Related Experiment Video
Updated: May 6, 2026
09:41
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
Published on: April 21, 2023
2.8K
Human motion intention based scaled teleoperation for orientation assistance in preshaping for grasping
Summary
This study introduces an algorithm using Hidden Markov Models (HMM) to assist users teleoperating remote grippers. The system predicts human motion intention for quicker, easier grasping, achieving 36% time savings.
Area of Science:
- Robotics
- Human-Computer Interaction
- Biomechanics
Background:
- Teleoperation of robotic grippers for grasping tasks requires precise control.
- Preshaping the gripper to match object geometry is crucial for successful grasping.
- Existing teleoperation methods may lack intuitive assistance for complex manipulation tasks.
Purpose of the Study:
- To develop and validate an algorithm for human motion intention-based assistance in teleoperated grasping.
- To enhance the efficiency and ease of preshaping a remote gripper for object manipulation.
- To reduce the time required for users to achieve the correct preshape configuration.
Main Methods:
- Utilizing human motion data from a remote arm to train a Hidden Markov Model (HMM) offline.
- Real-time processing of motion data through the HMM to infer user's intended preshape configuration.
- Implementing an algorithm that scales remote arm motion based on predicted intention to provide assistance.
Main Results:
- Successful validation of the human motion intention-based assistance algorithm with healthy subjects.
- Demonstrated significant improvements in preshaping speed and ease for users.
- Achieved an average time saving of 36% in grasping tasks.
Conclusions:
- Human motion intention prediction via HMMs effectively assists teleoperation for grasping.
- The proposed algorithm enhances user performance in preshaping remote grippers.
- This approach offers a promising direction for intuitive and efficient robotic manipulation.
