# Detecting Transitions from Stability to Instability in Robotic Grasping Based on Tactile Perception

## TL;DR

This paper introduces a tactile perception system that detects when a robot's grasp becomes unstable during object interactions.

## Contribution

A novel real-time dynamic state sensing network combining CNNs and LSTMs for robotic grasping stability detection.

## Key findings

- The network achieves 98.90% average classification accuracy in predicting stability transitions.
- It maintains high accuracy with previously unseen objects, showing robust generalization.
- The system operates in real-time with 31.84 ms per inference step.

## Abstract

Robots execute diverse load operations, including carrying, lifting, tilting, and moving objects, involving load changes or transfers. This dynamic process can result in the shift of interactive operations from stability to instability. In this paper, we respond to these dynamic changes by utilizing tactile images captured from tactile sensors during interactions, conducting a study on the dynamic stability and instability in operations, and propose a real-time dynamic state sensing network by integrating convolutional neural networks (CNNs) for spatial feature extraction and long short-term memory (LSTM) networks to capture temporal information. We collect a dataset capturing the entire transition from stable to unstable states during interaction. Employing a sliding window, we sample consecutive frames from the collected dataset and feed them into the network for the state change predictions of robots. The network achieves both real-time temporal sequence prediction at 31.84 ms per inference step and an average classification accuracy of 98.90%. Our experiments demonstrate the network’s robustness, maintaining high accuracy even with previously unseen objects.