Diseño e implementación de un proceso industrial accionado mediante gestos usando PLC, HMI y un ordenador Monoplaca

Abstract

The project aims to focus on the design and implementation of an industrial process that uses gesture recognition to actuate devices by integrating technologies such as PLC, HMI and a single board computer. The paper establishes the relevance of artificial intelligence and vision in optimizing industrial processes, in a global context where leading countries such as the United States, which contains strong investments in industrial automation, and Germany, which is a pioneer in the integration of cyber-physical systems, have invested in the infrastructure and development of these technologies. Furthermore, the urgent need for innovative approaches to interact with industrial equipment is emphasized by our solution, which serves as an alternative method of direct control for users with disabilities or physical limitations. The project has one goal, which is to develop a system that translates gestures into commands and uses images to build a dataset to train machine learning models to identify these gestures and establish a communication interface between the single-board computer and the PLC. By using dedicated access and fast hardware, such as Raspberry Pi, for this purpose, together with the use of today's industrial programming and control tools, a smooth and secure integration into automation environments is achieved. The proposed methodology is organized in different phases ranging from gestural research and selection to the final validation of the system in a pilot scenario. First, a data capture and annotation phase is completed, followed by the development and fine-tuning of the gesture processing, detection and segmentation algorithms. Then, the system is integrated by configuring industrial communication protocols, exhaustive tests are performed to maximize accuracy and response times, and finally validation is carried out in a controlled environment, demonstrating the feasibility and impact potential of this innovation in the field of labor automation.