Abstract
We present a distributed model for mobile robot spatial mapping and path planning. The method was implemented and tested on a sonar and compass-based physical mobile robot controlled by three competence layers:
-
Low-level navigation: a collection of reflex-like rules whose combination results in emergent collision-free edge-following.
-
Landmark detection: using the underlying reflexive navigation, the robot dynamically extracts procedurally-defined landmarks from the environment.
-
Spatial Mapping and Path Planning: the landmarks are used to construct a distributed topological map of the environment. The locations in the map are individual, independently acting processes. This implementation allows for localization in constant time. Spreading of activation is used to compute both topological and physical shortest paths in linear time.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
“Motivational Learning of Spatial Behavior”, Michael A. Arbib and Israel Lieblich, Systems Neuroscience, 1970.
“The Behavior Language; User's Guide”, Rodney A. Brooks, MIT Artificial Intelligence Lab Memo # 1227, April 1990.
“A Hardware Retargetable Distributed Layered Architecture for Mobile Robot Control”, Rodney A. Brooks, Proceedings IEEE Robotics and Automation, Raleigh, NC, April 1987, 106–110.
“Position Referencing and Consistent World Modeling for Mobile Robots”, R. Chatila and J. Laumond, Proceedings of the 1986 IEEE International Conference on Robotics and Automation, March 1985.
“Animal Cognition: The Representation of Space, Time and Number”, C. R. Gallistel, Annual Review of Psychology 1989, Vol. 40, Palo Alto, California, 155–189.
“Ethology: The Mechanisms and Evolution of Behavior”, James L. Gould, W. W. Norton and Company, New York, 1982.
“A Robust, Qualitative Approach to a Spatial Learning Mobile Robot”, Benjamin J. Kuipers and Yung-Tai Byun, SPIE Advances in Intelligent Robotics Systems Proceedings, November 1988.
“A Distributed Model for Mobile Robot Environment-Learning and Navigation”, Maja J Mataric, MIT Artificial Intelligence Lab Technical Report #1228, June 1990.
“Qualitative Sonar Based Environment Learning for Mobile Robots”, Maja J Mataric, SPIE 1989 Mobile Robots IV Proceedings, November, 1989.
“Navigating With a Rat Brain”, Maja J Mataric, Proceedings, Simulation of Adaptive Behavior: From Animals to Animats, MIT Press Bradford Books, 1990.
“Learning a Distributed Map Representation Based on Navigation Behaviors”, Maja J Mataric and Rodney A. Brooks, Proceedings of 1990 USA-Japan Symposium on Flexible Automation, June 1990.
“High Resolution Maps From Wide Angle Sonar”, Proceedings of the 1985 IEEE International Conference on Robotics and Automation, Hans P. Moravec and Alberto Elfes, St. Louis, MO, March 1985.
“Semantic Memory”, M. Ross Quillian, Semantic Information Processing, Marvin Minsky, ed., MIT Press, Cambridge, MA, 1969, 227–270.
“Animal Navigation”, Talbot H. Waterman, Scientific American Library, New York, 1989.
“'Matched Filters’ — Neural Models of the External World”, Rüdiger Wehner, J. comp. Physiol. A 161, 511–531.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mataric, M.J. (1991). Parallel, decentralized spatial mapping for robot navigation and path planning. In: Schwefel, HP., Männer, R. (eds) Parallel Problem Solving from Nature. PPSN 1990. Lecture Notes in Computer Science, vol 496. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0029779
Download citation
DOI: https://doi.org/10.1007/BFb0029779
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54148-6
Online ISBN: 978-3-540-70652-6
eBook Packages: Springer Book Archive