A fast decision making method for mandatory lane change using kernel extreme learning machine

Lane change maneuver on the highway is a complicated process. A quick and accurate decision for the maneuver is very important for a safe driving. This paper proposes a K-ELM (kernel extreme learning machine) based decision making method for mandatory lane changes. In this method, multiple driving variables that are essential for an accurate lane change are extracted and used as the inputs of an established K-ELM network to generate the right lane-changing decision. The K-ELM network is trained using a tenfold cross-validating approach with the vehicle trajectory data from the NGSIM (next generation simulation) data set on U.S. Highway 101 and Interstate 80. Simulation results demonstrate that the proposed method can generate the lane-changing decision with a 92.86% accuracy for merge events and a 94.36% accuracy for non-merge events. Compared with both the ELM and the SVM method, the proposed method is more accurate and faster in decision making.

[1]  Donald R Drew,et al.  GAP ACCEPTANCE IN THE FREEWAY MERGING PROCESS , 1967 .

[2]  Peter Hidas,et al.  Modelling vehicle interactions in microscopic simulation of merging and weaving , 2005 .

[3]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[4]  Harilaos N. Koutsopoulos,et al.  A microscopic traffic simulator for evaluation of dynamic traffic management systems , 1996 .

[5]  Kuldip K. Paliwal,et al.  Linear discriminant analysis for the small sample size problem: an overview , 2014, International Journal of Machine Learning and Cybernetics.

[6]  Matti Pietikäinen,et al.  Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Haris N. Koutsopoulos,et al.  Modeling Integrated Lane-Changing Behavior , 2003 .

[8]  Rahul Sukthankar,et al.  Evolving an intelligent vehicle for tactical reasoning in traffic , 1997, Proceedings of International Conference on Robotics and Automation.

[9]  Peter L. Bartlett,et al.  The Sample Complexity of Pattern Classification with Neural Networks: The Size of the Weights is More Important than the Size of the Network , 1998, IEEE Trans. Inf. Theory.

[10]  Hongming Zhou,et al.  Extreme Learning Machine for Regression and Multiclass Classification , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[11]  Raj Madhavan,et al.  PRIDE: a hierarchical, integrated prediction framework for autonomous on-road driving , 2006, Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006..

[12]  Jianping Wu,et al.  Lane changing on the motorway: factors affecting its occurrence, and their implications , 1998 .

[13]  Moshe Ben-Akiva,et al.  MODELS OF FREEWAY LANE CHANGING AND GAP ACCEPTANCE BEHAVIOR , 1996 .

[14]  Mashrur Chowdhury,et al.  Review of Microscopic Lane-Changing Models and Future Research Opportunities , 2013, IEEE Transactions on Intelligent Transportation Systems.

[15]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[16]  Majid Sarvi,et al.  Lane changing models: a critical review , 2010 .

[17]  Kuldip K. Paliwal,et al.  A feature selection method using improved regularized linear discriminant analysis , 2014, Machine Vision and Applications.

[18]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[19]  Fengjun Yan,et al.  Lane changing prediction at highway lane drops using support vector machine and artificial neural network classifiers , 2016, 2016 IEEE International Conference on Advanced Intelligent Mechatronics (AIM).

[20]  Kuldip K. Paliwal,et al.  Principal component analysis using QR decomposition , 2013, Int. J. Mach. Learn. Cybern..

[21]  P. G. Gipps,et al.  A MODEL FOR THE STRUCTURE OF LANE-CHANGING DECISIONS , 1986 .

[22]  Haris N. Koutsopoulos,et al.  Integrated driving behavior modeling , 2007 .