Integration of vision and force sensors for grasping
Peter K. Allen; Andrew T. Miller; Paul Y. Oh; Brian S. Leibowitz
- Integration of vision and force sensors for grasping
Allen, Peter K.
Miller, Andrew T.
Oh, Paul Y.
Leibowitz, Brian S.
- Computer Science
- Permanent URL:
- Book/Journal Title:
- MFI '96, 1996 IEEE/SICE/RSJ International Conference on Multisensor Fusion and Integration for Intelligent Systems, December 8-11, 1996, Washington, D.C., U.S.A.
- Publisher Location:
- Piscataway, N.J.
- This paper describes a set of methods that can be used to integrate real-time external vision sensing with internal force and position sensing to estimate contact forces by the fingers of a hand. Estimating these forces and contacts is essential to performing dextrous manipulation tasks. Most robotic hands are either sensorless or lack the ability to accurately and robustly report position and force information relating to contact. By adding external vision sensing, we can complement any internal sensors to more accurately estimate forces and contact positions. Experiments are described that use real-time visual trackers in conjunction with internal strain gauges and a new tactile sensor to accurately estimate finger contacts and applied forces for a three fingered robotic hand.
- Item views: