Reinforcement Learning for Collision-Free Motion Control

碩士 === 國立交通大學 === 控制工程系 === 82 === This thesis applies reinforcement learning to the collision- free motion control problem. There are two general reinforcement learning methods, the direct and indirect method. We present a direct method, t...

Full description

Bibliographic Details
Main Authors: Yii-Neng Hu, 胡以能
Other Authors: Chi-Cheng Jou
Format: Others
Language:en_US
Published: 1994
Online Access:http://ndltd.ncl.edu.tw/handle/92447171374639621352
id ndltd-TW-082NCTU0327029
record_format oai_dc
spelling ndltd-TW-082NCTU03270292016-07-18T04:09:34Z http://ndltd.ncl.edu.tw/handle/92447171374639621352 Reinforcement Learning for Collision-Free Motion Control 加強式學習於避碰運動控制之應用 Yii-Neng Hu 胡以能 碩士 國立交通大學 控制工程系 82 This thesis applies reinforcement learning to the collision- free motion control problem. There are two general reinforcement learning methods, the direct and indirect method. We present a direct method, the stochastic learning scheme, and an indirect method, the model-based learning scheme. The focus is on justify the effectiveness and efficiency of these schemes on the robot collision-free motion control problem. The results of simulations show that the stochastic learning scheme outperforms the model-based learning scheme, but both learning methods have their own limitations and disadvantages. It is suggested that reinforcement learning control is an effective alternative in dealing with less-structured control problems. Chi-Cheng Jou 周志成 1994 學位論文 ; thesis 74 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 國立交通大學 === 控制工程系 === 82 === This thesis applies reinforcement learning to the collision- free motion control problem. There are two general reinforcement learning methods, the direct and indirect method. We present a direct method, the stochastic learning scheme, and an indirect method, the model-based learning scheme. The focus is on justify the effectiveness and efficiency of these schemes on the robot collision-free motion control problem. The results of simulations show that the stochastic learning scheme outperforms the model-based learning scheme, but both learning methods have their own limitations and disadvantages. It is suggested that reinforcement learning control is an effective alternative in dealing with less-structured control problems.
author2 Chi-Cheng Jou
author_facet Chi-Cheng Jou
Yii-Neng Hu
胡以能
author Yii-Neng Hu
胡以能
spellingShingle Yii-Neng Hu
胡以能
Reinforcement Learning for Collision-Free Motion Control
author_sort Yii-Neng Hu
title Reinforcement Learning for Collision-Free Motion Control
title_short Reinforcement Learning for Collision-Free Motion Control
title_full Reinforcement Learning for Collision-Free Motion Control
title_fullStr Reinforcement Learning for Collision-Free Motion Control
title_full_unstemmed Reinforcement Learning for Collision-Free Motion Control
title_sort reinforcement learning for collision-free motion control
publishDate 1994
url http://ndltd.ncl.edu.tw/handle/92447171374639621352
work_keys_str_mv AT yiinenghu reinforcementlearningforcollisionfreemotioncontrol
AT húyǐnéng reinforcementlearningforcollisionfreemotioncontrol
AT yiinenghu jiāqiángshìxuéxíyúbìpèngyùndòngkòngzhìzhīyīngyòng
AT húyǐnéng jiāqiángshìxuéxíyúbìpèngyùndòngkòngzhìzhīyīngyòng
_version_ 1718351527430586368