Abstract: The proposed system has an infrastructure free approach for roadway detection and identification based on data collected through a smart phone application. There are very few systems available that provide detection of roadway obstacles using accelerometer. The proposed system concentrates on the problem of detecting and classifying roadway obstacles so as to differentiate between actionable and non actionable obstacles. The solution results can be obtained using machine learning algorithms which will be used to aggregate the acquired information. The proposed system is available on a social platform so that users would be able to provide indirect contributions towards development of smart city. The Data captured from the proposed system provides aggregated result of the obstacles irregularities being computed.

Keywords: Classification, machine learning, artificial intelligence, database system.