Foulds, Richard A. (Committee chair)
Lii, Neal Y. (Committee member)
Adamovich, Sergei (Committee member)
Roman, Max (Committee member)
Date:
2013-01
Keywords:
Telemanipulation
Robotic hand
Availability:
Unrestricted
Abstract:
This work presents a fuzzy logic based HandyMan (Hand gesture commands for grasping and manipulation) system to recognize the operator's hand gestures during task space telemanipulation. A combination of joint positions may be shared between at least two manipulation gestures. To avoid misinterpretation of gestures by the gesture recognition system, six new manipulation gestures were introduced. The gestures are produced by the user wearing a CyberGlove™.
This system replaces the previous HandyMan gesture recognition method. The output of the fuzzy system drives the state machines to implement the gestures with the robot hand.
The experimental results show that the proposed method can be used for telemanipulation of 15-DOF robot hand in task space. Manipulation in six degrees of freedom and pistol grip manipulation is achieved with a good repeatability percentage and no burst errors. The number of fingers used by the user does not affect the number of fingers used by the robot hand during manipulation. Hence, the same manipulation gesture can be used for 2 finger mode, 3 finger mode and 5 finger mode manipulation.
If you have any questions please contact the ETD Team, libetd@njit.edu.