Tactile Regrasp: Grasp Adjustments via Simulated Tactile Transformations

TL;DR

This paper introduces a tactile-based regrasp control policy that uses simulated tactile transformations and a neural network to improve grasp success rates in robotic manipulation.

Contribution

It presents a novel method combining tactile sensing, simulated transformations, and deep learning for local grasp adjustments in robotics.

Findings

Grasp quality network predicts grasp outcomes with 85% accuracy on known objects.

Regrasp policy increases grasp success rate by 70% on test objects.

Method effectively improves grasp stability through tactile simulation and neural evaluation.

Abstract

This paper presents a novel regrasp control policy that makes use of tactile sensing to plan local grasp adjustments. Our approach determines regrasp actions by virtually searching for local transformations of tactile measurements that improve the quality of the grasp. First, we construct a tactile-based grasp quality metric using a deep convolutional neural network trained on over 2800 grasps. The quality of each grasp, a continuous value between 0 and 1, is determined experimentally by measuring its resistance to external perturbations. Second, we simulate the tactile imprints associated with robot motions relative to the initial grasp by performing rigid-body transformations of the given tactile measurements. The newly generated tactile imprints are evaluated with the learned grasp quality network and the regrasp action is chosen to maximize the grasp quality. Results show that the grasp quality network can predict the outcome of grasps with an average accuracy of 85% on known objects and 75% on a cross validation set of 12 objects. The regrasp control policy improves the success rate of grasp actions by an average relative increase of 70% on a test set of 8 objects.