1. Praxis: a framework for AI-driven human action recognition in assembly.
- Author
-
Gkournelos, Christos, Konstantinou, Christos, Angelakis, Panagiotis, Tzavara, Eleni, and Makris, Sotiris
- Subjects
HUMAN activity recognition ,HIGH performance computing ,ARTIFICIAL intelligence ,PRAXIS (Process) ,SOFTWARE frameworks ,DIGITAL technology - Abstract
The role of Artificial intelligence in achieving high performance in manufacturing systems has been explored over the years. However, with the increasing number of variants in the factories and the advances in digital technologies new opportunities arise for supporting operators in the factory. The hybrid production systems stipulate the efficient collaboration of the workers with the machines. Human action recognition is a major enabler for intuitive machines and robots to achieve more efficient interaction with workers. This paper discusses a software framework called Praxis, aiming to facilitate the deployment of human action recognition (HAR) in assembly. Praxis is designed to provide a flexible and scalable architecture for implementing human action recognition in assembly lines. The framework has been implemented in a real-world case study originating for showcasing and validating the effectiveness of Praxis in real-life applications. It is deployed in an assembly use case for an air compression production industry. This study highlights the potential of the Praxis framework for promoting efficient human–robot collaboration (HRC) in modern manufacturing environments through HAR. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF