Higaki Takefumi

ヒガキタケフミ | Higaki Takefumi

Affiliation and department：
Job title： Assistant Professor

Research field (1)： Marine and maritime engineering

Research keywords (6)： Surrogate Model , Automatic Berthing , Imitation Learning , Inverse Reinforcement Learning , Automatic Collision Avoidance , Particle Method

Research theme for competitive and other funds (2)：

2022 - 2024 船舶の安全航行に関連する研究 -深層逆強化学習による熟練船長の模倣-
2021 - 2022 自動運航船を実現する自律操船 AIの開発:深層逆強化学習による熟練者の模倣

Papers (10)：

Takefumi Higaki, Yuki Tanabe, Hirotada Hashimoto, Takahito Iida. Enhancing Graph Neural Network-Based Surrogate Models for Lagrangian Fluid Simulations: Accurate Pressure Prediction, Flexible Boundary Condition Handling, and Time-Step Convergence. SSRN (preprint). 2024
Takefumi Higaki, Hirotada Hashimoto. Berthing Assistance Method for Automated Docking by Backward-Time Imitation Learning and Kernel Density Estimation Based on AIS Data. SSRN (preprint). 2024
Takefumi Higaki, Hiroki Nobe, Hirotada Hashimoto. Human-like Automatic Berthing System based on Imitative Trajectory Plan and Tracking Control. Proceedings of the OCEANS 2024-Singapore. 2024. 1-5
Takefumi Higaki, Hirotada Hashimoto, Hiroki Nobe. Online Berthing Trajectory Planner With Weak Constraint Based on AIS. SSRN (preprint). 2023
Takefumi Higaki, Hirotada Hashimoto. Human-like route planning for automatic collision avoidance using generative adversarial imitation learning. Applied Ocean Research. 2023. 138. 103620. 1-12

ｍore...

MISC (9)：

Yuki Tanabe, Takefumi Higaki, Hirotada Hashimoto. Proposal of a Versatile Surrogate Model for Particle Methods Using Graph Neural Network. Proceedings of The Annual Spring Meeting of JASNAOE 2024. 2024. 38. 487-488
Takefumi Higaki, Hiroki Nobe, Hirotada Hashimoto. Captains-Like Berthing System Based on Imitation Learning and Deep Reinforcement Learning. Proceedings of The Annual Spring Meeting of JASNAOE 2024. 2024. 38. 1-2
Takefumi Higaki, Hirotada Hashimoto. Human-Supporting Berthing Trajectory by Imitation Learning in Backward-Time Simulation. Proceedings of The Annual Autumn Meeting of JASNAOE 2023. 2023. 37. 65-66
Hirotada Hashimoto, Takefumi Higaki, Hitoshi Yoshioka. Recent Developments in Ship Collision Avoidance towards Autonomous Ships. Proceedings of The Annual Autumn Meeting of JASNAOE 2023. 2023. 37. 7-10
Kazuki Furusho, Hirotada Hashimoto, Takefumi Higaki. Development of Fluid-Solid Multiphase Flow Solver with Particle Method. Proceedings of The Annual Spring Meeting of JASNAOE 2023. 2023. 36. 483-484

ｍore...

Lectures and oral presentations (10)：

逆強化学習および模倣学習による船⻑らしい操船行動の獲得
(第140回テクノラボツアー 2024)
Investigation and Imitation of Expert Decision-Making by Captains
(International Symposium of the Graduate School of Engineering, Osaka Metropolitan University 2023)
Data-Driven Approaches for Safe Navigation
(US-Japan Joint Workshop on Ship Hydrodynamics 2023)
熟練船長による避航操船行動の解明と模倣
(日本船舶海洋工学会関西支部学生研究発表会 2022)
Our Recent Studies on Automatic Navigation
(Research Topics Exchange with TU Berlin 2022)

ｍore...

Education (3)：

2021 - 2024 Osaka Prefecture University Graduate School of Engineering Division of Aerospace and Marine-System Engineering
2017 - 2019 Osaka University Graduate School of Engineering Division of Global Architecture
2013 - 2017 Osaka University School of Engineering Division of Global Architecture

Professional career (1)：

Ph.D. (Eng.) (Osaka Prefecture University)

Work history (3)：

2024/04 - 現在 Osaka Metropolitan University Assistant Professor
2022/04 - 2024/03 Japan Society for the Promotion of Science Research Fellowships for Young Researchers (DC1)
2019/04 - 2021/07 SoftBank Corp.

Awards (6)：

2023/09 - International Symposium of the Graduate School of Engineering, Osaka Metropolitan University Excellent Poster Award Investigation and Imitation of Expert Decision-Making by Captains
2023/03 - 大阪府立大学工学研究科学生顕彰
2022/12 - 日本船舶海洋工学会関西支部支部長賞(奨励) 熟練船長による避航操船行動の解明と模倣
2022/05 - 日本船舶海洋工学会若手優秀講演賞自動避航操船のための最適航路計画の策定 -逆強化学習による熟練船長の模倣-
2019/03 - 日本船舶海洋工学会奨学褒章

Show all

Association Membership(s) (2)：

The Japanese Society for Artificial Intelligence , The Japan Society of Naval Architects and Ocean Engineers

※ Researcher’s information displayed in J-GLOBAL is based on the information registered in researchmap. For details, see here.

Return to Previous Page