Related Experiment Video
Updated: Dec 20, 2025
09:41
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
Published on: April 21, 2023
2.1K
Proprioceptive Estimation of Forces Using Underactuated Fingers for Robot-Initiated pHRI
Joaquin Ballesteros1, Francisco Pastor2, Jesús M Gómez-De-Gabriel2
1Department of Computer Languages and Science, University of Malaga, Escuela de Ingeniería Informática, 29071 Málaga, Spain.
Sensors (Basel, Switzerland)
|May 24, 2020
Summary
This study introduces a novel method for estimating human-robot interaction forces using only a robot gripper's built-in sensors. This approach enables adaptive shared control in limb manipulation without extra hardware.
Area of Science:
- Robotics
- Human-Robot Interaction
- Control Systems
Background:
- Accurate estimation of human-exerted forces is crucial for safe and effective physical Human-Robot Interaction (pHRI).
- Existing methods often require additional, costly force sensors, limiting widespread adoption.
- Adaptive shared control strategies in limb manipulation necessitate real-time force feedback.
Purpose of the Study:
- To develop and validate a method for estimating interaction forces in pHRI using only proprioceptive sensing within a robot gripper.
- To enable adaptive shared control for limb manipulation tasks by accurately inferring human forces.
- To demonstrate a hardware-agnostic and computationally inexpensive approach for force estimation.
Main Methods:
- Utilized an underactuated gripper with proprioceptive sensing to measure forces exerted by a human limb in the frontal plane.
- Employed a regression method trained on experimental data, correlating phalanx angles and actuator signals (motor current, position, torque) with interaction forces.
- Focused on leveraging existing gripper sensor feedback without supplementary hardware.
Main Results:
- The proposed regression method successfully estimated human-exerted forces with sufficient accuracy for adaptive shared control applications.
- The approach demonstrated effectiveness using only the gripper's inherent motor feedback and joint angle data.
- The method proved computationally inexpensive, allowing for low processing times suitable for continuous human-adapted pHRI.
Conclusions:
- Proprioceptive sensing within a gripper is a viable and cost-effective alternative to dedicated force sensors for estimating interaction forces in pHRI.
- The developed method facilitates adaptive shared control in limb manipulation, enhancing safety and user experience.
- This approach significantly reduces hardware requirements and computational load, promoting broader implementation of intelligent robotic systems.
