DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Kim, Beomjoon | - |
dc.contributor.advisor | 김범준 | - |
dc.contributor.author | Kim, Minchan | - |
dc.date.accessioned | 2023-06-22T19:31:11Z | - |
dc.date.available | 2023-06-22T19:31:11Z | - |
dc.date.issued | 2023 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1032319&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/308177 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 김재철AI대학원, 2023.2,[iii, 30 p. :] | - |
dc.description.abstract | We proposed an algorithm that allows robots to manipulate non-graspable objects. Existing works such as planners require complex contact modeling, and are relatively slow at finding a plan. Our method is based on reinforcement learning, because contact modeling is not necessary and it is faster at finding actions. However, simple application of reinforcement learning is data-inefficient because a robot wastes exploration to merely touch the object. Many existing works address this issue by using a reward that encourages the robot to put its end-effector close to the object. On the other hand, we introduce the pre-contact policy, which makes initial contact with the object. Another policy called the post-contact policy manipulates the object to the goal. We show that the use of trained pre-contact policy is better than not using any initial contact or using random initial contacts. Also, we test whether our method is more data-efficient than using contact-encouraging reward in challenging manipulation problems. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Non-prehensile manipulation▼aReinforcement learning▼aSim-to-real | - |
dc.subject | 비파지형 조작▼a강화학습▼a심투리얼 | - |
dc.title | Non-prehensile manipulation by learning how to make an initial contact | - |
dc.title.alternative | 초기 접촉 방법의 학습을 통한 파지 불가능한 물체의 조작 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :김재철AI대학원, | - |
dc.contributor.alternativeauthor | 김민찬 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.