Democratising scientific software in physics education: AI-assisted custom instrumentation for teachers

Full abstract

Read the full abstract

Abstract Published Arduino-based mechanics experiments reveal a continued accessibility paradox: while sensor hardware costs are minimal, the software required for instrumentation control and data analysis can be financially restrictive for school practitioners, limiting access to enhanced analytical techniques. This paper contributes to addressing this gap by showing how AI-assisted Python development enables KS4-5 secondary school physics teachers, without formal programming training, to extend Arduino-based experiments using entirely free software. A worked example is presented in which a complete graphical user interface for spring oscillation measurements is developed with large language models. The development process, challenges encountered, and implications for classroom practice are discussed. The resulting system provides real-time data acquisition, automated analysis, and data visualisation, including curve fitting to anharmonic oscillator models and numerical derivative calculations. Static and dynamic spring constant measurements show good agreement (19.94 ± 0.13 N m −1 and 20.44 ± 0.14 N m −1 respectively), validating the approach. The method generalises to a wide range of Arduino-based experiments and demonstrates that sophisticated data acquisition and analysis are now accessible to individual educators at low cost. A short demonstration video of the apparatus and workflow is available at https://zenodo.org/records/17980784.