Road traffic is affected by the presence of obstacles and defects present on the road, e.g., potholes, cracks, and so on. Properly managing these issues is the task of drivers. However, in many situations, the presence of these conditions results in changes in the mechanical condition of the car, due to a delayed reaction. The proposed driver assistance app detects and classifies road defects in real time and alerts drivers. It also records their locations and shares this information with drivers and managing authorities for timely intervention. This project begins by identifying and classifying roadway damage using machine learning, then analyzing data to find safe paths around obstacles. Training datasets come from an existing repository and are supplemented with data processed for the Constanta city area.