Novel Pothole Detection System Using YOLOv8 Algorithm

Road humps and potholes are major causes of car damage, traffic inefficiencies, and road safety risks. Repairs are delayed by the labour-intensive, sluggish, and sometimes reactive nature of traditional techniques for identifying and maintaining road infrastructure. Using IoT-based ultrasonic sensors and YOLOv8 machine learning algorithm, this study proposes a novel method for the automatic identification of potholes and humps. The system uses a machine learning model to classify road conditions after using an Arduino Uno microcontroller to gather real-time surface data. To support a cooperative infrastructure monitoring network, the system also has an easy-to-use web interface where anyone may report road anomalies. With a 92 % identification accuracy, the model showed promise for greatly enhancing the efficacy and efficiency of road repair activities. In addition to facilitating prompt repairs, the suggested method promotes preventative maintenance and urban planning, which leads to more efficient and sustainable city growth. In the future, the system will be expanded to bigger road networks, predictive analytics will be included, and interaction with autonomous car navigation systems will be investigated. This strategy is a significant advancement in using technology to improve infrastructure management and road safety.