Towards Personalized Autonomous Driving: Deep Reinforcement Learning from Human Feedback

• Item type: SIG Technical Reports(1)
• 公開日: 2021-09-08
• タイトル: Towards Personalized Autonomous Driving: Deep Reinforcement Learning from Human Feedback
• タイトル / 言語: en
• タイトル: Towards Personalized Autonomous Driving: Deep Reinforcement Learning from Human Feedback
• 言語: eng
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Faculty of Science and Engineering, Waseda University
• 著者所属: Faculty of Science and Engineering, Waseda University
• 著者所属: Presently with National Institute of Informatics／Faculty of Science and Engineering, Waseda University
• 著者所属: Presently with National Institute of Informatics／Faculty of Science and Engineering, Waseda University
• 著者所属(英): en
• 著者所属(英): Faculty of Science and Engineering, Waseda University
• 著者所属(英): en
• 著者所属(英): Faculty of Science and Engineering, Waseda University
• 著者所属(英): en
• 著者所属(英): Presently with National Institute of Informatics / Faculty of Science and Engineering, Waseda University
• 著者所属(英): en
• 著者所属(英): Presently with National Institute of Informatics / Faculty of Science and Engineering, Waseda University
• 著者名: Jiali, Ling; Jialong, Li; Kenji, Tei; Shinichi, Honiden
• 著者名(英): Jiali, Ling; Jialong, Li; Kenji, Tei; Shinichi, Honiden
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: In modern society, personalization is one of the important indicators to attract customers. And this is the same in the field of autonomous driving. Personalized autonomous driving can not only meet the different passengers' riding preferences but also relieve the pressure and distrust caused by autonomous driving to a certain extent. In this research, We regard human as another agent, and vehicles and humans are in a cooperative relationship. And we propose a composite reward model based on reinforcement learning, which combines the passenger's feedback on autonomous driving behavior. The system proposed in this study can learn personalized driving behavior based on passenger feedback.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: In modern society, personalization is one of the important indicators to attract customers. And this is the same in the field of autonomous driving. Personalized autonomous driving can not only meet the different passengers' riding preferences but also relieve the pressure and distrust caused by autonomous driving to a certain extent. In this research, We regard human as another agent, and vehicles and humans are in a cooperative relationship. And we propose a composite reward model based on reinforcement learning, which combines the passenger's feedback on autonomous driving behavior. The system proposed in this study can learn personalized driving behavior based on passenger feedback.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA11135936
• 書誌情報: 研究報告知能システム（ICS）; 巻 2021-ICS-204, 号 11, p. 1-3, 発行日 2021-09-08
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-885X
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会