Skip to content
Back to outputs

Learning for a robot: deep reinforcement learning, imitation learning, transfer learning

Research output: Contribution to journalLiterature reviewpeer-review

Standard

Learning for a robot : deep reinforcement learning, imitation learning, transfer learning. / Hua, Jiang; Zeng, Liangcai; Li, Gongfa; Ju, Zhaojie.

In: Sensors (Switzerland), Vol. 21, No. 4, 1278, 11.02.2021, p. 1-21.

Research output: Contribution to journalLiterature reviewpeer-review

Harvard

Hua, J, Zeng, L, Li, G & Ju, Z 2021, 'Learning for a robot: deep reinforcement learning, imitation learning, transfer learning', Sensors (Switzerland), vol. 21, no. 4, 1278, pp. 1-21. https://doi.org/10.3390/s21041278

APA

Vancouver

Author

Hua, Jiang ; Zeng, Liangcai ; Li, Gongfa ; Ju, Zhaojie. / Learning for a robot : deep reinforcement learning, imitation learning, transfer learning. In: Sensors (Switzerland). 2021 ; Vol. 21, No. 4. pp. 1-21.

Bibtex

@article{e9715e8d80cb456fbc661ef32bc96978,
title = "Learning for a robot: deep reinforcement learning, imitation learning, transfer learning",
abstract = "Dexterous manipulation of the robot is an important part of realizing intelligence, but manipulators can only perform simple tasks such as sorting and packing in a structured environment. In view of the existing problem, this paper presents a state-of-the-art survey on an intelligent robot with the capability of autonomous deciding and learning. The paper first reviews the main achievements and research of the robot, which were mainly based on the breakthrough of automatic control and hardware in mechanics. With the evolution of artificial intelligence, many pieces of research have made further progresses in adaptive and robust control. The survey reveals that the latest research in deep learning and reinforcement learning has paved the way for highly complex tasks to be performed by robots. Furthermore, deep reinforcement learning, imitation learning, and transfer learning in robot control are discussed in detail. Finally, major achievements based on these methods are summarized and analyzed thoroughly, and future research challenges are proposed.",
keywords = "Adaptive and robust control, Deep reinforcement learning, Dexterous manipulation, Imitation learning, Transfer learning",
author = "Jiang Hua and Liangcai Zeng and Gongfa Li and Zhaojie Ju",
year = "2021",
month = feb,
day = "11",
doi = "10.3390/s21041278",
language = "English",
volume = "21",
pages = "1--21",
journal = "Sensors",
issn = "1424-8220",
publisher = "Multidisciplinary Digital Publishing Institute",
number = "4",

}

RIS

TY - JOUR

T1 - Learning for a robot

T2 - deep reinforcement learning, imitation learning, transfer learning

AU - Hua, Jiang

AU - Zeng, Liangcai

AU - Li, Gongfa

AU - Ju, Zhaojie

PY - 2021/2/11

Y1 - 2021/2/11

N2 - Dexterous manipulation of the robot is an important part of realizing intelligence, but manipulators can only perform simple tasks such as sorting and packing in a structured environment. In view of the existing problem, this paper presents a state-of-the-art survey on an intelligent robot with the capability of autonomous deciding and learning. The paper first reviews the main achievements and research of the robot, which were mainly based on the breakthrough of automatic control and hardware in mechanics. With the evolution of artificial intelligence, many pieces of research have made further progresses in adaptive and robust control. The survey reveals that the latest research in deep learning and reinforcement learning has paved the way for highly complex tasks to be performed by robots. Furthermore, deep reinforcement learning, imitation learning, and transfer learning in robot control are discussed in detail. Finally, major achievements based on these methods are summarized and analyzed thoroughly, and future research challenges are proposed.

AB - Dexterous manipulation of the robot is an important part of realizing intelligence, but manipulators can only perform simple tasks such as sorting and packing in a structured environment. In view of the existing problem, this paper presents a state-of-the-art survey on an intelligent robot with the capability of autonomous deciding and learning. The paper first reviews the main achievements and research of the robot, which were mainly based on the breakthrough of automatic control and hardware in mechanics. With the evolution of artificial intelligence, many pieces of research have made further progresses in adaptive and robust control. The survey reveals that the latest research in deep learning and reinforcement learning has paved the way for highly complex tasks to be performed by robots. Furthermore, deep reinforcement learning, imitation learning, and transfer learning in robot control are discussed in detail. Finally, major achievements based on these methods are summarized and analyzed thoroughly, and future research challenges are proposed.

KW - Adaptive and robust control

KW - Deep reinforcement learning

KW - Dexterous manipulation

KW - Imitation learning

KW - Transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85100642428&partnerID=8YFLogxK

U2 - 10.3390/s21041278

DO - 10.3390/s21041278

M3 - Literature review

AN - SCOPUS:85100642428

VL - 21

SP - 1

EP - 21

JO - Sensors

JF - Sensors

SN - 1424-8220

IS - 4

M1 - 1278

ER -

ID: 26408511