As the core component of art education, the teaching effect of art appreciation is often limited by the problems of teachers' experience, uneven distribution of resources and insufficient teaching interaction in the traditional mode. The purpose of this study is to develop an algorithm-driven robot-aided system. By constructing a semantic analysis model of art works based on deep learning, the style classification, emotional labeling and cultural background of art works are realized, and an educational robot with emotional interaction ability is designed to support multimodal interaction and dynamic demonstration such as voice, gesture and expression, so as to enhance learners' aesthetic ability, learning interest and creativity. The system adopts a modular and hierarchical architecture, including an art works analysis module, a multimodal interaction module, a teaching strategy generation module and a robot execution module. The key technologies include multi-modal analytic network of art works, emotional calculation model, dynamic teaching strategy generation and robot action planning. The experimental results show that the system is excellent in the accuracy of art works analysis, the response speed of multi-modal interaction and user acceptance, and the improvement of teaching effect. The students in the experimental group are significantly better than those in the control group in terms of aesthetic ability, classroom participation and creativity.
Shi et al. (2026) studied this question.