Real-time control using tactile feedback and fuzzy logic

Tactile sensors are a class of robotic sensors capable of measuring contact interaction between a robotic gripper and a grasped object. Tactile sensors are an important source of dynamic information for real-time gripper control during manipulation tasks. To be effective, this dynamic information must be used as part of a real-time control scheme because many of the conditions indicated by the tactile sensor require immediate attention (typically in milliseconds or less). One important application of dynamic tactile data is the tracking of objects in real time. This paper discusses a tracking system that was implemented to prove the feasibility of using a tactile sensor and fuzzy logic to track objects in real time. The system as implemented used a tactile sensor, an X-Y table, a personal computer and custom fuzzy-logic software. It successfully tracked objects at speeds in excess of 1 ms/sup -1/.<<ETX>>

[1]  Thomas H. Speeter A Tactile Sensing System for Robotic Manipulation , 1990, Int. J. Robotics Res..

[2]  Brad L. Hutchings,et al.  Multiple-layer cross-field ultrasonic tactile sensor , 1994, Proceedings of the 1994 IEEE International Conference on Robotics and Automation.

[3]  Ronald S. Fearing,et al.  The dynamic response of a tactile sensor , 1990, Proceedings., IEEE International Conference on Robotics and Automation.

[4]  Mark H. Lee,et al.  A Survey of Robot Tactile Sensing Technology , 1989, Int. J. Robotics Res..

[5]  Hendrik Van Brussel,et al.  A sensory controlled gripper system , 1990, Robotics Auton. Syst..

[6]  E. Cox,et al.  Fuzzy fundamentals , 1992, IEEE Spectrum.

[7]  Pradeep K. Khosla,et al.  Using Tactile Data for Real-Time Feedback , 1991, Int. J. Robotics Res..