A Non-Gaussian Kalman Filter With Application to the Estimation of Vehicular Speed

Single inductive loop detectors (ILDs), which provide online measurements of traffic volume and occupancy, are widely used devices in road systems. Because of the nature of traffic flow, fast estimation and forecasting of vehicular speed using the data collected by an ILD are crucial to online road traffic management. Here we formulate statistical inference for vehicular speed as a dynamic generalized linear model with a reciprocal inverse Gaussian observational distribution. The formulation motivates us to extend the Gaussian Kalman filter to this non-Gaussian scenario. This results in a set of simple recursive formulas in which the current estimate of the parameter of interest is updated as a weighted harmonic average of the previous estimate and the current observation. By applying the developed non-Gaussian Kalman filter to analyze traffic data collected by an ILD, we provide a competitive alternative for estimating vehicular speed at a minimum computational cost.

[1]  A. Basu,et al.  The Inverse Gaussian Distribution , 1993 .

[2]  Joseph E. Cavanaugh,et al.  Regression Models for Time Series Analysis , 2004 .

[3]  G. Kitagawa.  Monte Carlo Filter and Smoother for Non-Gaussian Nonlinear State Space Models , 1996 .

[4]  Pravin Varaiya,et al.  Measuring Traffic , 2008, 0804.2982.

[5]  Byron J. T. Morgan,et al.  Applied Stochastic Modelling , 2001 .

[6]  Donald J Dailey,et al.  A statistical algorithm for estimating speed from single loop volume and occupancy measurements , 1999 .

[7]  Benjamin Kedem,et al.  Regression models for time series analysis , 2002 .

[8]  David B. Dunson,et al.  Bayesian Data Analysis , 2010 .

[9]  M. West,et al.  Dynamic Generalized Linear Models and Bayesian Forecasting , 1985 .

[10]  R J Salter.  HIGHWAY TRAFFIC ANALYSIS AND DESIGN :(2ND ED. ) , 1989 .

[11]  Peter Bonsall,et al.  Understanding Traffic Systems: Data Analysis and Presentation , 1996 .

[12]  Siem Jan Koopman,et al.  Time Series Analysis of Non-Gaussian Observations Based on State Space Models from Both Classical and Bayesian Perspectives , 1999 .

[13]  Adolf D. May,et al.  Traffic Flow Fundamentals , 1989 .

[14]  N. L. Johnson,et al.  Continuous Univariate Distributions. , 1995 .

[15]  Stanley Gershwin,et al.  Estimation of roadway traffic density on freeways using presence detector data , 1979, 1979 18th IEEE Conference on Decision and Control including the Symposium on Adaptive Processes.

[16]  Michael A. West,et al.  Bayesian Forecasting and Dynamic Models (2nd edn) , 1997, J. Oper. Res. Soc..

[17]  A. Brix.  Bayesian Data Analysis, 2nd edn , 2005 .

[18]  Baibing Li,et al.  Bayesian Inference for Origin-Destination Matrices of Transport Networks Using the EM Algorithm , 2005, Technometrics.

[19]  G. Kitagawa.  Non-Gaussian State—Space Modeling of Nonstationary Time Series , 1987 .

[20]  M. West,et al.  Bayesian forecasting and dynamic models , 1989 .

[21]  Nicholas G. Polson,et al.  A Monte Carlo Approach to Nonnormal and Nonlinear State-Space Modeling , 1992 .

[22]  R. J. Salter,et al.  Highway Traffic Analysis and Design , 1974 .

[23]  Randolph W. Hall,et al.  Handbook of transportation science , 1999 .

[24]  Martin L. Hazelton.  Estimating Vehicle Speed from Traffic Count and Occupancy Data , 2004 .