ABSTRACT
Road quality assessment is a time-consuming and labour-intensive activity that requires surveying the road and capturing observations on paper and then reporting them on a cartographic map. Curl Analytics adopted an innovative pilot approach in collaboration with the government of Andhra Pradesh, where drones were flown over the road to capture images of the road, and the depth of the potholes was measured using the onboarded LiDAR. Machine learning methods were used to classify road segments into good, average, bad, or mud roads, along with information on the line markings. This information, along with the dimensions of potholes, is reported in digital format to the state authorities for real-time monitoring of road conditions and taking up repair activities.
