Road Roughness Measurement using Laser Profiler and Smartphone Sensors and Finding the Effects of Road Anomalies on Smartphone Roughness
In today's world in both developing and developed countries road infrastructure is majorly used for transportation, and its use is increasing exponentially by the day. This increase in the use of road facilities is causing deterioration at a very rapid pace. To measure the amount of deterioration accurately road maintenance agencies detect the roughness of the roads through expensive equipment like laser profilers and optical profilers which is very costly and in most developing countries is unavailable. To cope with this issue modern researchers have focused on developing pavement roughness models by using smartphone sensor data. This approach is relatively very cheap as compared to expensive road profilers but it can be very inaccurate at times. There can be various causes for the inaccurate roughness measurement but mostly it's because of the pavement anomalies which are encountered during profiling of the road. If these anomalies are not taken into consideration during the profiling of a road, these can cause huge errors in the roughness values. This research focuses on developing road roughness model by highlighting these anomalies and removing them from the roughness profile to obtain a true roughness profile. To validate this method the smartphone roughness profile was then measured with the accurate laser profiler. An Artificial Neural Network (ANN) is established for the identification of the most commonly occurring road anomalies. Analysis results show that the roughness profile from which the anomalies were removed showed a lot of similarity with the accurate laser profiler.
Keywords - Laser Profiler, MATLAB Simulink, Road Roughness Models, Smartphone Acceleration Sensor