Related Experiment Video
Updated: Jul 6, 2025
09:41
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
Published on: April 21, 2023
1.6K
EEG-controlled tele-grasping for undefined objects
Minki Kim1, Myoung-Su Choi2, Ga-Ram Jang2
1Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea.
Frontiers in Neurorobotics
|January 8, 2024
Summary
This study introduces a robot grasping system using real-time electroencephalography (EEG) and shared autonomy. It enables human operators to make critical decisions for grasping uncertain objects, enhancing robotic capabilities in unstructured environments.
Area of Science:
- Robotics
- Neuroscience
- Human-Computer Interaction
Background:
- Fully autonomous grasping struggles with undefined objects in unstructured environments.
- Real-time human decision-making is crucial for uncertain robotic tasks.
- Teleoperation systems require intuitive human control interfaces.
Purpose of the Study:
- To develop a teleoperation system for robot grasping of undefined objects.
- To integrate real-time electroencephalography (EEG) measurements for human control.
- To implement shared autonomy allowing human decisions throughout the grasping process.
Main Methods:
- A teleoperation system utilizing shared autonomy and real-time EEG.
- Steady-state visually evoked potentials (SSVEP) generated by six flickering blocks for human input.
- Division of the grasping task into predefined substeps (approach, posture/force selection, grasp, transport, release).
- A graphical user interface (GUI) displaying task status and control options.
Main Results:
- Demonstrated successful tele-grasping of various objects using human decisions via EEG.
- Enabled human control over gripper 3D movement, grasping posture (4 options), and grip force (3 levels).
- The system effectively integrates human decision-making into sequential robotic tasks.
Conclusions:
- The developed EEG-based teleoperation system enhances robot grasping capabilities for undefined objects.
- Shared autonomy allows for flexible human intervention in complex robotic tasks.
- This approach is adaptable to other sequential teleoperation tasks requiring intricate human decisions.
