DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Shim, David Hyunchul | - |
dc.contributor.advisor | 심현철 | - |
dc.contributor.author | Seong, Hyunki | - |
dc.date.accessioned | 2022-04-15T07:58:12Z | - |
dc.date.available | 2022-04-15T07:58:12Z | - |
dc.date.issued | 2021 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=948427&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/295096 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 로봇공학학제전공, 2021.2,[iv, 37 p. :] | - |
dc.description.abstract | Driving at an unsignalized intersection is a complicated driving scenario that requires both safety and traffic efficiency. In the intersection, the driving policy does not simply maintain the safe distance to all vehicles, but should pay more attention to the vehicles that are likely to cross with the ego vehicle and make its decision considering their intentions. Our goal is to train an attention-based driving policy for handling intersection scenarios using deep reinforcement learning. By leveraging the attention, our policy is able to learn how to focus on spatially and temporally more important features in its egocentric observation and perform complex driving strategies at the congested intersection environment. We transfer the policy model trained in a high fidelity simulator to a full-scale vehicle system, and conduct experiments to evaluate our model in simulated and real-world environments. Our model successfully performs various intersection scenarios even with noisy sensory data and delayed response. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Reinforcement Learning▼aDecision Making▼aDeep learning▼aAutonomous Driving | - |
dc.subject | 강화학습▼a의사결정▼a딥러닝▼a자율주행 | - |
dc.title | Learning to drive at unsignalized intersections using deep reinforcement learning | - |
dc.title.alternative | 심층강화학습을 이용한 비보호 교차로 자율 주행 학습 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :로봇공학학제전공, | - |
dc.contributor.alternativeauthor | 성현기 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.