IoT-complex for Monitoring and Analysis of Motor Highway Condition Using Artificial Neural Networks

Monitoring the road condition has acquired a critical significance during recent years. A few major factors add up to the importance of current research: (a) the information retrieved may support the decision-making process of drivers to the use or avoidance of certain highways; (b) smooth road surface causes less damage to the car chassis and suspension system; (c) the dependability of the car's control system remains; (d) valid information on the road surface quality is the basis for updating the knowledge base of the road management companies and organizations and thus challenges them for regular surface reviews and repairs. The tool considered in the paper is the real-time IoT-complex with Android application that automatically collects the data from the mobile triaxial accelerometer and gyroscope, shows the road trace on a geographic map using GPS and sends all recorded entries to the cloud-based computation algorithms. Different types of artificial neural networks are applied to training data to classify road segments and build the model. The experimental results show a consistent accuracy of 90 and higher percent. Using this approach the expected output is the visualization of the road quality map of a selected region. Hence, the constructive feedback may be provided to drivers and local authorities. The long-term benefit from this system is the performing of the road network state comparison throughout various time intervals and checking up on the road construction projects, whether or not they meet the assigned quality prerequisites.