期刊
MICROMACHINES
卷 13, 期 2, 页码 -出版社
MDPI
DOI: 10.3390/mi13020293
关键词
6D pose estimation; real object judgment; pixel-wise voting network; 6D grasping robotic system
This paper presents a method for realizing an autonomous real-time 6D robotic grasping system on Kinova Gen3, integrating object detection, pose estimation, and grasping plan techniques. The system utilizes pixel-wise voting network (PV-net) for estimating the object's 6D pose, and a rapid analytical method on point cloud to judge the authenticity of the detected object. The system demonstrates stable and robust performance in various installation positions and heavily cluttered scenes.
A robotic system that can autonomously recognize object and grasp it in a real scene with heavy occlusion would be desirable. In this paper, we integrate the techniques of object detection, pose estimation and grasping plan on Kinova Gen3 (KG3), a 7 degrees of freedom (DOF) robotic arm with a low-performance native camera sensor, to implement an autonomous real-time 6 dimensional (6D) robotic grasping system. To estimate the object 6D pose, the pixel-wise voting network (PV-net), is applied in the grasping system. However, the PV-net method can not distinguish the object from its photo through only RGB image input. To meet the demands of a real industrial environment, a rapid analytical method on a point cloud is developed to judge whether the detected object is real or not. In addition, our system shows a stable and robust performance in different installation positions with heavily cluttered scenes.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据