Related Experiment Video
Updated: Mar 6, 2026
09:41
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
Published on: April 21, 2023
2.3K
3-D-Gaze-Based Robotic Grasping Through Mimicking Human Visuomotor Function for People With Motion Impairments
IEEE Transactions on Bio-Medical Engineering
|March 10, 2017
Summary
This study introduces a new 3-D gaze-based system for intuitive human-robot interaction. Users with mobility impairments can now control robots for object manipulation simply by looking at targets.
Area of Science:
- Robotics
- Human-Computer Interaction
- Computer Vision
Background:
- Traditional human-robot interaction methods can be challenging for individuals with motion impairments.
- Gaze direction inherently contains information about user intent for object manipulation.
Purpose of the Study:
- To develop a novel 3-D gaze-based interaction modality for intuitive robot control.
- To enable users with motion impairments to command robots by looking at real-world objects.
Main Methods:
- Developed a novel gaze vector method for accurate 3-D coordinate estimation of gazed-at objects.
- Designed an interpretation framework mimicking human visuomotor functions for enhanced gaze control.
Main Results:
- Achieved high tracking accuracy using the developed gaze vector method.
- Demonstrated successful robotic arm control for object grasping via direct gaze.
Conclusions:
- 3-D gaze tracking offers an effective and intuitive interaction modality for robotic object manipulation.
- This technology holds significant promise for assisting individuals with disabilities and the elderly in human-robot interaction.
