traffic data, implications for incident detection , 1989 .

[5]  Bhagwant Persaud,et al.  CONGESTION IDENTIFICATION ASPECTS OF THE MCMASTER INCIDENT DETECTION ALGORITHM , 1990 .

[6]  Asim Karim,et al.  INCIDENT DETECTION ALGORITHM USING WAVELET ENERGY REPRESENTATION OF TRAFFIC PATTERNS , 2002 .

[7]  Vedagiri Perumal,et al.  Evaluation of Pedestrian Mid-Block Road Crossing Behavior Using an Artificial Neural Network (ANN) , 2014 .

[8]  Ge Hong-Xia,et al.  A modified coupled map car-following model based on application of intelligent transportation system and control of traffic congestion , 2007 .

[9]  H J Payne,et al.  DEVELOPMENT AND TESTING OF INCIDENT DETECTION ALGORITHMS, VOLUME 2: RESEARCH METHODOLOGY AND DETAILED RESULTS , 1976 .

[10]  Samir A. Ahmed,et al.  APPLICATION OF TIME-SERIES ANALYSIS TECHNIQUES TO FREEWAY INCIDENT DETECTION , 1982 .

[11]  Yixuan Li,et al.  Motorway incident detection using probe vehicles , 2005 .

[12]  Joseph D. Crabtree,et al.  Dedicated Short-Range Communications Technology for Freeway Incident Detection , 2007 .

[13]  Jiang Gui.  Design of Freeway Automatic Incident Detection System and Algorithm , 2001 .

[14]  Der-Horng Lee,et al.  Mobile Sensor and Sample-Based Algorithm for Freeway Incident Detection , 2002 .

[15]  Hojjat Adeli,et al.  Wavelet‐Clustering‐Neural Network Model for Freeway Incident Detection , 2003 .

[16]  Jye-Chyi Lu,et al.  Wavelet-based SPC procedure for complicated functional data , 2006 .

[17]  Phil Charles.  Traffic incident management: best practice , 2007 .

[18]  Dipti Srinivasan,et al.  Training Neural Networks to Detect Freeway Incidents by Using Particle Swarm Optimization , 2004 .

[19]  Hussein Dia,et al.  Development and evaluation of neural network freeway incident detection models using field data , 1997 .

[20]  Vedagiri Perumal,et al.  Evaluation of pedestrian mid-block road crossing behaviour using artificial neural network , 2014 .

[21]  A. R. Cook,et al.  DETECTION OF FREEWAY CAPACITY-REDUCING INCIDENTS BY TRAFFIC-STREAM MEASUREMENTS , 1974 .

[22]  Di Chen,et al.  Wavelet-Based Data Reduction Techniques for Process Fault Detection , 2006, Technometrics.

[23]  Moshe Levin,et al.  INCIDENT DETECTION: A BAYESIAN APPROACH , 1978 .

[24]  Jing-Huai Gao,et al.  Traffic incident acoustic recognition method based on wavelet decomposition and support vector machine , 2010 .

[25]  Dipti Srinivasan,et al.  DEVELOPMENT AND ADAPTATION OF CONSTRUCTIVE PROBABILISTIC NEURAL NETWORK IN FREEWAY INCIDENT DETECTION , 2002 .

[26]  Lee D. Han,et al.  A wavelet-based freeway incident detection algorithm with adapting threshold parameters , 2011 .

[27]  Teng Jianfu,et al.  Research of gyro signal de-noising with stationary wavelets transform , 2003, CCECE 2003 - Canadian Conference on Electrical and Computer Engineering. Toward a Caring and Humane Technology (Cat. No.03CH37436).

[28]  D J Bowers,et al.  THE ASTRID/INGRID INCIDENT DETECTION SYSTEM FOR URBAN AREAS , 1995 .