Keio University

Research Center for Artificial Intelligence and Big Data

Publish: June 30, 2025
KGRI

Center Director: Takahira Yamaguchi (Professor, Faculty of Science and Technology)

Main Campus: Yagami

Center Overview

In the CREST project, "Development and Social Implementation of PRINTEPS, a Framework for Building Practical Intelligent Applications," we are developing the PRINTEPS platform. This platform builds diverse general intelligences that can co-evolve with humans by integrating five types of elemental intelligence: knowledge reasoning, voice dialogue, image sensing, motion, and machine learning (including deep learning).

In terms of practical application, we will implement and evaluate a robot café and a teaching assistant robot as applications of multi-task, multi-robot systems.

Keywords and Main Research Themes

Artificial intelligence, image sensing, intelligent robots, big data

Fiscal Year 2019 Business Plan

■ Activities Continuing from the Previous Fiscal Year: Background, Rationale, and Goals

Currently, PRINTEPS implements six types of intelligent elemental technologies: knowledge reasoning, voice dialogue, continuous dialogue, image sensing of people and objects, intelligent limb movements, and machine learning (which is relevant to all). However, as these elemental intelligences are only simply linked, we aim to create more appealing services by enabling bidirectional and multiple interactions among them.

■ New Activity Goals, Content, and Implementation Background for Fiscal Year 2019

We plan to continue our joint research with Toyota Motor Corporation related to the HSR (Human Support Robot) this fiscal year.

Fiscal Year 2019 Business Report

■ Implementation Details, Research Results, and Degree of Achievement for the Fiscal Year Business Plan

As a practical application of PRINTEPS: (1) At Suginami Ward Hamadayama Elementary School, we had the AI robot SOTA, which uses ontologies, participate in student group discussions. We confirmed that SOTA could answer student questions, contribute to the discussion's development, and vitalize the group discussions. (2) We also prototyped an udon chef AI robot. We confirmed that it could evaluate customer requests based on nutritional knowledge, recommend and prepare better ingredient combinations for the customer, allowing us to envision a future society.

Achievements in Social Contribution, Including Published Papers, Conference Presentations, and Events

  1. Takeshi Morita, Naoya Takahashi, Mizuki Kosuda, Takahira Yamaguchi, "A Teaching Assistant Robot Design Tool Base on Knowledge Chunks Reuse", The 2nd IEEE International Workshop on Advances in AI and Machine Learning: Research & Practice, pp.68-73, 2019.

  2. Michael Walch, Takeshi Morita, Dimitris Karagiannis, Takahira Yamaguchi, "A knowledge-based conceptual modelling approach to bridge design thinking and intelligent environments", Proceedings of the 12th International Conference on Knowledge Science, Engineering and Management (KSEM 2019), LNCS, Vol. 11775, Springer, pp.524-536, 2019.

  3. Takeshi Morita, Naho Kashiwagi, Ayanori Yorozu, Hideo Suzuki, Takahira Yamaguchi, "Evaluation of a Multi-Robot Cafe based on Service Quality Dimensions", The Review of Socionetwork Strategies, Springer, Vol. 13, Issues. 24, pp.1-22, 2019.

  4. Hiroki Ono, Kaito Koike, Takeshi Morita, Takahira Yamaguchi, "Ontologies-Based Pupil Robot Interaction with Group Discussion", Proceedings of the 23rd International Conference on Knowledge-Based and Intelligent Information & Engineering Systems, 2019.

  5. Takeshi Morita, Naoya Takahashi, Mizuki Kosuda, Takahira Yamaguchi, "A Knowledge Chunk Reuse Support Tool Based on Heterogeneous Ontologies", Proceedings of the 11th International Conference on Knowledge Engineering and Ontology Development, 2019.

  6. Ayanori Yorozu, Genya Ishigami, Masaki Takahashi, Ridge-tracking for Strawberry Harvesting Support Robot According to Farmer's Behavior, The 12th Conference on Field and Service Robotics (FSR2019)

Notable Achievements Through Center Activities

The research results for PRINTEPS, a platform that allows end-users to build diverse general intelligence applications, were featured in the October 2019 issue of the UK web magazineOpen Access Government.

User-Centric Platform PRINTEPS for Unified AI Robot Applications

Project Members

Principal Investigator

Takahira Yamaguchi

ProfessorDepartment of Industrial and Systems Engineering, Faculty of Science and Technology

Hideo Saito

ProfessorDepartment of Information and Computer Science, Faculty of Science and Technology

Yuko Koshino

Project Research AssociateGraduate School of Science and Technology