TeachingBot: Robot Teacher for Human Handwriting

Zhimin Hou,Cunjun Yu,David Hsu,Haoyong Yu

2023-09-21

Abstract:Teaching physical skills to humans requires one-on-one interaction between the teacher and the learner. With a shortage of human teachers, such a teaching mode faces the challenge of scaling up. Robots, with their replicable nature and physical capabilities, offer a solution. In this work, we present TeachingBot, a robotic system designed for teaching handwriting to human learners. We tackle two primary challenges in this teaching task: the adaptation to each learner's unique style and the creation of an engaging learning experience. TeachingBot captures the learner's style using a probabilistic learning approach based on the learner's handwriting. Then, based on the learned style, it provides physical guidance to human learners with variable impedance to make the learning experience engaging. Results from human-subject experiments based on 15 human subjects support the effectiveness of TeachingBot, demonstrating improved human learning outcomes compared to baseline methods. Additionally, we illustrate how TeachingBot customizes its teaching approach for individual learners, leading to enhanced overall engagement and effectiveness.

Robotics

What problem does this paper attempt to address?

The paper aims to address the following issues:

1. **Personalized Teaching Needs**: Human learners exhibit significant differences in writing styles, leading to varying learning preferences. Therefore, a robotic teaching system needs to adapt to each learner's unique style.

2. **Balance of Guidance Intensity**: It is challenging to determine the appropriate level of physical guidance. Insufficient guidance fails to provide adequate support to the learner, while excessive guidance may increase the learner's dependency on the teacher, hindering their ability to independently master the skills.

To address these issues, the research team proposed the TeachingBot system, a robotic teacher specifically designed to teach human writing skills. TeachingBot achieves its goals through the following steps:

1. **Collecting Learner Writing Samples**: First, it collects the writing trajectory data of human learners and uses a Gaussian Mixture Model (GMM) to encode the learners' writing styles.

2. **Generating Personalized Teaching Trajectories**: Based on the learned learner styles and reference patterns, it uses probabilistic learning methods to generate personalized teaching trajectories.

3. **Variable Impedance Control**: Based on the generated teaching trajectories, TeachingBot interacts with human learners through a Variable Impedance Controller to provide an appropriate level of physical guidance. Specifically, it increases impedance when the learner deviates from the reference trajectory and decreases impedance when the deviation is small to encourage learner engagement.

Experimental results show that compared to traditional methods, TeachingBot can significantly improve learners' writing skills, better adapt to individual differences, and enhance the overall learning experience and effectiveness.