A Real-Time Robotic Movement Control System Based on Hand Gesture Recognition Using Human Computer Interaction (HCI) Principles

This paper presents a real-time robotic movement control system (RMCS) utilizing hand gesture recognition through a Cascaded CNN-SVM approach. The system was designed to enhance human-robot interaction by allowing users to control RMs seamlessly through gestures. Extensive testing was conducted on a dataset comprising various gestures, achieving an impressive overall recognition accuracy of 93.6%. Specific gestures, such as “Wave” and “Point,” demonstrated recognition accuracies of 95.2% and 93.8%, respectively. The system was also evaluated for response times, with an average response latency of 125 ms, ensuring efficient real-time operation. User feedback highlighted high levels of satisfaction, with average ratings of 4.4 for intuitiveness and 4.5 for the likelihood to recommend the system. These results indicate that the Cascaded CNN-SVM model is effective in recognizing gestures with high precision while maintaining low latency, making it suitable for practical applications in RMCS.