Robotics Dexterous Grasping: The Methods Based on Point Cloud and Deep Learning

被引：36

作者：

Duan, Haonan ^{[1
,2
,3
]}

Wang, Peng ^{[1
,3
,4
]}

Huang, Yayu ^{[1
,3
]}

Xu, Guangyun ^{[1
,3
]}

Wei, Wei ^{[1
,3
]}

Shen, Xiaofei ^{[1
,3
]}

机构：

[1] Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing, Peoples R China

[2] Univ Pittsburgh, Dept Informat Sci, Sch Comp & Informat, Pittsburgh, PA 15260 USA

[3] Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China

[4] Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Shanghai, Peoples R China

来源：

FRONTIERS IN NEUROROBOTICS | 2021年 / 15卷

基金：

中国国家自然科学基金;

关键词：

robotics; dexterous grasping; point cloud; deep learning; review; 3-DIMENSIONAL OBJECT RECOGNITION; NEURAL-NETWORKS; POSE ESTIMATION; MANIPULATION; MODEL; REGISTRATION; AFFORDANCES; STRATEGIES; DATASET; PICKING;

D O I：

10.3389/fnbot.2021.658280

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Dexterous manipulation, especially dexterous grasping, is a primitive and crucial ability of robots that allows the implementation of performing human-like behaviors. Deploying the ability on robots enables them to assist and substitute human to accomplish more complex tasks in daily life and industrial production. A comprehensive review of the methods based on point cloud and deep learning for robotics dexterous grasping from three perspectives is given in this paper. As a new category schemes of the mainstream methods, the proposed generation-evaluation framework is the core concept of the classification. The other two classifications based on learning modes and applications are also briefly described afterwards. This review aims to afford a guideline for robotics dexterous grasping researchers and developers.

引用

页数：27

共 230 条

[1] A Haptic Shared-Control Architecture for Guided Multi-Target Robotic Grasping [J].