2020 Fiscal Year Research-status Report
Development of artificial agents for the classroom and study of their efficiency
| Project/Area Number |
19K12167
|
|---|
| Research Institution | Tokyo University of Agriculture and Technology |
|---|
Principal Investigator |
Venture Gentiane 東京農工大学, 工学(系)研究科(研究院), 卓越教授 (30538278)
|
|---|
| Project Period (FY) |
2019-04-01 – 2022-03-31
|
| Keywords | Machine learning / Human behavior analysis / online teaching |
|---|
| Outline of Annual Research Achievements |
Three main tasks were achieved:
1. All the collected data last year are then processed to be analyzed in machine learning algorithms. This resulted in an offline behavior analysis software of students’ behaviors, enabling us to choose the relevant features for the next step.
2. The relevant features found in 1. were used to develop a real-time algorithm to identify the changes in students’ behavior. Again, this is based on an implementation of machine learning algorithm in real-time for classification. We tested several machine learning architecture to find the most relevant one.
3. A mimicking algorithm that allows the robot to reproduce the lecturer behaviors offline to generate easily robot's movement.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
The research was little impacted by the pandemic as this year it was mainly a task of programming and data analysis.
|
| Strategy for Future Research Activity |
We will now work further on the robotics development for a robot lecturer focusing on important behaviors of the lecturer. The behavior
of the robot will not be pre-programmed but will adaptively change depending on the students’ behavior measurement using the system developed in C. The whole system will then be tested with different teaching styles and different audiences in Japan. Evaluations will be conducted, and statistical analysis of the results between human lecturer and robot lecturer will be compared to assess the validity of the teaching robot.
|
