Kato Laboratory focuses on research on the integration of information networks and robotics, the fundamental concept for which is "Network + Sensing + Robotics."
Recent years have been witnessed an increase of interest in cyber-physical systems (CPSs) and Internet of Things (IoT). These systems acquire real-world contexts captured using various sensors in a virtual space (as mathematical models constructed in computers), analyze the sensor data using the model, and then based on the analysis results, actuate the real-world system by running actuators (such as motors). To realize such a CPS society, we design advanced network systems by combining information network technologies (including IoT) and robotics into one that can facilitate the collaboration between humans and their environment.
Mathematical models (statistical and probabilistic methods)，computer simulation
Machine learning, data analysis and prediction
Real-world sensing and its analysis (mainly, mathematical models for robotics)
Robot service platform
IoT service platform etc.
1. Real-world sensing and its analysis
We are conducting research on various sensing technologies, analysis methods for the sensor data and their applications in robot services. Examples include a gait estimation method using a walking pattern obtained from a laser range scanner (using machine learning), a method for predicting the destination of pedestrians from their trajectories (using machine learning), time series data analysis of vital sensor data (using Fourier transform analysis), and pedestrian models for robotics (using Bayesian inference, deep learning etc.). We are also developing a delay-conscious communication model for mobile robot navigation based on research on Integration of information networks and robotics. Recently, we have also been working on semiautomatic preprocessing methods for data analysis.
H. Nagashima and Y. Kato: APREP-DM: a Framework for Automatic the Pre-Processing of a Sensor Data Analysis based on CRISP-DM, IEEE PerFoT 2019, pp.555-560, 2019.
Y. Kato, T. Ikeda, S. Okano and N. Matsuhira, A Method Predicting the Destination of a Pedestrian for Voice Guidance of Communication Robots, IPSJ Journal, Vol.60, No.2, pp.572-580, 2019. （in Japanese）
N. Sakata, Y. Kinoshita and Y. Kato, Predicting a Pedestrian Trajectory Using Seq2Seq for Mobile Robot Navigation, IEEE IECON 2018, pp.4300-4305, 2018.
Y.Kato,Y.Nagano and H.Yokoyama, A Pedstrian Model in Human-Robot Coexisting Environment for Mobile Robot Navigation, IEEE/SICE SII 2017, pp.992-997, 2017.
S. Akiyama and Y. Kato, A Method for Estimating Stress and Relaxed States Using a Pulse Sensor for QOL Visualization, Smart Sensors Networks (Chap.12), F. Xhafa et al. (Eds.), Academic Press, pp.261-290, 2017.
S. Sakai, S. Kimura, T. Ikeda, D. Nomiyama, N. Matsuhira and Y. Kato, A Classification Method of Elderly and Young People Using Walking Pattern Obtained from a Laser Range Scanner, IPSJ Journal, Vol.58, No.2, pp.375-383, 2017. （in Japanese）
Y. Kato and M. Tanaka, Performance Evaluation of Remote Navigation with Network Delay for Low-cost Mobile Robots, IEEE MFI 2016, pp.615-620, 2016.
2. Robot service platform
We are conducting a study of protocol specification for robot services using the Internet and robot service integration platforms. In addition to proposing platform technologies, we are also promoting research focused on practical applications, including a robot remote-navigation system with virtual reality and a campus tour service using remote-control robots. These studies are being conducted as part of the activities of RSi (Robot Service initiative) which is a consortium of robotics companies， SIG on Robot Services using Computer Networks of the Robotics Society of Japan, Challenge for Realizing Early Profits of the Japan Society for Artificial Intelligence, etc.
M. Narita, Y. Tsuchiya, Y. Kato and Y. Murakawa, Summary of Challenge for Realizing Early Profits: Cloud-based Robot Service Platform, Journal of JSAI, Vol.33, No.6, p.814, 2018. （in Japanese）
Y. Kato, A Remote Navigation System for a Simple Tele-presence Robot with Virtual Reality, IEEE IROS 2015, pp.4524-4529, 2015.
M. Narita, T. Izui, S. Nakagawa, Y. Tsuchiya, N. Matsuhira and Y. Kato, Development Framework for Non-experts Targetting the Network Robot Service, Journal of the Robotics Society of Japan, Vol.33, No.10, pp.807-817, 2015. （in Japanese）
T. Kusu, M. Takahashi, Y. Nomoto, Y. Ito, Y. Tsuchiya, M. Narita and Y. Kato, An RPG-like Campus Tour Service using Remote Control Robots, IEEE IECON 2013, Vienna, Austria, pp.8251-8256, 2013.
Y. Kato, T. Izui, Y. Tsuchiya, M. Narita, M. Ueki, Y. Murakawa and K. Okabayashi, RSi-Cloud for Integrating Robot Services with Internet Services, IEEE IECON 2011, pp.2164-2169, 2011.
M. Narita, Y. Murakawa, C. Akiguchi, Y. Kato and T. Yamaguchi, Push Communication for Network Robot Services and RSi/RTM Interoperability, FUZZ-IEEE 2009, pp.1480-1485, 2009.
3. IoT service platform
We are conducting a study on an information-processing platform focused on information flow that combines and utilizes the data stream acquired and generated by IoT devices, including mobile and sensing devices, in real time. We are also constructing novel fundamental technologies by combining next-generation network technologies with information flow (i.e., flow of information contents). Currently, we hava an ongoing research on a secure IoT service infrastructure for strengthening trust in an information society.
Fumi Ito, Eri Ozawa and Yuka Kato: Design of Robot Service Functions for a Framework Establishing Human-Machine Trust, AINA 2019, pp.1193-1204, 2019.
T. Suganuma, K. Yasumoto and Y. Kato, A Concept of a Framework Establishing Human-Machine Trust for Secure IoT Services, IPSJ SIG Technical Report, Vol. 2018-DPS-175, pp.1-7, 2018. (in Japanese)