The development of energy-efficient robotic systems has become increasingly important, particularly in servo-driven robotic arms where continuous PWM signals lead to unnecessary energy loss and thermal stress. This study proposes an adaptive pulse-width modulation (PWM) algorithm designed to activate servo signals only during motion phases and automatically deactivate them once the target position is reached. A four-degree-of-freedom robotic arm prototype was developed using an ESP32 microcontroller, MG90S servos, and ACS712-based current monitoring to evaluate power efficiency under conventional continuous PWM and the proposed adaptive control. Experimental results demonstrate a 28–33% reduction in average power consumption, a decrease of 6–8 °C in servo operating temperature, and the preservation of positional accuracy within ±5%. These findings confirm that significant energy savings and thermal improvements can be achieved without modifying hardware components. The proposed algorithm offers a practical, lightweight, and software-based optimization approach suitable for educational, research, and low-power robotic applications. This work introduces a distinct adaptive activation strategy that fully disables PWM in steady-state conditions, representing a low-cost and effective contribution to sustainable servo control.
Copyrights © 2025