GENERATIVE AI-ASSISTED TEACHING FOR MICROCONTROLLER PROGRAMMING AND DEBUGGING: MODEL DESIGN AND PRACTICE
TL;DR
An AI-assisted teaching model for microcontroller programming practice based on three functional modules, namely code-skeleton generation, compilation/logic-error diagnosis, and personalized question answering is built and preliminary teaching practice indicates that the model helps shorten students' debugging time and improve learning autonomy and classroom satisfaction.
Abstract
The course "Principles and Applications of Microcontrollers" is highly practice-oriented and covers a broad range of knowledge domains. In program writing and debugging, students commonly face a steep learning curve, time-consuming debugging, and limited access to personalized instructor guidance. Taking generative artificial intelligence (Generative AI) as an entry point, this paper builds an AI-assisted teaching model for microcontroller programming practice based on three functional modules, namely code-skeleton generation, compilation/logic-error diagnosis, and personalized question answering, and conducts concrete teaching-practice exploration in terms of instructional design, laboratory cases, and implementation procedures. Preliminary teaching practice indicates that the model helps shorten students' debugging time and improve learning autonomy and classroom satisfaction. At the same time, it also carries risks such as the reliability of AI-generated code and students' over-reliance on AI, which require instructors to guide and regulate through instructional organization and assessment design. The teaching model and implementation pathway proposed in this paper can serve as a reference for teaching reform in similar practice-oriented courses.