Markov Chain Monte Carlo based inverse modeling of traffic flows using GPS data

In large scale deployments of traffic flow models, estimation of the model parameters is a critical but cumbersome task. A poorly calibrated model leads to erroneous estimates in data--poor environments, and limited forecasting ability. In this article we present a method for calibrating flow model parameters for a discretized scalar conservation law using only velocity measurements. The method is based on a Markov Chain Monte Carlo technique, which is used to approximate statistics of the posterior distribution of the model parameters. Numerical experiments highlight the difficulty in estimating jam densities and provide a new approach to improve performance of the sampling through re-parameterization of the model. Supporting source code for the numerical experiments is available for download at https://github.com/dbwork/MCMC-based-inverse-modeling-of-traffic .

[1]  Andreas Hegyi,et al.  Freeway traffic estimation within particle filtering framework , 2007, Autom..

[2]  P. I. Richards.  Shock Waves on the Highway , 1956 .

[3]  Markos Papageorgiou,et al.  RENAISSANCE – A Unified Macroscopic Model-Based Approach to Real-Time Freeway Network Traffic Surveillance , 2006 .

[4]  R. Horowitz,et al.  Methodological calibration of the cell transmission model , 2004, Proceedings of the 2004 American Control Conference.

[5]  A. Bayen,et al.  A traffic model for velocity data assimilation , 2010 .

[6]  Sylvia Richardson,et al.  Markov Chain Monte Carlo in Practice , 1997 .

[7]  Roberto Horowitz,et al.  Probabilistic Graphical Models of Fundamental Diagram Parameters for Simulations of Freeway Traffic , 2011 .

[8]  B. De Moor,et al.  Model predictive control with repeated model fitting for ramp metering , 2002, Proceedings. The IEEE 5th International Conference on Intelligent Transportation Systems.

[9]  Rinaldo M. Colombo,et al.  A Hölder continuous ODE related to traffic flow , 2003, Proceedings of the Royal Society of Edinburgh: Section A Mathematics.

[10]  J. Lebacque.  THE GODUNOV SCHEME AND WHAT IT MEANS FOR FIRST ORDER TRAFFIC FLOW MODELS , 1996 .

[11]  Mauro Garavello,et al.  Traffic Flow on Networks , 2006 .

[12]  Benedetto Piccoli,et al.  A Tracking Algorithm for Car Paths on Road Networks , 2008, SIAM J. Appl. Dyn. Syst..

[13]  E. Somersalo,et al.  Statistical and computational inverse problems , 2004 .

[14]  Andreas Hegyi,et al.  Dual EKF State and Parameter Estimation in Multi-Class First-Order Traffic Flow Models , 2008 .

[15]  Jaimyoung Kwon,et al.  Automatic Calibration of the Fundamental Diagram and Empirical Observations on Capacity , 2009 .

[16]  A. Bayen,et al.  On sequential data assimilation for scalar macroscopic traffic flow models , 2012 .

[17]  Geert Wets,et al.  Model Predictive Control for Ramp Metering Combined with Extended Kalman Filter-Based Traffic State Estimation , 2006, 2006 IEEE Intelligent Transportation Systems Conference.

[18]  Benedetto Piccoli,et al.  A Fluid Dynamic Approach for Traffic Forecast from Mobile Sensor Data , 2010 .

[19]  Jingwen Yan.  Parameter Identification of Freeway Traffic Flow Model and Adaptive Ramp Metering , 2009, 2009 Second International Symposium on Electronic Commerce and Security.

[20]  M J Lighthill,et al.  On kinematic waves II. A theory of traffic flow on long crowded roads , 1955, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences.

[21]  Markos Papageorgiou,et al.  Real-Time Freeway Traffic State Estimation Based on Extended Kalman Filter: A Case Study , 2007, Transp. Sci..

[22]  Markos Papageorgiou,et al.  Parameter identification for a traffic flow model , 1979, Autom..

[23]  Bart De Schutter,et al.  A comparison of filter configurations for freeway traffic state estimation , 2006, 2006 IEEE Intelligent Transportation Systems Conference.

[24]  Stef Smulders,et al.  Control of freeway traffic flow by variable speed signs , 1990 .

[25]  R. LeVeque.  Numerical methods for conservation laws , 1990 .