Share Email Print

Proceedings Paper

Robot training through incremental learning
Author(s): Robert E. Karlsen; Shawn Hunt; Gary Witus
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The real world is too complex and variable to directly program an autonomous ground robot's control system to respond to the inputs from its environmental sensors such as LIDAR and video. The need for learning incrementally, discarding prior data, is important because of the vast amount of data that can be generated by these sensors. This is crucial because the system needs to generate and update its internal models in real-time. There should be little difference between the training and execution phases; the system should be continually learning, or engaged in "life-long learning". This paper explores research into incremental learning systems such as nearest neighbor, Bayesian classifiers, and fuzzy c-means clustering.

Paper Details

Date Published: 23 May 2011
PDF: 9 pages
Proc. SPIE 8045, Unmanned Systems Technology XIII, 804504 (23 May 2011); doi: 10.1117/12.884092
Show Author Affiliations
Robert E. Karlsen, U.S. Army Tank Automotive Research, Development and Engineering Ctr. (United States)
Shawn Hunt, U.S. Army Tank Automotive Research, Development and Engineering Ctr. (United States)
Gary Witus, Turing Associates (United States)

Published in SPIE Proceedings Vol. 8045:
Unmanned Systems Technology XIII
Douglas W. Gage; Charles M. Shoemaker; Robert E. Karlsen; Grant R. Gerhart, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?