Modular Actuator for Multimodal Proprioceptive and Kinesthetic Feedback of Robotic Hands
Sungwoo Park, Myo-Taeg Lim, Donghyun Hwang
AI summary
Problem
Robotic hands lack the space and wiring capacity to integrate sensors for proprioceptive and kinesthetic feedback, hindering precise force control and the ability to assess intrinsic object properties like stiffness and roughness in unstructured environments.
Approach
The authors developed a 10-gram modular actuator featuring a high-ratio multistage gear mechanism and coreless motor, with all sensing and control electronics integrated internally, and implemented an adaptive velocity estimator alongside a simplified Reaction Torque Observer to compute multimodal feedback in real time.
Key results
- Compact 25×10×24 mm, 10 g actuator with fully integrated sensing and control electronics
- Adaptive velocity estimator and simplified RTOB enable real-time position, velocity, current, and torque estimation
- Measurement errors of 5.8 mrad for position, 0.19 rad/s for velocity, and 0.011 N·m for torque
- Accurate detection of object surface shape, roughness, and stiffness without additional sensors
Why it matters
Provides a scalable, sensor-free feedback solution that enhances dexterity and adaptability for compact robotic hands operating in dynamic, unstructured environments.
Abstract
This study addresses the challenge of implementing proprioceptive and kinesthetic (PK) feedback in robotic hands, essential for grasping and manipulation tasks in unstructured environments. We developed a compact modular actuator featuring a low-module, high-transmission-ratio multistage gear mechanism that measures 25×10×24 mm, weighs only 10 grams, and maintains moderate backdrivability. The actuator provides multimodal PK feedback, capturing position, velocity, current, and torque data, which are critical for performing various grasping and manipulation tasks. To enable precise motion and force control, we introduced a new adaptive velocity estimator and a simplified Reaction Torque Observer (RTOB). Comprehensive experiments demonstrated the actuator’s ability to accurately detect surface shape, roughness, and stiffness of target objects, eliminating the need for additional sensors or space. Experimental results confirmed the actuator’s precision, achieving measurement errors of 5.8 mrad for position, 0.19 rad/s for velocity, and 0.011 Nꞏm for torque. These findings highlight the actuator’s ability to leverage proprioceptive information, significantly enhancing the functionality and adaptability of robotic hands in diverse and dynamic scenarios.