Welcome to my page!
My name is Masashi Hamaya (濵屋 政志). I am a senior researcher at OMRON SINIC X Corporation. I obtained a Ph.D. degree (Eng.) at Osaka University in 2019. My current research is controlling soft robots for assemblies.
Research interests: Soft Robotics, Exoskeletons, Human-Robot Interaction, Machine Learning, and Reinforcement Learning
Contact: masashi.hamaya [at] sinicx.com
Last update: 7th Nov. 2023
News
Our paper has been accepted for NeurIPS 2023!
Three papers have been accepted on IROS 2022!
Education
Apr 2019 : Ph.D. of Engineering, Graduate School of Frontier Bioscience Osaka University
Mar 2016 : Master of Engineering, Graduate School of Frontier Bioscience Osaka University
Mar 2014: Bachelor of Engineering, Department of Mechanical Engineering Kobe University
Employment
Apr 2019 - : Senior researcher, OMRON SINIC X Corporation
Sep 2019 - : Collaborative researcher, Dept. of Brain Robot Interface, ATR Computational Neuroscience Labs
Jul 2017 - Sep 2017: Visiting scholar, Carnegie Mellon University
Apr 2017 - 2019: JSPS research fellow (DC2)
Apr 2014 - 2019: Internship, Dept. of Brain Robot Interface, ATR Computational Neuroscience Labs
Apr 2014 - 2019: Osaka University Humanware Innovation Program
Publication
Journal
Felix von Drigalski, Kazumi Kasaura, Cristian C Beltran-Hernandez, Masashi Hamaya, Kazutoshi Tanaka, Takamitsu Matsubara, "Uncertainty-aware manipulation planning using gravity and environment geometry," IEEE Robotics and Automation Letters, vol. 7, no. 4, pp. 11942-11949 , 2022
Reina Ishikawa, Masashi Hamaya, Felix von Drigalski, Kazutoshi Tanaka, Atsushi Hashimoto, "Learning by breaking: food fracture anticipation for robotic food manipulation,” IEEE ACCESS, pp. 99321-99329, 2022
Robert Lee, Masashi Hamaya, Takayuki Murooka, Yoshihisa Ijiri, Peter Corke," Sample-Efficient Learning of Deformable Linear Object Manipulation in the Real World Through Self-Supervision," IEEE Robotics and Automation Letters, vol. 7, no.1, pp. 573-580, 2021
Masashi Hamaya, Kazutoshi Tanaka, Yoshiya Shibata, Felix Wolf Hans Erich von Drigalski, Chisato Nakashima, Yoshihisa Ijiri, "Robotic Learning From Advisory and Adversarial Interactions Using a Soft Wrist," IEEE Robotics and Automation Letters, vol. 6, no.2, pp. 3878-3885, 2021 [Open Access]
Masashi Hamaya, Takamitsu Matsubara, Tatsuya Teramae, Tomoyuki Noda, and Jun Morimoto, "Design of physical user–robot interactions for model identification of soft actuators on exoskeleton robots," The International Journal of Robotics Research, vol. 40, no. 1, pp. 397-410, 2021 [Open Access] [URL]
Masashi Hamaya, Takamitsu Matsubara, Tomoyuki Noda, Tatsuya Teramae, and Jun Morimoto, "Learning Assistive Strategies for Exoskeleton Robots from User-Robot Physical Interaction," Pattern Recognition Letters, vol. 99, pp. 67-76, 2017 [Open Access] [URL]
Barkan Ugurlu, Corinne Doppmann, Masashi Hamaya, Paolo Forni, Tatsuya Teramae, Tomoyuki Noda, and Jun Morimoto, "Variable Ankle Stiffness Improves Balance Control: Experiments on a Bipedal Exoskeleton," Transactions on Mechatronics, IEEE/ASME, vol. 21, no. 1, pp.79-87, 2016 [URL]
International Conference
Yueh-hua Wu, Xiaolong Wang, Masashi Hamaya, "Elastic Decision Transformer,” Conference on Neural Information Processing Systems (NeurIPS 2023), accepted, 2023, .
Joaquín Royo-Miquel, Masashi Hamaya, Cristian Beltran Hernandez, Kazutoshi Tanaka, “Learning Robotic Assembly by Leveraging Physical Softness and Tactile Sensing,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2023),2023
Yuki Kadokawa, Masashi Hamaya, Kazutoshi Tanaka, “Learning Robotic Powder Weighing from Simulation for Laboratory Automation,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2023), 2023
Yusaku Nakajima, Masashi Hamaya, Kazutoshi Tanaka, Takafumi Hawai, Felix von Drigalski, Yasuo Takeichi, Yoshitaka Ushiku, Kanta Ono, “Robotic Powder Grinding with Audio-Visual Feedback for Laboratory Automation in Materials Science,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2023), 2023
Kazutoshi Tanaka, Masashi Hamaya, "Twist Snake: Plastic table-top cable-driven robotic arm with all motors located at the base link," IEEE International Conference on Robotics and Automation (ICRA2023), 2023
Rinto Yagawa, Reina Ishikawa, Masashi Hamaya, Kazutoshi Tanaka, Atsushi Hashimoto, Hideo Saito, “Learning Food Picking without Food: Fracture Anticipation by Breaking Reusable Fragile Objects,” IEEE International Conference on Robotics and Automation (ICRA2023), pp. 917-923, 2023
Yusaku Nakajima, Masashi Hamaya, Yuta Suzuki, Takafumi Hawai, Felix von Drigalski, Kazutoshi Tanaka, Yoshitaka Ushiku, Kanta Ono, “Robotic Powder Grinding with a Soft Jig for Laboratory Automation in Material Science,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2022), pp. 2320-2326, 2022
Takayuki Murooka, Masashi Hamaya, Felix von Drigalski, Kazutoshi Tanaka, Yoshihisa Ijiri, "Iterative Backpropagation Disturbance Observer with Forward Dynamics Model," IEEE International Conference on Automation Science and Engineering (CASE 2021), pp. 373-378, 2021.
Kazutoshi Tanaka, Masashi Hamaya, Devwrat Joshi, Felix von Drigalski, Ryo Yonetani, Takamitsu Matsubara, Yoshihisa Ijiri, "Learning Robotic Contact Juggling," IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2021), pp 958-964, 2021.
Felix von Drigalski, Kennosuke Hayashi, Yifei Huang, Ryo Yonetani, Masashi Hamaya, Kazutoshi Tanaka, Yoshihisa Ijiri, "Precise Multi-Modal In-Hand Pose Estimation using Low-Precision Sensors for Robotic Assembly," IEEE International Conference on Robotics and Automation (ICRA 2021), pp. 4055-4061, 2021.
Felix von Drigalski, Devwrat Joshi, Takayuki Murooka, Kazutoshi Tanaka, Masashi Hamaya, Yoshihisa Ijiri, "An analytical diabolo model for robotic learning and control," IEEE International Conference on Robotics and Automation (ICRA 2021), pp. 4055-4061, 2021.
Kazutoshi Tanaka, Ryo Yonetani, Masashi Hamaya, Robert Lee, Felix von Drigalski, Yoshihisa Ijiri, "Trans-am: Transfer learning by aggregating dynamics models for soft robotic assembly," IEEE International Conference on Robotics and Automation (ICRA 2021), pp. 4627-4633, 2021.
Masashi Hamaya, Kazutoshi Tanaka, Yoshiya Shibata, Felix Wolf Hans Erich von Drigalski, Chisato Nakashima, Yoshihisa Ijiri, "Robotic Learning From Advisory and Adversarial Interactions Using a Soft Wrist," IEEE International Conference on Soft Robotics (RoboSoft 2021), 2021.
Takayuki Murooka, Masashi Hamaya, Felix von Drigalski, Kazutoshi Tanaka, Yoshihisa Ijiri, "EXI-Net: EXplicitly/Implicitly Conditioned Network for Multiple Environment Sim-to-Real Transfer," Conference on Robot Learning (CoRL 2020), 2020. [paper]
Masashi Hamaya, Felix von Drigalski, Takamitsu Matsubara, Kazutoshi Tanaka, Robert Lee, Chisato Nakashima, Yoshiya Shibata, Yoshihisa Ijiri, "Learning Soft Robotic Assembly Strategies from Successful and Failed Demonstration," IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2020), pp 8309-8315, 2020.
Kazutoshi Tanaka, Felix von Drigalski, Masashi Hamaya, Robert Lee, Chisato Nakashima, Yoshiya Shibata, Yoshihisa Ijiri, "A Compact, Cable-driven, Activatable Soft Wrist with Six Degrees of Freedom for Assembly Tasks," IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2020), pp 8752-8757, 2020.
Mohammadamin Barekatain, Ryo Yonetani, and Masashi Hamaya, "MULTIPOLAR: Multi-Source Policy Aggregation for Transfer Reinforcement Learning between Diverse Environmental Dynamics”, International Joint Conference on Artificial Intelligence 2020 (IJCAI 2020), accepted [Project page]
Masashi Hamaya, Robert Lee, Kazutoshi Tanaka, Felix von Drigalski, Chisato Nakashima, Yoshiya Shibata, and Yoshihisa Ijiri, "Learning Robotic Assembly Tasks with Lower Dimensional Systems byLeveraging Physical Softness and Environmental Constraints ," 2020 IEEE International Conference on Robotics and Automation (ICRA2020), pp 7747-7753, 2020 [paper]
Felix von Drigalski, Shohei Taniguchi, Robert Lee, Takamitsu Matsubara, Masashi Hamaya, Kazutoshi Tanaka, and Yoshihisa Ijiri, "Contact-based in-hand pose estimation using Bayesian state estimationand particle filtering ," 2020 IEEE International Conference on Robotics and Automation (ICRA2020), pp 7294-7299, 2020
Masashi Hamaya, Takamitsu Matsubara, Jun-ichiro Furukawa, Yuting Sun, Satoshi Yagi, Tatsuya Teramae, Tomoyuki Noda, and Jun Morimoto, "Exploiting Human and Robot Muscle Synergies for Human-in-the-loop Optimization of EMG-based Assistive Strategies," 2019 IEEE International Conference on Robotics and Automation (ICRA2019), pp 549-555, 2019 [URL]
Masashi Hamaya, Takamitsu Matsubara, Tomoyuki Noda, Tatsuya Teramae, and Jun Morimoto, “User-Robot Collaborative Excitation for PAM Model Identification in Exoskeleton Robots," 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2017), pp. 3063-3068, Vancouver, 2017 [URL]
Masashi Hamaya, Takamitsu Matsubara, Tomoyuki Noda, Tatsuya Teramae, and Jun Morimoto, “Learning Task-Parametrized Assistive Strategies for Exoskeleton Robots by Multi-Task Reinforcement Learning," 2017 IEEE International Conference on Robotics and Automation (ICRA2017), pp. 5907-5912, Singapore, 2017 [URL]
Masashi Hamaya, Takamitsu Matsubara, Tomoyuki Noda, Tatsuya Teramae, and Jun Morimoto, "Learning Assistive Strategies from Few User-Robot Interactions: Model-based Reinforcement Learning Approach," 2016 IEEE International Conference on Robotics and Automation (ICRA2016), pp. 3346-3351. Stockholm, 2016 [URL]
Giuseppe Lisi, Masashi Hamaya, Tomoyuki Noda, and Jun Morimoto, "Dry-Wireless EEG and Asynchronous Adaptive Feature Extraction towards a Plug-and-Play Co-Adaptive Brain Robot Interface," 2016 IEEE International Conference on Robotics and Automation (ICRA2016), pp. 959-966, Stockholm, 2016 [URL]
Corinne Doppmann, Barkan Ugurlu, Masashi Hamaya, Tatsuya Teramae, Tomoyuki Noda, and Jun Morimoto, "Towards Balance Recovery Control for Lower Body Exoskeleton Robots with Variable Stiffness Actuators: Spring-Loaded Flywheel Model," 2015 IEEE International Conference on Robotics and Automation (ICRA2015), pp. 5551 - 5556, Seattle, May 2015 [URL]
解説記事
牛久祥孝, 濱屋政志, 米谷竜,”スマートファクトリー R&D —オムロンサイニックエックスにおける研究事例—,”人工知能,37巻,3号,pp. 299-305, 2022
濵屋 政志,田中 一敏,フェリクス フォン ドリガルスキ, 井尻 善久,”工場現場の組立応用に向けたソフトロボット運動学習,”日本ロボット学会誌,39 巻 ,7 号,pp. 609-612,2021
濱屋政志,松原崇充,森本淳,ガウス過程回帰の基礎から応用 外骨格ロボットに搭載された空圧人工筋の協調同定 -ガウス過程に基づく能動学習のアプローチ,”システム/制御/情報 ,”62巻 ,10 号,pp. 423-428,2021
Invited Talk
濵屋政志,「柔軟要素を持つロボットのモデルベース学習」,ロボット工学セミナー ロボットのための強化学習/深層強化学習,May 2022
濵屋政志,「柔軟要素を持つロボットによる組立作業学習」,日刊工業新聞社 オンラインセミナー AI利用によるロボットマニピュレーションと産業応用,Mar 2022
濵屋政志,「ソフトロボットによる部品組立作業学習」,Vision Engineering Workshop 2021 基調講演,Dec 2021
濵屋政志,「柔軟要素を持つロボットによる部品組立作業に向けたモデルベース強化学習 」,人工知能学会 人工知能セミナー 強化学習とその実用化,Nov 2021
濵屋政志,「柔軟要素を持つロボットによる組立作業学習 」,日本ロボット学会 ロボット工学セミナー 組立動作の自動制御技術,Sep 2021
濵屋政志,「ソフトロボットを活用した工場現場における組み立て動作学習」,日本ロボット学会ヒューロビント研究専門委員会 若手ロボティクス研究会,Jan 2021
Yoshihisa Ijiri and Masashi Hamaya, "Soft-Robotic Learning for Industrial Assembly," Knowledge Based Reinforcement Learning Workshop, International Joint Conference on Artificial Intelligence (IJCAI 2020), Jan 2021 [URL]
Award
IEEE Kansai Section Student Paper Award, Feb 2018
ベストプレゼンテーション賞 大阪大学大学院 生命機能研究科 博士論文公聴会, Mar 2019
IEEE Robotics and Automation Society Japan Joint Chapter Young Award, May 2019
Research projects
Design of physical user–robot interactions for model identification of soft actuators on exoskeleton robots , IJRR 2019
Recent breakthroughs in wearable robots, such as exoskeleton robots with soft actuators and soft exosuits, have enabled the use of safe and comfortable movement assistance. However, modeling and identification methods for soft actuators used in wearable robots have yet to be sufficiently explored. In this study, we propose a novel approach for obtaining accurate soft actuator models through the design of physical user–robot interactions for wearable robots, in which the user applies external forces to the robot. To obtain an accurate soft actuator model from the limited amount of data acquired through an interaction, we leverage an active learning framework based on Gaussian process regression. We conducted experiments using a two-degree-of-freedom upper-limb exoskeleton robot with four pneumatic artificial muscles (PAMs). Experimental results showed that physical interactions between the exoskeleton robot and the user were successfully designed to allow PAM models to be identified. Furthermore, we found that data acquired through an interaction could result in more accurate soft actuator models for the exoskeleton robots than data acquired without a physical interaction between the exoskeleton robot and the user.
Exploiting Human and Robot Muscle Synergies for Human-in-the-loop Optimization of EMG-based Assistive Strategies, ICRA 2019, IEEE Robotics and Automation Society Japan Joint Chapter Young Award
In this study, we propose a novel human-in-the-loop optimization approach for exoskeleton robot control. We develop a method to optimize widely-used Electromyography (EMG)-based assistive strategies. If we use multiple EMG channels to control multi-DoF robots, optimization process becomes complex and requires a large amount of data. To make the optimization tractable, we exploit the synergies both of the human muscles and artificial muscles of the exoskeleton robots to reduce the number of parameters of the assistive strategies. We show that we can extract the synergies not only from the user’s muscle activities but from pneumatic artificial muscle (PAMs) contractions of the exoskeleton robot.
Then, we adopt a Bayesian optimization method to acquire the parameters for assisting human movements by iteratively identifying the user’s preferences of the assistive strategies. We conducted experiments to evaluate our proposed method with a PAMs-driven upper-limb exoskeleton robot. Our method successfully learned assistive strategies from the human-in-the-loop optimization with a practicable number of interactions.
Learning assistive strategies for exoskeleton robots from user-robot physical interaction, Pattern Recognit. Lett. 2017
With exoskeleton robots, an assistive strategy is a key ingredient. Since interactions between users and exoskeleton robots are bidirectional, the assistive strategy design problem is complex and challenging. We explore a data-driven learning approach for designing assistive strategies for exoskeletons from user-robot physical interaction. We formulate the learning problem of assistive strategies as a policy search problem and exploit a data-efficient model-based reinforcement learning framework. Instead of explicitly providing the desired trajectories in the cost function, our cost function only considers the user’s muscular effort measured by electromyography signals (EMGs) to learn the assistive strategies. We applied our method to a 1-DoF exoskeleton robot and conducted a series of experiments with human subjects. Our experimental results demonstrated that our method learned proper assistive strategies that explicitly considered the bidirectional interactions between a user and a robot with only 60 seconds of interaction.
Pneumatic Artificial Muscle (PAM) identification for exoskeleton robots by user-root collaboration, IROS 2017
Pneumatic Artificial Muscle (PAM) actuators have been used as exoskeletons because of their inherited compliance and high power-weight ratio. However, creating accurate models remains difficult mainly due to the compliance issue; the model can be changed by the force applied by the user. Therefore, both user and robot actions need to be considered for sufficient excitation of PAMs that are equipped in exoskeleton robots, unlike typical rigid actuators that can only be sufficiently excited by robot actions. We propose a user-robot collaborative excitation approach for PAM model identification as an active learning framework for sequentially collecting data by deriving and executing optimal user and robot actions at each step with Gaussian processes . Experimental results show that our method can more efficiently identify the PAM model than a standard model identification method that does not use any data acquired through user-robot collaboration.