Classifier fusion for outdoor obstacle detection

This work describes an approach for using several levels of data fusion in the domain of autonomous off-road navigation. We are focusing on outdoor obstacle detection, and we present techniques that leverage on data fusion and machine learning for increasing the reliability of obstacle detection systems. We are combining color and infrared (IR) imagery with range information from a laser range finder. We show that in addition to fusing data at the pixel level, performing high level classifier fusion is beneficial in our domain. Our general approach is to use machine learning techniques for automatically deriving effective models of the classes of interest (obstacle and non-obstacle for example). We train classifiers on different subsets of the features we extract from our sensor suite and show how different classifier fusion schemes can be applied for obtaining a multiple classifier system that is more robust than any of the classifiers presented as input. We present experimental results we obtained on data collected with both the experimental unmanned vehicle (XUV) and a CMU developed robotic tractor.

[1]  Gunnar Rätsch,et al.  Soft Margins for AdaBoost , 2001, Machine Learning.

[2]  Dean A. Pomerleau,et al.  Progress in neural network-based vision for autonomous robot driving , 1992, Proceedings of the Intelligent Vehicles `92 Symposium.

[3]  Peter L. Bartlett,et al.  Boosting Algorithms as Gradient Descent , 1999, NIPS.

[4]  Tommy Chang,et al.  Feature detection and tracking for mobile robots using a combination of ladar and color images , 2002, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292).

[5]  Jean-Yves Bouguet,et al.  Camera calibration toolbox for matlab , 2001 .

[6]  Tommy Chang,et al.  Repository of sensor data for autonomous driving research , 2003, SPIE Defense + Commercial Sensing.

[7]  Roberto Manduchi,et al.  Terrain perception for DEMO III , 2000, Proceedings of the IEEE Intelligent Vehicles Symposium 2000 (Cat. No.00TH8511).

[8]  M. Perrone.  Improving regression estimation: Averaging methods for variance reduction with extensions to general convex measure optimization , 1993 .

[9]  L. Cooper,et al.  When Networks Disagree: Ensemble Methods for Hybrid Neural Networks , 1992 .

[10]  Paul A. Viola,et al.  Boosting Image Retrieval , 2004, International Journal of Computer Vision.

[11]  Leo Breiman,et al.  Stacked regressions , 2004, Machine Learning.

[12]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[13]  Robert E. Schapire,et al.  A Brief Introduction to Boosting , 1999, IJCAI.

[14]  Steven A. Shafer,et al.  An architecture for sensor fusion in a mobile robot , 1986, Proceedings. 1986 IEEE International Conference on Robotics and Automation.

[15]  Alonzo Kelly,et al.  Real-Time, Multi-Perspective Perception for Unmanned Ground Vehicles , 2003 .

[16]  David H. Wolpert,et al.  Stacked generalization , 1992, Neural Networks.

[17]  Robert A. Jacobs,et al.  Hierarchical Mixtures of Experts and the EM Algorithm , 1993, Neural Computation.

[18]  Mark Ollis,et al.  Structural method for obstacle detection and terrain classification , 2003, SPIE Defense + Commercial Sensing.

[19]  Robert E. Schapire,et al.  The Boosting Approach to Machine Learning An Overview , 2003 .

[20]  Martial Hebert,et al.  Terrain Classification Techniques From Ladar Data For Autonomous Navigation , 2002 .

[21]  C. M. Shoemaker,et al.  The Demo III UGV program: a testbed for autonomous navigation research , 1998, Proceedings of the 1998 IEEE International Symposium on Intelligent Control (ISIC) held jointly with IEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA) Intell.

[22]  Padhraic Smyth,et al.  An Evaluation of Linearly Combining Density Estimators via Stacking , 1998 .

[23]  Nasser M. Nasrabadi,et al.  Combination of two learning algorithms for automatic target recognition , 1997, Proceedings of International Conference on Image Processing.