A Real-time Robotic Grasp Approach with Oriented Anchor Box

TL;DR

This paper presents a real-time vision-based robotic grasp detection method using oriented anchor boxes and a novel angle regression mechanism, achieving high accuracy and speed in both simulation and real-world tests.

Contribution

It introduces the Orientation Anchor Box Mechanism and Angle Matching for improved grasp angle prediction, advancing the accuracy and efficiency of robotic grasp detection.

Findings

Achieves 98.8% accuracy in image-wise split and 97.8% in object-wise split.

Runs at 67 FPS on GTX 1080Ti, outperforming current methods.

Demonstrates 90% success rate in real-world robotic experiments.

Abstract

Grasp is an essential skill for robots to interact with humans and the environment. In this paper, we build a vision-based, robust and real-time robotic grasp approach with fully convolutional neural network. The main component of our approach is a grasp detection network with oriented anchor boxes as detection priors. Because the orientation of detected grasps is significant, which determines the rotation angle configuration of the gripper, we propose the Orientation Anchor Box Mechanism to regress grasp angle based on predefined assumption instead of classification or regression without any priors. With oriented anchor boxes, the grasps can be predicted more accurately and efficiently. Besides, to accelerate the network training and further improve the performance of angle regression, Angle Matching is proposed during training instead of Jaccard Index Matching. Five-fold cross validation results demonstrate that our proposed algorithm achieves an accuracy of 98.8% and 97.8% in image-wise split and object-wise split respectively, and the speed of our detection algorithm is 67 FPS with GTX 1080Ti, outperforming all the current state-of-the-art grasp detection algorithms on Cornell Dataset both in speed and accuracy. Robotic experiments demonstrate the robustness and generalization ability in unseen objects and real-world environment, with the average success rate of 90.0% and 84.2% of familiar things and unseen things respectively on Baxter robot platform.