Track-based self-supervised classification of dynamic obstacles

• Published in: Autonomous robots Vol. 29; no. 2; pp. 219 - 233
• Main Authors: Katz, Roman, Nieto, Juan, Nebot, Eduardo, Douillard, Bertrand
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01-08-2010; Springer Nature B.V
• Subjects: Architecture; Artificial Intelligence; Autonomous; Classification; Computer Imaging; Control; Engineering; Labels; Mechatronics; Moving obstacles; Obstacles; Pattern Recognition and Graphics; Retraining; Robotics; Robotics and Automation; Robots; Segmentation; Sensors; Trains; Urban environments; Vision; Learning and adaptive systems; Intelligent vehicles; Obstacle classification; Self-supervised learning
• ISSN: 0929-5593; 1573-7527
• DOI: 10.1007/s10514-010-9193-0

This work introduces a self-supervised architecture for robust classification of moving obstacles in urban environments. Our approach presents a hierarchical scheme that relies on the stability of a subset of features given by a sensor to perform an initial robust classification based on unsupervised techniques. The obtained results are used as labels to train a set of supervised classifiers. The outcomes obtained with the second sensor can be used for higher level tasks such as segmentation or to refine the within-clusters discrimination. The proposed architecture is evaluated for a particular realization based on range and visual information which produces track-based labeling that is then employed to train supervised modules that perform instantaneous classification. Experiments show that the system is able to achieve 95% classification accuracy and to maintain the performance through on-line retraining when working conditions change.