Innovative Glove Design for Enhanced Robotic Hand Control using PWM Techniques

The demand for robotic systems to assist individuals and undertake complex tasks has grown, particularly in enhancing mobility for those with physical disabilities and ensuring safety in hazardous environments. Traditional sensors in robotic hand control systems are often costly, fixed, and lack adaptability. This research presents an innovative glove design integrating rotary potentiometers and three advanced Pulse Width Modulation (PWM) techniques for controlling robotic hands. The glove's increased rotational range, up to 150°, offers enhanced precision and control, making it suitable for diverse applications. We explore the efficacy of Arduino MEGA2650, 555 timers, and Artificial Neural Networks (ANNs) in generating PWM signals. The Arduino provides flexibility, 555 timers offer cost-effective precision, and ANNs enable adaptive control in complex environments. The study demonstrates that rotary potentiometers offer a reliable and cost-effective alternative to flex sensors, with consistent performance across multiple uses. Furthermore, integrating deep learning models trained on Electromyography (EMG) signals can significantly improve the accuracy and reliability of robotic hand control systems, especially in prosthetic applications. The findings suggest that the proposed system not only broadens the applications of robotic hands but also enhances their functionality and user-friendliness.