Deep Learning a Grasp Function for Grasping under Gripper Pose Uncertainty

TL;DR

This paper introduces a novel deep learning approach that predicts a grasp function over all possible grasp poses from depth images, enabling robust grasping under significant gripper pose uncertainty by smoothing the grasp scores with an uncertainty model.

Contribution

It proposes a new method to predict grasp scores for all poses, improving robustness to pose uncertainty compared to traditional single-pose prediction methods.

Findings

The learned grasp function is more robust to gripper pose uncertainty.

Simulation-based training enables extensive data generation without real-world experiments.

Experimental results show improved grasp success under pose uncertainty.

Abstract

This paper presents a new method for parallel-jaw grasping of isolated objects from depth images, under large gripper pose uncertainty. Whilst most approaches aim to predict the single best grasp pose from an image, our method first predicts a score for every possible grasp pose, which we denote the grasp function. With this, it is possible to achieve grasping robust to the gripper's pose uncertainty, by smoothing the grasp function with the pose uncertainty function. Therefore, if the single best pose is adjacent to a region of poor grasp quality, that pose will no longer be chosen, and instead a pose will be chosen which is surrounded by a region of high grasp quality. To learn this function, we train a Convolutional Neural Network which takes as input a single depth image of an object, and outputs a score for each grasp pose across the image. Training data for this is generated by use of physics simulation and depth image simulation with 3D object meshes, to enable acquisition of sufficient data without requiring exhaustive real-world experiments. We evaluate with both synthetic and real experiments, and show that the learned grasp score is more robust to gripper pose uncertainty than when this uncertainty is not accounted for.