Abstract: This project integrates microcontroller technology with stepper motors to achieve precise syringe control, establishing an artificial intelligence-based titration analysis system utilizing computer vision. The system implements an innovative visual approach employing the ResNet neural network classification algorithm to intelligently identify titration endpoints through indicator color changes. During the 23rd-level analytical chemistry laboratory course, 40 students successfully completed the experiment within a 4-hour session, demonstrating the system’s user-friendly operation and ease of learning. In the inaugural Intelligent Experiment Challenge competition, 19 teams comprising 57 participants not only completed the assigned tasks but also showcased innovative potential, with 4 teams independently developing novel hardware systems and software algorithms. To enhance cost-effectiveness and scalability, the project utilizes commercial components and 3D printing technology. This technological integration not only pioneers new directions for teaching innovation in fundamental analytical chemistry experiments but also establishes a robust foundation for future research in intelligent experimental systems.

Key words: Convolutional neural network (CNN), Computer vision (CV), Intelligent titration