Visual-tactile sensor, robot gripper including the same, and control method for the robot gripper

Robotic Gripper with Vision-Tactile Sensor Capturing Tensegrity Matrix Deformation

This technology is a vision-tactile sensor utilizing a tensegrity structure. It acquires physical information about an object by visually capturing the deformation of a matrix plane—composed of multiple contact points connected by tension members—in response to external pressure. A mirror and camera inside the sensor record the geometric changes of the plane, which are then analyzed by a processor to determine the object's gripping state and shape.

Conventional grippers rely heavily on external cameras, requiring additional space and struggling to accurately recognize complex shapes. They also have limitations in real-time detection of subtle contact information or anomalies during the gripping process.

This technology implements a tensegrity-based vision-tactile sensor mounted on the joints of a gripper finger. An internal camera captures the deformation of the matrix-arranged contact points upon contact with an object, and a processor calculates the position, posture, and orientation of these points to recognize the object and adjust finger movement. Applicable to logistics picking, precision assembly, and service robots, it enhances 3D shape estimation accuracy by providing real-time contact information at the moment of grasping.

‍

‍

Key Features:

• Vision-tactile sensor comprising a main housing with a bottom opening and a tensegrity plane installed within the opening
• Multiple contact points arranged in a matrix structure forming the tensegrity plane to detect object contact
• Tension members connecting at least two of the multiple contact points to form the tensegrity structure
• Tensegrity plane that changes shape as the position, posture, and orientation of the contact points shift upon contact with an object

‍

This invention was developed with support from the 2022 Regional Industry-Linked University Open-Lab Promotion Program funded by the Ministry of Science and ICT.