Scratch for Sports: Athletic Drills as a Platform for Experiencing, Understanding, and Developing AI-Driven Apps
DOI:
https://doi.org/10.1609/aaai.v37i13.26901Keywords:
Sports, Physical Computing, Gesture Detection, K-12 Education, Object Detection, Scratch, Soccer, BasketballAbstract
Culturally relevant and sustaining implementations of computing education are increasingly leveraging young learners' passion for sports as a platform for building interest in different STEM (Science, Technology, Engineering, and Math) concepts. Numerous disciplines spanning physics, engineering, data science, and especially AI based computing are not only authentically used in professional sports in today's world, but can also be productively introduced to introduce young learnres to these disciplines and facilitate deep engagement with the same in the context of sports. In this work, we present a curriculum that includes a constellation of proprietary apps and tools we show student athletes learning sports like basketball and soccer that use AI methods like pose detection and IMU-based gesture detection to track activity and provide feedback. We also share Scratch extensions which enable rich access to sports related pose, object, and gesture detection algorithms that youth can then tinker around with and develop their own sports drill applications. We present early findings from pilot implementations of portions of these tools and curricula, which also fostered discussion relating to the failings, risks, and social harms associated with many of these different AI methods – noticeable in professional sports contexts, and relevant to youths' lives as active users of AI technologies as well as potential future creators of the same.Downloads
Published
2024-07-15
How to Cite
Kumar, V., & Worsley, M. (2024). Scratch for Sports: Athletic Drills as a Platform for Experiencing, Understanding, and Developing AI-Driven Apps. Proceedings of the AAAI Conference on Artificial Intelligence, 37(13), 16011-16016. https://doi.org/10.1609/aaai.v37i13.26901
Issue
Section
EAAI Symposium: Resources for Teaching AI in K-12