Spatial Features-Based Slip Detection in Neuromorphic Vision Tactile Sensors.

Area of Science:

• Robotics
• Neuromorphic Engineering
• Sensor Technology

Background:

• Intelligent robotics and human-machine interaction are rapidly advancing, driven by neuromorphic technologies.
• Detecting incipient slip is critical for stable grip and object safety during robotic manipulation.

Purpose of the Study:

• To propose a novel framework for incipient slip detection using a bioinspired neuromorphic vision-based tactile sensor.
• To analyze spatial features for discriminating slip from other manipulation actions.

Main Methods:

• Development of a bioinspired neuromorphic vision-based tactile sensor.
• Extraction and analysis of spatial features for slip detection.
• Classification of manipulation actions using a Random Forest classifier.

Main Results:

• The proposed spatial features effectively discriminate incipient slip from pressing and proximity actions.
• The framework achieved an overall accuracy of 94.19% on unseen datasets.
• Successful detection of incipient slip under minimal contact conditions on smooth surfaces.

Conclusions:

• The novel framework demonstrates high accuracy in detecting incipient slip.
• This technology has the potential to significantly enhance precision in robotic manipulation.
• The bioinspired sensor and spatial feature analysis offer a promising approach for intelligent robotics.