ML approach to identify and classify potential road anomalies using smartphone sensors.
Our main goal is to use sensors such as accelerometer, gyroscope, etc to spot anomalies on the road by analizing the time series that stem from the sampling of such integrated sensors. Such time series will be generated by a smartphone running a simple Flutter app while driving in a vehicle.
We will collect data and train a Machine Learning Model to recognize patterns in those time series that can be an indication of an anomaly being present.
We intend to use GPS as a way to stablish ground truth once a vehicle drive over an anomaly and to provide information as to where an anomaly is located approximately.