○International Award

Mr. Ryoichi TAKASE
He received the B.E. degree from Osaka Prefecture University, Japan, in 2017. He received the M.E. and Ph.D. degrees in Engineering from the University of Tokyo, Japan, in 2019 and 2022, respectively. He is currently with Hitachi, Ltd. His research interests include theory and application of reinforcement learning and robust control for safety-critical systems.

Mr. Nobuyuki YOSHIKAWA
He graduated from Keio University in 2014 and received degrees of a Bachelor of technology and a Master of medicine, and he joined Mitsubishi Electric Corporation. His responsibility in Mitsubishi Electric is research of machine learning and quantum computing.

Prof. Takeshi TSUCHIYA
He received Ph.D. degree in Engineering from the University of Tokyo in 2000. After the graduation, he entered the National Aeronautical Laboratory of Japan (presently, JAXA) and became a lecturer in 2002, an Associate Professor in 2006, and a Professor in 2015. He has been working in the fields of flight control system, numerical optimization of aerospace design, and unmanned aerial robotics.

受賞論文「Estimating Bounded Uncertain Model for Stability-Certified Reinforcement Learning」
This article presents the methods of identifying uncertain plants used for stability-certified reinforcement learning (RL). The uncertain plant is the interconnection of a perturbation and a nominal plant represented by a state-space model. By assuming that the perturbation is bounded, two types of methods are proposed to identify the uncertain plant. The first one is an optimization-based approach proposed for known nonlinear systems. The state-space model is given by solving an optimization problem of seeking matrices satisfying the bounded condition. The second method is a learning-based
approach proposed for unknown systems. The state-space model is estimated by interacting with environments, which means that the second method is suitable for conjunction with stability-certified RL. In the numerical experiments, the identified uncertain models are used for stability analysis of feedback systems with a neural network controller. The results show that both feedback systems provide enough stability with similar regions of attractions (ROAs). This means that the learning-based method enables us to discuss the stability of neural network controllers even for unknown systems.

Mr. Takeshi EMOTO
He received his B.E. and M.E. degrees in mechanical engineering from Hokkaido University, Japan in 1997 and 1999, respectively. He is currently a doctoral student at Graduate School of Engineering, Hokkaido University, Japan. His research primarily focuses on the development of automatic inspection system for railcar. He is a student member of SICE and JSME.

Mr. Abhijeet RAVANKAR
After completing his M.S. in Computer Science & Engineering, he received Ph.D. degree from Hokkaido University in 2017. He was a MEXT scholarship recipient for three times from the government of Japan. He worked with Panasonic Corporation, Japan, from 2011 to 2014 as a computer vision engineer. After briefly working a Post-Doc researcher at Hokkaido University, he joined Kitami Institute of Technology as an Assistant Professor in 2018.Since 2021, he is working as an Associate Professor and the head of Robotics & AI Laboratory. His research interests include robot navigation, multi-robot systems, computer vision, and artificial intelligence. He is a member of IEEE and JSME.

He (Member, IEEE, SICE, JSME) received a Master’s degree (2012) and a PhD (2015) in Human Mechanical Systems and Design Engineering from Hokkaido University, Japan. He was the MEXT scholarship recipient (2010?2015) from the government of Japan. He is presently working as a specially appointed Associate Professor at the Department of Robotics, Tohoku University, Japan. He was an Assistant Professor at the Graduate School of Engineering, Hokkaido University (2015?2021) and a Lecturer at Tohoku University, Japan (2021?2022). His research interests include mobile robot navigation, simultaneous localization and mapping (SLAM), system integration, multi-robot systems, service robots, motion and path planning, artificial intelligence and machine learning, semantic scene understanding, and decision-making for robot sensing under uncertainty.

Dr. Takanori EMARU
He received his M.E. and Ph.D. degrees in electrical engineering from Hokkaido University, Japan, in 1998 and 2002, respectively. He was a Research Fellow of the Japan Society for the Promotion of Science at the University of Electro-Communications, Japan, from 2004 to 2006. He was a Lecturer at Osaka Electro-Communication University from 2006 to 2007. Currently, he is an Associate Professor at Hokkaido University, Japan. His research interests include the areas of robotics, navigation, sensor, and nonlinear signal processing. He is a member of IEEE, RSJ, SICE, and JSME.

Prof. Yukinori KOBAYASHI
He received his B.E., M.E., and Ph.D. degrees in Mechanical Engineering from Hokkaido University, Japan, in 1981, 1983 and 1986, respectively. He is currently the President at the National Institute of Technology, Tomakomai College, Japan, and an Emeritus Professor of Hokkaido University, Japan. His research interests include vibration control of flexible structure, control problem of robots having flexibility, path planning, and navigation of mobile robots, vibration analysis, and nonlinear vibrations of continuous systems. He is a member of JSME, SICE, RSJ, and EAJ.

受賞論文「Automatic Dimensional Inspection System of Railcar Wheelset for Condition Monitoring」
The wheelset is designed to maintain the stable and safe running of railcars and requires periodic maintenance. Failure of a wheelset can lead to major incidents (derailment in the worst case scenario). Therefore, wheelset inspection is categorized as one of the most important tasks in railcar maintenance. However, with the shortage of skilled labor and aging society, the work force becomes more scarce, and the replacing manual inspection with an automatic inspection system is becoming more crucial. Railcar maintenance continues to rely mainly on traditional manual and visual inspection, which requires a considerable workforce. Therefore, a reliable automatic inspection system is strongly desired to reduce the cost of inspection and workforce while ensuring a stable accuracy of measurements.
This research proposes an automatic inspection system for condition monitoring of wheelsets. Traditional manual inspection of wheelsets uses special equipment. Highly skilled technicians must also perform manual measurements, and stable precision is difficult to achieve. To overcome these problems, we proposed a method and equipment for the automatic inspection of wheelsets under various constraints.
In this research, we propose concrete installation conditions and equipment types under the following constraints.
・Effective measurable range of laser sensors
・Obstacles installed near the wheelset
・Structure gauge of the railcars
・Railcar pass speed
This paper proposes the automatic inspection system concept and describe its setup. The novelties of our approach are twofold: first, we propose equipment specifications within the abovementioned restrictions; second, we confirm the feasibility of automatic inspection of the wheelset.
In this research, we confirmed the automatic inspection of WBTB and WTP under static conditions using an actual wheelset. We demonstarated that laser sensors can substitute traditional manual inspection. Our goal is to develop an automatic inspection system for condition monitoring of wheelsets.

○Young Author's Award

She received the B.E. degree in engineering from Keio University in 2022. She is currently in the Master’s course at the Graduate School of Science and Technology, Keio University. Her research interests include feedback control of biocircuits.

受賞論文「Model-Based Analysis and Rapid Prototyping of Genetic Circuits with Decoy Transcription Factor Binding Sites」
Decoy transcription factor binding sites in biocircuits have recently received attentions as a versatile tool for tuning the parameters of biocircuits in synthetic biology. In this paper, we propose a combined modeling and experimental approach to accelerate the tuning of transcription rates by decoy transcription factor binding sites. Specifically, we derive an analytic expression between the concentrations of the decoy transcription factor binding site and the transcription factor-bound DNA that synthesizes target proteins. We then measure the effect of decoy transcription factor binding sites on transcription using a cell-free protein synthesis system. Finally, we compare model-based predictions and experimental results to verify the model.