Simplifying Edge AI: A Block-Based Development Platform for IoT Applications
Alexander Koß, Guido Dartmann, Klaus-Uwe Gollmer, Lejla Begic Fazlic · 2025
This paper aims to develop a visual programming environment that allows users with limited programming knowledge to interact with AI models. This environment is implemented on an Edge-AI platform. The platform, referred to as the OZY board, is designed to simplify and enhance the accessibility of AI applications in practice. The developed visual programming environment focuses on creating an intuitive user interface, making it easier for nonprogrammers to work with AI models through block-based programming, reducing the complexity of machine learning development. The study includes a thorough literature review, evaluation of suitable libraries, and the creation of use cases. UI prototypes are developed, followed by technical implementation and user testing. The results of these tests are used to further refine the application. This research contributes to the field by addressing the challenges of accessibility and usability in AI technologies and provides practical solutions for extending AI access to a wider user base, especially in education to demonstrate simple machine learning algorithms.