Jun Umeda
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jun Umeda has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Toshifumi Fujiwara
National Maritime Research Institute (NMRI)
3 shared publications - Kangsoo Kim
Offshore Advanced Technology Department, National Maritime Research Institute, National Institute of Maritime, Port and Aviation Technology, 6-38-1 Shinkawa, Mitaka, Tokyo 181-0004, Japan
2 shared publications - Shogo Inaba2 shared publications
- Takumi Sato2 shared publications
- Burak Yıldız
Yildiz Technical University
1 shared publication - Hirotada Hashimoto
Osaka Metropolitan University
1 shared publication
Latest Publications
- Open Access
- 1 June 2024
- Published by Elsevier in Ocean Engineering
- Vol. 308
Experimental validation of data-driven reactive control strategy for wave energy converters: A Gaussian process regression approach
- 1 June 2024
- Published by Elsevier in Ocean Engineering
- Vol. 308
Open AccessExperimental validation of data-driven reactive control strategy for wave energy converters: A Gaussian process regression approach
- Open Access
- 1 April 2023
- Published by Springer Nature in Journal of Marine Science and Technology
- Vol. 28 (2) , 439-451
Development of model-based and model-free reactive control scheme: considering copper loss and movable-floater-displacement constraint for a wave energy converter
The utilization of wave energy is expected since ocean wave energy has a high potential. The improvement of the feasibility of wave energy converters requires control that maximizes the electric output energy, including the copper loss under a displacement constraint. Several model-based and model-free reactive controls have been developed. Although model-based reactive control attains high performance, it struggles to deal with modeling errors and forecasting wave excitation forces. On the other hand, the model-free reactive control can adapt to dynamic modeling, including modeling errors; however, it requires a vast amount of learning data and considerable time and effort to consider the displacement constraint. Model-based and model-free reactive controls each have advantages and disadvantages. Combined model-based and model-free reactive controls are desirable to freely switch between the model-based and model-free reactive controls based on various ocean situations. In this study, two equivalent model-based and model-free reactive controls that can consider the copper loss and displacement constraints without forecasting the wave excitation forces were proposed. The model-free reactive control was compared with the model-based reactive control and a conventional control using numerical simulations in irregular waves. The results of the simulation show that the proposed model-based reactive control achieves superior performance compared to that of the conventional control. The proposed model-free reactive control achieved comparable performance to that of the proposed model-based reactive control under various wave conditions. Moreover, the proposed model-free reactive control decreased the required training trials. The development of the two equivalent control schemes will lead to the proposal of combined model-based and model-free reactive controls in the future.
- 1 April 2023
- Published by Springer Nature in Journal of Marine Science and Technology
- Vol. 28 (2) , 439-451
Open AccessDevelopment of model-based and model-free reactive control scheme: considering copper loss and movable-floater-displacement constraint for a wave energy converter
The utilization of wave energy is expected since ocean wave energy has a high potential. The improvement of the feasibility of wave energy converters requires control that maximizes the electric output energy, including the copper loss under a displacement constraint. Several model-based and model-free reactive controls have been developed. Although model-based reactive control attains high performance, it struggles to deal with modeling errors and forecasting wave excitation forces. On the other hand, the model-free reactive control can adapt to dynamic modeling, including modeling errors; however, it requires a vast amount of learning data and considerable time and effort to consider the displacement constraint. Model-based and model-free reactive controls each have advantages and disadvantages. Combined model-based and model-free reactive controls are desirable to freely switch between the model-based and model-free reactive controls based on various ocean situations. In this study, two equivalent model-based and model-free reactive controls that can consider the copper loss and displacement constraints without forecasting the wave excitation forces were proposed. The model-free reactive control was compared with the model-based reactive control and a conventional control using numerical simulations in irregular waves. The results of the simulation show that the proposed model-based reactive control achieves superior performance compared to that of the conventional control. The proposed model-free reactive control achieved comparable performance to that of the proposed model-based reactive control under various wave conditions. Moreover, the proposed model-free reactive control decreased the required training trials. The development of the two equivalent control schemes will lead to the proposal of combined model-based and model-free reactive controls in the future.
- 29 March 2023
- Published by Springer Nature
- p. 475-492
Study on Bilge Keel Component of Roll Damping for Non-periodic Motion
- 29 March 2023
- Published by Springer Nature
- p. 475-492
Study on Bilge Keel Component of Roll Damping for Non-periodic Motion
- Open Access
- 16 July 2021
- Published by ASME International
Path Following Control of Autonomous Underwater Vehicle Using Nonlinear Model Predictive Control
The autonomous underwater vehicles (AUV) have been widely used for various missions such as ocean sampling and underwater resource survey and so on. The design of a motion controller of AUVs plays an important role in the successes of the missions. In this study, a nonlinear model predictive controller for path following control of the AUV subject to inputs constrains, thruster
- 16 July 2021
- Published by ASME International
Open AccessPath Following Control of Autonomous Underwater Vehicle Using Nonlinear Model Predictive Control
The autonomous underwater vehicles (AUV) have been widely used for various missions such as ocean sampling and underwater resource survey and so on. The design of a motion controller of AUVs plays an important role in the successes of the missions. In this study, a nonlinear model predictive controller for path following control of the AUV subject to inputs constrains, thruster
- Open Access
- 16 July 2021
- Published by ASME International
Sea Trials and Simulations for Leader Follower Formation Control of Cruising-Type Auvs
Autonomous underwater vehicle (AUV) is promising technology for surveying and exploration of the oceans and many relevant technologies are studied around the world. Japanese government promotes the R&D project of developing the system and strategy for simultaneous deployment and operation of multiple AUVs. National maritime research institute, Japan (NMRI) developed a
- 16 July 2021
- Published by ASME International
Open AccessSea Trials and Simulations for Leader Follower Formation Control of Cruising-Type Auvs
Autonomous underwater vehicle (AUV) is promising technology for surveying and exploration of the oceans and many relevant technologies are studied around the world. Japanese government promotes the R&D project of developing the system and strategy for simultaneous deployment and operation of multiple AUVs. National maritime research institute, Japan (NMRI) developed a
- Open Access
- 1 January 2020
- Published by Japan Society of Naval Architects and Ocean Engineers in Journal of the Japan Society of Naval Architects and Ocean Engineers
- Vol. 31, 229-238
Deep Reinforcement Learning Control to Maximize Output Energy for a Wave Energy Converter
This paper presents a deep reinforcement learning control method to maximize output energy for a point absorber type wave energy converter (WEC) with a linear generator. Conventional control methods require the dynamic model of the WEC. Modeling errors of the dynamic model, however, make energy absorption smaller and cause incorrect control. The proposed method, which is a model-free control method learns the optimal damping and stiffness coefficients based on experiences. In the proposed control method, damping and stiffness coefficients are able to vary in time-domain depending on the incident waves by deep reinforcement learning. The performance of the proposed control method is investigated through numerical simulation in both regular and irregular waves. Compared with the conventional control method, averaged output power increased, and the power fluctuation decreased without the dynamic model. It is understood that the proposed method is more effective than the conventional control method.
- 1 January 2020
- Published by Japan Society of Naval Architects and Ocean Engineers in Journal of the Japan Society of Naval Architects and Ocean Engineers
- Vol. 31, 229-238
Open AccessDeep Reinforcement Learning Control to Maximize Output Energy for a Wave Energy Converter
This paper presents a deep reinforcement learning control method to maximize output energy for a point absorber type wave energy converter (WEC) with a linear generator. Conventional control methods require the dynamic model of the WEC. Modeling errors of the dynamic model, however, make energy absorption smaller and cause incorrect control. The proposed method, which is a model-free control method learns the optimal damping and stiffness coefficients based on experiences. In the proposed control method, damping and stiffness coefficients are able to vary in time-domain depending on the incident waves by deep reinforcement learning. The performance of the proposed control method is investigated through numerical simulation in both regular and irregular waves. Compared with the conventional control method, averaged output power increased, and the power fluctuation decreased without the dynamic model. It is understood that the proposed method is more effective than the conventional control method.