Gesture Recognition for Human-Robot Interaction Through Virtual Characters

Human-Robot Interaction (HRI) has grown increasingly important with the integration of various types of robots into industrial and daily life aspects. Non-verbal communication plays a vital role in HRI and heavily relies on gesture recognition. The ability of robot’s interpretation of human gestures is essential for enhancing the overall user experience. To achieve precision and effectiveness of gesture recognition in HRI, the incorporation of Virtual Reality (VR) technology offers promising avenues. In this article, we describe the development of a gesture recognition pipeline tailored to HRI in Precision Agriculture scenarios. Its key feature is the use of Virtual characters and Machine learning techniques to allow agricultural robots to recognize gestures performed by both digital agents and humans. We address challenges by presenting a set of gesture definitions, data generation, and system evaluation using varies combinations of real and virtual simulated data. Our results demonstrate how the adoption of virtual simulations can significantly increase the system’s accuracy and efficiency.