Automated extraction of driver behaviour primitives using Bayesian agglomerative sequence segmentation

The low-level building blocks of driver behaviour have been shown to exhibit statistical patterns such as periods of turning, braking and acceleration, as well as different combinations of these. Collectively, these patterns can be regarded as a language of “driving primitives.” This allows us to reason about more meaningful driving maneuvers, e.g. overtaking, parking, by treating them as sequences of primitives. In this paper we introduce a method for automatically finding the boundaries between primitives, which is important when analysing large volumes of raw sensor data that can be generated in ITS applications. Our method is cost-effective, completely unsupervised and requires minimal preprocessing. We demonstrate the potential of our approach via an experiment with genuine data from an inertial measurement unit.

[1]  Kevin P. Murphy,et al.  Modeling changing dependency structure in multivariate time series , 2007, ICML '07.

[2]  Reza Shadmehr,et al.  Learning of action through adaptive combination of motor primitives , 2000, Nature.

[3]  Tadahiro Taniguchi,et al.  Semiotic prediction of driving behavior using unsupervised double articulation analyzer , 2012, 2012 IEEE Intelligent Vehicles Symposium.

[4]  P. Diaconis,et al.  Conjugate Priors for Exponential Families , 1979 .

[5]  James M. Rehg,et al.  A Variational inference method for Switching Linear Dynamic Systems , 2005 .

[6]  Rafael Toledo-Moreo,et al.  IMM-Based Lane-Change Prediction in Highways With Low-Cost GPS/INS , 2009, IEEE Transactions on Intelligent Transportation Systems.

[7]  Tadahiro Taniguchi,et al.  Unsupervised drive topic finding from driving behavioral data , 2013, 2013 IEEE Intelligent Vehicles Symposium (IV).

[8]  Christoph Bregler,et al.  Learning and recognizing human dynamics in video sequences , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[9]  Milton Abramowitz,et al.  Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables , 1964 .

[10]  Jan Bulla,et al.  hsmm - An R package for analyzing hidden semi-Markov models , 2010, Comput. Stat. Data Anal..

[11]  Gm Gero Walter,et al.  Bayesian linear regression , 2009 .

[12]  Shigeru Okuma,et al.  Modeling of driver's collision avoidance maneuver based on controller switching model , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[13]  A. Dawid.  Some matrix-variate distribution theory: Notational considerations and a Bayesian application , 1981 .

[14]  R. Muirhead.  Aspects of Multivariate Statistical Theory , 1982, Wiley Series in Probability and Statistics.

[15]  Pietro Perona,et al.  Decomposition of human motion into dynamics-based primitives with application to drawing tasks , 2003, Autom..

[16]  Michael I. Jordan,et al.  The Sticky HDP-HMM: Bayesian Nonparametric Hidden Markov Models with Persistent States , 2009 .

[17]  Tatsuya Suzuki,et al.  Multi-scale driving behavior modeling using hierarchical PWARX model , 2010, 13th International IEEE Conference on Intelligent Transportation Systems.

[18]  N. L. Johnson,et al.  Multivariate Analysis , 1958, Nature.

[19]  Tatsuya Suzuki,et al.  Modeling and Recognition of Driving Behavior Based on Stochastic Switched ARX Model , 2007, IEEE Transactions on Intelligent Transportation Systems.