Fuzzy Logic Based Speed Control System For Brush Drive Motor In High Rise Building Glass Cleaning Robot

Abstract

Abstract Window cleaning on high-rise buildings is a high-risk task for workers, especially regarding accidents at heights. As the number of high-rise buildings increases, the demand for safer and more efficient cleaning systems becomes more urgent. This study develops a robotic window cleaning system that integrates the YOLOv5 object detection algorithm with a fuzzy logic-based motor speed controller. The system is designed to detect stains on the glass using a camera and adjust the brush motor speed based on the level of dirt detected. The main goal of this research is to improve energy efficiency and reduce the risk of work-related accidents during manual cleaning at height. The methods used include hardware design based on Jetson Nano, implementation of YOLOv5 for stain detection, and the application of fuzzy logic to control motor speed. Testing on 30 image samples shows a stain detection accuracy of 90.82%, while the use of fuzzy logic can save power consumption by up to 38% compared to conventional control methods. The conclusion of this study is that the integration of YOLOv5 and fuzzy logic in the robotic window cleaner provides significant energy savings and enhances work safety in high-rise buildings. Keywords: window cleaning robot, YOLOv5, fuzzy logic, motor control, stain detection