NTRS - NASA Technical Reports Server

Back to Results
Real-time qualitative reasoning for telerobotic systemsThis paper discusses the sensor-based telerobotic driving of a car in a-priori unknown environments using 'human-like' reasoning schemes implemented on custom-designed VLSI fuzzy inferencing boards. These boards use the Fuzzy Set theoretic framework to allow very vast (30 kHz) processing of full sets of information that are expressed in qualitative form using membership functions. The sensor-based and fuzzy inferencing system was incorporated on an outdoor test-bed platform to investigate two control modes for driving a car on the basis of very sparse and imprecise range data. In the first mode, the car navigates fully autonomously to a goal specified by the operator, while in the second mode, the system acts as a telerobotic driver's aid providing the driver with linguistic (fuzzy) commands to turn left or right, speed up, slow down, stop, or back up depending on the obstacles perceived by the sensors. Indoor and outdoor experiments with both modes of control are described in which the system uses only three acoustic range (sonar) sensor channels to perceive the environment. Sample results are presented that illustrate the feasibility of developing autonomous navigation modules and robust, safety-enhancing driver's aids for telerobotic systems using the new fuzzy inferencing VLSI hardware and 'human-like' reasoning schemes.
Document ID
Document Type
Conference Paper
Pin, Eancois G. (Oak Ridge National Lab. TN, United States)
Date Acquired
September 6, 2013
Publication Date
February 1, 1993
Publication Information
Publication: NASA. Lyndon B. Johnson Space Center, The Sixth Annual Workshop on Space Operations Applications and Research (SOAR 1992)
Subject Category
Distribution Limits
Work of the US Gov. Public Use Permitted.

Related Records

IDRelationTitle19930022908Analytic PrimaryThe Sixth Annual Workshop on Space Operations Applications and Research (SOAR 1992)