A Unified CBR Architecture for Robot
Navigation

Susan Eileen Fox

Mathematics/Computer Science Department, Macalester College
1600 Grand Avenue, Saint Paul, MN 55105
fox@macalester.edu
http://www.macalester.edu/~fox


Abstract. A delivery robot navigating in the real world and contending
with multiple goals and priorities benefits from combining deliberative
and reactive planning. Deliberative planners anticipate and optimize actions, 
and manage multiple goals at once. Reactive planners respond flexibly 
and moment-to-moment to a changing world, based on incomplete
and flawed knowledge. This project proposes using case-based reasoning 
to fully integrate high-level and low-level planning techniques, along
with other reasoning tasks of the system. The resulting robot allows
case selection to mediate between methods. In addition, the robot learns
by recording new reactive behaviors, storing new plans it creates, and
introspective learning of retrieval features.
References

1.	P. Agre and D. Chapman. Pengi: An implementation of a theory of activity.
In Proceedings of the Sixth Annual National Conference on Artificial Intelligence,
Seattle, WA, July 1987. AAAI.
2.	R. Brooks. Intelligence without representation. In Proceedings of the Workshop on
the Foundations of Artificial Intelligence, Cambridge, MA, 1987. MIT.
3.	R. J. Firby. Adaptive Execution in Complex Dynamic Worlds. PhD thesis, Yale
University, Computer Science Department, 1989. Technical Report 672.
4.	S. Fox. Introspective Reasoning for Case-Based Planning. PhD thesis, Indiana
University, Computer Science Department, 1995. IUCS: Technical Report 462.
5.	S. Fox. A new model of reflective introspective learning. In Proceedings of the
Eleventh International Florida Artificial Intelligence Research Symposium Conference, 1998.
6.	S. Fox and D. Leake. Learning to refine indexing by introspective reasoning. In
Proceedings of the First International Conference on Case-Based Reasoning, Sesimbra, 
Portugal, October 1995.
7.	A. Goel, T. Callantine, M. Shankar, and B. Chandrasekaran. Representation,
organization, and use of topographic models of physical spaces for route planning.
In Proceedings of the Seventh IEEE Conference on AI Applications, pages 308314.
IEEE Computer Society Press, 1991.
8.	K. Haigh and M. Veloso. Planning, execution and learning in a robotic agent. In
The Fourth International Conference on Artificial Intelligence Planning Systems
1998 (AIPS98), pages 120127, Pittsburgh, Pennsylvania, June 1998.
9.	L. Meeden. Towards Planning: Incremental Investigations into Adaptive Robot
Control. PhD thesis, Indiana University, Computer Science Department, 1994.
10.	R. Murphy, K. Hughes, and E. Noll. An explicit path planner to facilitate reactive
control and terrian preferences. In 1996 IEEE International Conference on Robotics
and Automation, volume 3, pages 20672072, Minneapolis, MN, April 1996.
11.	N. J. Nilsson. Shakey the robot. Technical Report 323, AI Center, SRI International, 
Menlo Park, CA, 1984.
12.	R. Oehlmann and P. Edwards. Introspection planning: Representing metacognitive
experience. In Proceedings of the 1995 AAAI Spring Symposium on Representing
Mental States and Mechanisms, Stanford, CA, March 1995. AAAI. In press.
13.	A. Ram and A. Francis. Multi-plan Retrieval and Adaptation in an Experience-based 
Agent, chapter 10. AAAI Press, Menlo Park, CA, 1996.
14.	A. Ram and J. Santamaria. Multistrategy learning in reactive control systems for
autonomous robotic navigation. Informatica, 17(4) :347369, 1993.
15.	E. D. Sacerdoti. A structure for plans and behavior. Technical Report 109, SRI
Artificial Intelligence Center, 1975.
