Gesture Recognition for Mechatronic System Control Using Motion-Capture Suit Rokoko SmartSuit Pro II

Abstract

This study explores the processing and evaluation of data collected from a motion-capture suit to facilitate gesture recognition for controlling mechatronic devices. In the context of advanced human-machine interaction (HMI) and human-robot interaction (HRI), the research investigates the utilization of motion-capture technology as a tool for capturing and interpreting operator gestures. The study begins with the selection of a cost-effective motion-capture suit, followed by a comprehensive analysis of data acquisition techniques. A program is then designed to process and evaluate the captured data using computational intelligence methods, including Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks. The results demonstrate the efficacy of these models in accurately recognizing gestures, which are then translated into control commands for mechatronic devices. This work contributes to the development of more intuitive and efficient interfaces for human-machine communication, with potential applications in various fields, including robotics, healthcare, and industrial automation.