Towards advanced robotic manipulation

被引：2

作者：

Sanchez, Francisco Roldan ^{[1
,2
]}

Redmond, Stephen ^{[1
,3
]}

McGuinness, Kevin ^{[1
,2
]}

O'Connor, Noel ^{[1
,2
]}

机构：

[1] Insight SFI Res Ctr Data Analyt, Dublin, Ireland

[2] Dublin City Univ, Dublin, Ireland

[3] Univ Coll Dublin, Dublin, Ireland

来源：

2022 SIXTH IEEE INTERNATIONAL CONFERENCE ON ROBOTIC COMPUTING, IRC | 2022年

基金：

爱尔兰科学基金会;

关键词：

robotic manipulation; deep reinforcement learning; artificial intelligence;

D O I：

10.1109/IRC55401.2022.00058

中图分类号：

TP301 [理论、方法];

学科分类号：

081202 ;

摘要：

Robotic manipulation and control has increased in importance in recent years. However, state of the art techniques still have limitations when required to operate in real world applications. This paper explores Hindsight Experience Replay both in simulated and real environments, highlighting its weaknesses and proposing reinforcement-learning based alternatives based on reward and goal shaping. Additionally, several research questions are identified along with potential research directions that could be explored to tackle those questions.

引用

页码：302 / 305

页数：4

共 23 条

[21] Learning Dense Rewards for Contact-Rich Manipulation Tasks [J].

Wu, Zheng ;

Lian, Wenzhao ;

Unhelkar, Vaibhav ;

Tomizuka, Masayoshi ;

Schaal, Stefan .

2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021), 2021, :6214-6221

[22]

Yoneda T, 2021, Arxiv, DOI arXiv:2101.02842

[23] Object Detection With Deep Learning: A Review [J].

Zhao, Zhong-Qiu ;

Zheng, Peng ;

Xu, Shou-Tao ;

Wu, Xindong .

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2019, 30 (11) :3212-3232

← 1 2 3 →