Learning in dynamic environments with Ensemble Selection for autonomous outdoor robot navigation
Authors:
- 1 September 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 620-627
Abstract
Autonomous robot navigation in unstructured outdoor environments is a challenging area of active research. The navigation task requires identifying safe, traversable paths which allow the robot to progress toward a goal while avoiding obstacles. Machine learning techniques - in particular, classifier ensembles - are well adapted to this task, accomplishing near-to-far learning by augmenting near-field stereo readings in order to identify safe terrain and obstacles in the far field. Composition of the ensemble and subsequent combination of model outputs in this dynamic problem domain remain open questions. Recently, Ensemble selection has been proposed as a mechanism for selecting and combining models from an existing model library and shown to perform well in static domains. We propose the adaptation of this technique to the time-evolving data associated with the outdoor robot navigation domain. Important research questions as to the composition of the model library, as well as how to combine selected modelspsila outputs, are addressed in a two-factor experimental evaluation. We evaluate the performance of our technique on six fully labeled datasets, and show that our technique outperforms memoryless baseline techniques that do not leverage past experience.Keywords
This publication has 13 references indexed in Scilit:
- Mining Concept-Drifting and Noisy Data Streams Using Ensemble Classifiers2009
- Online Learning of Multiple Perceptual Models for Navigation in Unknown Terrain2008
- Long-Term learning using multiple models for outdoor autonomous robot navigation2007
- Trust region Newton methods for large-scale logistic regression2007
- Getting the Most Out of Ensemble SelectionIEEE International Conference on Data Mining (ICDM), 2006
- The DARPA LAGR program: Goals, challenges, methodology, and phase I resultsJournal of Field Robotics, 2006
- Ensemble selection from libraries of models2004
- A Simple Generalisation of the Area Under the ROC Curve for Multiple Class Classification ProblemsMachine Learning, 2001
- The Elements of Statistical Learning2001
- Learning in the presence of concept drift and hidden contextsMachine Learning, 1996