Detection of Potholes and Speed Breaker for Autonomous Vehicles

Abstract

As it is now the generation of autonomous vehicles, Every automotive company is aiming to build an autonomous vehicle, so an application to detect potholes and speed breakers is really necessary because poor road conditions are a constant concern since they annoy the public. Inefficient roadways increase traffic in a city. This directly bothers the traveller. Additionally, it harms a number of important car parts, which has a long-term negative impact on how well the car performs. Despite all of these inconveniences and harm, accidents are the main problem caused by bad roads that need to be addressed. The study highlights how crucial pothole and speed breaker detection are for enhancing driver safety and road infrastructure. The YOLOv5 implementation is used for the pothole detection and speed breaker models. By testing real-time data, the model provided an accuracy of 85 percent for potholes and 83.8 percent for speed breakers. More training and validation of photos can increase accuracy, which can then be used in the real world with sensors to indicate detection. This type operates automatically, assisting motorists with safe driving. More sensors might be added to the model to produce data on climate-vulnerable regions, which could then be used to design suspension systems and other car parts. Any unintended failure of the roads in the database may be promptly updated by cars with this model, serving as a warning.