Using a Cognitive Architecture to Solve Simultaneous Localization and Mapping (SLAM) Problems

Report No. ARL-MR-0639
Authors: Troy D. Kelley
Date/Pages: May 2006; 18 pages
Abstract: The cognitive production system architecture, the Atomic Components of Thought-Rational (ACT-R), was applied to the problem of simultaneous localization and mapping (SLAM). The map space was represented as a pseudo-topological map of salient features, which was then evaluated with a production system architecture (ACT-R). Results showed that after the robot was placed in five unique locations, it was able to recognize them as spaces that it had previously visited. Results also showed that intrinsic error associated with robotic sensor readings was overcome with the use of the simulated human memory activation equations of ACT-R.
Distribution: Approved for public release
  Download Report ( 0.094 MBytes )
If you are visually impaired or need a physical copy of this report, please visit and contact DTIC.
 

Last Update / Reviewed: May 1, 2006