Fuzzy Logic in Autonomous Robot Navigation
a case study

Alessandro Saffiotti

Center for Applied Autonomous Sensor Systems
Dept. of Technology, University of Örebro, Sweden

The contents of this note appear as Chapter G6 ``Autonomous Robot Navigation'' in the Handbook of Fuzzy Computation, E. Ruspini, P. Bonissone and W. Pedrycz, Eds. (Oxford Univ. Press and IOP Press, 1998).
The development of techniques for autonomous navigation constitutes one of the major trends in the current research on mobile robotics. In this case study, we discuss how fuzzy computation techniques have be used in the SRI International mobile robot Flakey to address some of the difficult issues posed by autonomous navigation: (i) how to design basic behaviors; (ii) how to coordinate behaviors to execute full navigation plans; and (iii) how to use approximate map information. Our techniques have been validated in both in-house experiments and public events. The use of fuzzy logic has resulted in smooth motion control, robust performance in face of errors in the prior knowledge and in the sensor data, and principled integration between different layers of control.


This paper in PostScript format
gzip compressed (92 Kbytes)
uncompressed (364 Kbytes).

Other links
Survey paper on using fuzzy logic in autonomous robotics
Bibliography on the use of fuzzy logic in the robot Flakey
Flakey's home page at SRI International
Flakey in Action (movies)

The work on Flakey was done in collaboration with Enrique Ruspini and Kurt Konolige while the author was visiting the AI Center of SRI International. Support was partly provided by a grant from the Italian National Council for Research, and partly by SRI International.

Last updated: August 7, 1997, by A. Saffiotti