Japanese / English

Education

• Apr. 2019 - Mar. 2021 Meiji University, Japan
  • Master of Engineering in Mechanical Engineering
  • GPA: 4.0/4.0
  • Supervisor: Prof. Yoji Kuroda (in Autonomous Mobile Systems Laboratory)
  • Thesis: “Real-Time Self-Posture Estimation Using DNN Learning Landscape Knowledge” (written in Japanese)
  • Research activities:
    • Innovated self-posture estimation algorithms for mobile robots using:
      • neural network that takes camera data as input.
      • neural network that takes LiDAR data as input.
      • neural network that takes camera and LiDAR data as input.
      • neural network that takes camera data as input and common LiDAR features.
    • Innovated Landmark SLAM (Simultaneous Localization and Mapping) using planar features measured by LiDAR.
• Aug. 2017 - Jan. 2018 California State University, Long Beach, USA
  • Exchange Program in Mechanical & Aerospace Engineering
  • GPA: 3.0/4.0
    • Design Analysis ME Systems I: 3.0
    • Flight Mechanics: 3.0
    • Propulsion: 3.0
    • Power Plant Design: 3.0
• Apr. 2015 - Mar. 2019 Meiji University, Japan
  • Bachelor of Engineering in Mechanical Engineering
  • GPA: 3.75/4.0 (awarded)
  • Supervisor: Prof. Yoji Kuroda (in Autonomous Mobile Systems Laboratory)
  • Thesis: “Real-Time 6DoF Pose Estimation Using Relative Posture with Respect to Building Walls” (written in Japanese)
  • Research activities:
    • Innovated self-posture estimation algorithms for mobile robots based on common knowledge that planes of buildings are vertical.
    • Competed in autonomous mobile robot competition called Tsukuba Challenge.

Professional Experience

• Apr. 2021 - Present Komatsu Ltd., Japan
  Automation System Engineer/Researcher
  • Developing obstacle detection system for dump trucks using cameras and deep learning.
  • Developing obstacle detection system for dump trucks using LiDARs and radars.
  • Innovated blade control algorithm for bulldozers using reinforcement learning.
  • Developed bulldozer simulator.
  • Innovated odometry estimation algorithm for bulldozers using machine learning.
  • Developed obstacle detection system for bulldozers using LiDARs and radars.
  • Developed mapping system for bulldozers using GNSS and LiDARs.

Teaching/Research Assistant Experience

• Apr. 2019 - Mar. 2020 Meiji University, Japan
  Research assistant in project “Research and Development of Intelligent Behavior Emergence in Robots through Knowledge Structuring,” organized by New Energy and Industrial Technology Development Organization (NEDO)
• Apr. 2019 - Sep. 2019 Meiji University, Japan
  Teaching assistant in course “Mechanical Engineering Experiments 2”

Awards

• Nov. 2024 Best Presentation Award, Annual Conference of the IEEE Industrial Electronics Society (IECON)
  Awarded to one presentation in each conference.
• Mar. 2021 Miura Award, Japan Society of Mechanical Engineers (JSME)
  Awarded annually to one student from each mechanical engineering department at universities in Japan who has achieved the most research accomplishments.
• Mar. 2019 Hatakeyama Award, Japan Society of Mechanical Engineers (JSME)
  Awarded annually to one student from each mechanical engineering department at universities in Japan who earned outstanding credits.
• Mar. 2018 Hasegawa Hisao Award for The Best Presentation of Bachelor’s Thesis, Meiji University
  Awarded annually to three undergraduate students in Mechanical Engineering at Meiji University.
• Apr. 2017 Hasegawa Hisao Award for Outstanding Academic Achievement, Meiji University
  Awarded annually to one undergraduate student in Mechanical Engineering at Meiji University.

Scholarship

• Apr. 2019 - Mar. 2021 Exemption from Return for Particularly Outstanding Achievement, Japan Student Services Organization (JASSO)
  Awarded annually to exceptional graduate students, amounting to 2,213,000 JPY (approximately 19,955 USD at that time).
• Apr. 2019 - Mar. 2021 Graduate Research Encouragement Scholarship, Meiji University Awarded annually to exceptional graduate students, covering half of the tuition, amounting to 760,000 JPY (approximately 6,853 USD at that time).
• Apr. 2017 - Mar. 2018 Academic Achievement Scholarship, Meiji University
  Awarded annually to exceptional undergraduate students, covering a fourth of the tuition, amounting to 282,000 JPY (approximately 2,525 USD at that time).

Research Interests and Skills

• Autonomous Robotics, Machine learning, Self-localization, SLAM, Reinforcement Learning, Object Detection
• IMU, LiDAR, Camera, GNSS, Mining Machinery (e.g., Bulldozer, Dump Truck)
• C/C++, Python, ROS, PyTorch, Docker

Publications

Under Peer Review

• Hikaru Sawafuji, Ryota Ozaki, Takuto Motomura, Toyohisa Matsuda, Masanori Tojima, Kento Uchida, and Shinichi Shirakawa, “Machine Learning Based Self-Localization Estimation Using Internal Sensors for Automating Bulldozers,” Journal of Field Robotics (currently under peer review).

Journals

• Ryota Ozaki, Naoya Sugiura, and Yoji Kuroda, “LiDAR DNN based self-attitude estimation with learning landscape regularities,” ROBOMECH Journal, Vol.8, No.26, Dec. 20, 2021.
• Ryota Ozaki, and Yoji Kuroda, “Real-time self-attitude estimation using visual images and/or structures of the environments,” Transactions of the JSME (in Japanese), Vol.87, No.903, pp.21-00098, Dec. 25, 2021 (in Japanese).
• Ryota Ozaki, and Yoji Kuroda, “EKF-based self-attitude estimation with DNN learning landscape information,” ROBOMECH Journal, Vol.8, No.9, Mar. 6, 2021.
• Ryota Ozaki, and Yoji Kuroda, “EKF-based real-time self-attitude estimation with camera DNN learning landscape regularities,” IEEE Robotics and Automation Letters (RA-L), Vol.6, No.2, pp.1737-1744, Feb. 19, 2021.
  • Video Abstract
  • This was also presented in the 2021 IEEE International Conference on Robotics and Automation (ICRA 2021).
• Ryota Ozaki, and Yoji Kuroda, “Real-time 6DoF localization with relative poses to walls of buildings,” Transactions of the JSME (in Japanese), Vol.85, No.875, pp.19-00065, Jul. 25, 2019 (in Japanese).

International Conferences (w/ Peer Review)

• Ryota Ozaki, Takuto Motomura, Yutaka Nakayama, and Shouta Mori, “Blade Control for Bulk Dozer Push by Bulldozer using Reinforcement Learning,” Annual Conference of the IEEE Industrial Electronics Society (IECON), Nov. 5, 2024, Best Presentation Award.
• Ryota Ozaki, and Yoji Kuroda, “EKF-based real-time self-attitude estimation with camera DNN learning landscape regularities,” IEEE International Conference on Robotics and Automation (ICRA), Jun. 1, 2021.
  • Video Abstract
  • This was presented as Conference Option of IEEE Robotics and Automation Letters (RA-L).
• Ryota Ozaki, and Yoji Kuroda, “DNN-Based Self-Attitude Estimation by Learning Landscape Information,” IEEE/SICE International Symposium on System Integrations (SII), pp.733-738, Jan. 13, 2021.
• Ryota Ozaki, and Yoji Kuroda, “6-DoF EKF SLAM with Global Planar Features in Artificial Environments,” IEEE/SICE International Symposium on System Integrations (SII), pp.531-535, Jan. 14, 2020.

Japanese Domestic Conferences (w/ Peer Review)

• Ryota Ozaki, and Yoji Kuroda, “Camera-LiDAR-DNN-based self-attitude estimation with learning landscape regularities,” 26th Robotics Symposia, pp.294-250, Mar. 17, 2021 (in Japanese).
• Ryota Ozaki, and Yoji Kuroda, “Landmark SLAM using planar features in artificial environments,” 25th Robotics Symposia, pp.316-317, Mar. 15, 2020 (in Japanese).
• Ryota Ozaki, and Yoji Kuroda, “Pose Estimation with Relative Poses to Walls of Buildings,” 24th Robotics Symposia, pp.120-121, Mar. 15, 2019 (in Japanese).

Japanese Domestic Conferences (w/o Peer Review)

• Ryota Ozaki, Shouta Mori, Yutaka Nakayama, and Takuto Motomura, “Evaluating productivity of bulldozer using simulator,” Robotics and Mechatronics Conference, May 31, 2024 (in Japanese).
• Hikaru Sawafuji, Ryota Ozaki, Takuto Motomura, Toyohisa Matsuda, Masanori Tojima, Kento Uchida, and Shinichi Shirakawa, “Application of Machine Learning to Self-Position Estimation of Bulldozers Using Internal Sensors,” 38th Annual Conference of the Japanese Society for Artificial Intelligence, May 31, 2024 (in Japanese).
• Ryota Ozaki, and Yoji Kuroda, “EKF-SLAM with planar landmarks in artificial environments,” 20th SICE System Integration Division Annual Conference, pp.165-166, Dec. 12, 2019 (in Japanese).
• Ryota Ozaki, and Yoji Kuroda, “Pose Estimation for mobile robots with vertical walls,” Robotics and Mechatronics Conference, Jun. 6, 2019 (in Japanese).
• Ryota Ozaki, and Yoji Kuroda, “6DoF Localization with Relative Poses to Walls of Buildings,” 58th Japan Society of Mechanical Engineer (JSME) Kanto branch graduation thesis meeting, Mar. 18, 2019 (in Japanese).
• Kazuya Onda, Tomotaka Oishi, Jumpei Arima, Ryota Ozaki, Toshihiro Hayata, and Yoji Kuroda, “Development of navigation system using Edge-Node Map and intersection shape matching,” Tsukuba Challenge Symposium, pp.101-106, Jan. 19, 2019 (in Japanese).

Patents

• Ryota Ozaki, and Takuto Motomura, “{NOT YET DISCLOSED},” Pending.
• Ryota Ozaki, Takuto Motomura, and Toyohisa Matsuda, “{NOT YET DISCLOSED},” Pending.
• Ryota Ozaki, Toyohisa Matsuda, and Takuto Motomura, “{NOT YET DISCLOSED},” Pending.
• Ryota Ozaki, Takuto Motomura, and Toyohisa Matsuda, “{NOT YET DISCLOSED},” Pending.
• Ryota Ozaki, Takuto Motomura, and Toyohisa Matsuda, “{NOT YET DISCLOSED},” Pending.
• Masataka Ozaki, Yoshihiro Tagawa, and Ryota Ozaki, “System for creating current landform data of work site and method for creating current landform data of work site,” Pending.
• Masataka Ozaki, Ryota Ozaki, and Shoma Sato, “Work site detection system, and work site detection method,” Pending.
• Masataka Ozaki, Toshihide Mineushiro, Ryota Ozaki, and Takuto Motomura, “Control system for work machine and control method for work machine,” Pending.