Implementation of Human Gestures in the Control of Collaborative Robots

The current paper explores the possibilities of introducing a gesture control method to command and collaborate with robots. In the first part, aspects of the current limitations of tele-robotic systems and the communication models adopted in the industry are presented. As for the method used to detect hand gestures, it was selected a machine learning model and shrank to run on a conventional microcontroller. By leveraging machine learning techniques, the precision and repeatability of the gesture-based commands were significantly improved. In the final part of the paper, the implementation stages are presented alongside the results of the gesture detection model. The proposed solution not only enhances the efficiency of the tele-robotics systems but also paves the way for resource efficiency on microcontrollers. Furthermore, the paper presents practical applications according to industrial technologies. The adaptability and versatility of the developed model make it a promising candidate for a wide range of applications.