Simultaneous Sensory Feedback Strategy for Force and Position of Gripper Based on TENS

Abstract

We present a transcutaneous electrical nerve stimulation (TENS) based sensory feedback strategy to simultaneously transmit two senses through a single channel for giving force and position information of the gripper to the user. The two sensations were delivered by mixing two differently shaped current signals. The force of the gripper was delivered by high-frequency stimulation (pressure) with pulse width modulation. The position was delivered by low-frequency stimulation (tapping). Users can feel the force/position of the gripper and control it without relying on visual/auditorial sense. We conducted three experiments (force level control, object size, and stiffness recognition) and one demonstration. All results were compared with the result of the two-finger pinching experiment. Ten health subjects participated, and their eyes and ears were blocked. The result in force level control showed that the error of the proposed method (NRMSE: 0.103 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula> 0.073) and fingers (0.009 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula> 0.072) had a difference of less than 4 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> . The proposed method showed higher accuracy (+5.93 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> ) and faster response (−0.41 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$s$</tex-math></inline-formula> ) in size recognition, and lower accuracy (−7.4 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\%$</tex-math></inline-formula> ), and slower (+0.32 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$s$</tex-math></inline-formula> ) in stiffness recognition than fingers. Therefore, the proposed method implemented the sense of force/position at a similar performance level of the fingers. Additionally, the spatial limitation was overcome by reducing the number of electrodes needed.