ChatGPT-powered Inquiry-based Learning Model of Training for Intelligent Car Racing Competition

Abstract

In this study, we explore the application of an inquiry-based learning model powered by ChatGPT in the context of intelligent car racing competition training. We address four key aspects: (1) the construction of a knowledge and skill acquisition process through student interactions with ChatGPT to facilitate the progressive development of problem-solving strategies and approaches; (2) project-based learning for interdisciplinary students participating in the competition, where students are grouped in accordance with their backgrounds and engage in tasks such as vehicle design and optimization, electrical drive and control algorithm adaptation, and sensor circuit design and calibration; (3) the paradigm shift in the role of teachers, transitioning from knowledge providers to co-coaches alongside ChatGPT, allowing teachers to allocate more time to monitor the progress of different student groups and design learning objectives; and (4) knowledge building and prompt engineering during different stages of the training process, where students employ various questions and prompts to interact with ChatGPT, thereby constructing domain-specific knowledge and improving the quality and effectiveness of knowledge acquisition. By leveraging ChatGPT as a conversational agent, students engage in a dynamic learning process that fosters their understanding of research problems and nurtures their problem-solving skills. Integrating an inquiry-based approach, project-based learning, and teacher-student collaboration with ChatGPT empowers students to acquire essential knowledge and cultivate critical thinking abilities, contributing to their overall growth and readiness for intelligent car racing competitions. The findings of this study shed light on the efficacy of ChatGPT-powered inquiry-based learning models in preparing students for complex and interdisciplinary challenges in the field of intelligent car racing.