HapCompass: A Rotational Haptic Device for Contact-Rich Robotic Teleoperation
Xiangshan Tan, Jingtian Ji, Tianchong Jiang, Pedro Lopes, Matthew Walter
AI summary
Problem
Existing wearable teleoperation interfaces lack intuitive directional haptic feedback, relying on non-directional vibrations or complex actuator arrays that cause perceptual interference and leave operators insensitive to critical contact forces.
Approach
The authors developed HAPCOMPASS, a low-cost wearable ring that mechanically rotates a single linear resonant actuator to render 2D directional pulling sensations, integrated with a hand-tracking system to translate robot contact forces into intuitive haptic cues.
Key results
- Novel low-cost wearable device rendering 2D directional cues via a single rotating LRA
- Quantitatively validated accurate directional perception by human operators
- Increased task success rates and reduced completion time and maximum contact forces
- Improved downstream imitation learning policy success and reduced peak contact forces
Why it matters
Provides an accessible, high-fidelity haptic feedback solution that bridges the sensory gap in robotic teleoperation, benefiting researchers and practitioners in teleoperation, haptics, and robot learning.
Abstract
The contact-rich nature of manipulation makes it a significant challenge for robotic teleoperation. While haptic feedback is critical for contact-rich tasks, providing intuitive directional cues within wearable teleoperation interfaces re- mains a bottleneck. Existing solutions, such as non-directional vibrations from handheld controllers, provide limited infor- mation, while vibrotactile arrays are prone to perceptual interference. To address these limitations, we propose HAP- COMPASS, a novel, low-cost wearable haptic device that renders 2D directional cues by mechanically rotating a single linear resonant actuator (LRA). We evaluated HAPCOMPASS’s ability to convey directional cues to human operators and showed that it increased the success rate, decreased the completion time and the maximum contact force for teleoperated manipula- tion tasks when compared to vision-only and non-directional feedback baselines. Furthermore, we conducted a preliminary imitation-learning evaluation, suggesting that the directional feedback provided by HAPCOMPASS enhances the quality of demonstration data and, in turn, the trained policy. We release 1Toyota Technological Institute at Chicago (TTIC), {vincenttann, jijingtian, tianchongj, mwalter}@ttic.edu 2University of Chicago, pedrolopes@uchicago.edu the design of the HAPCOMPASS device along with the code that implements our teleoperation interface: https://ripl. github.io/HapCompass/.