This application helps the authorities to fetch the pothole data within a locality via people’s involvement. Data collection via crowd-sourcing helps to further improve the pothole detection. A cost-effective solution was developed, which provides a detailed description of a pothole. This contextual information of the pothole played a vital role in categorizing
pothole based on its destructiveness. An empirical model was developed that processes the provided detailed description of potholes in order to set a priority to its repair.
Another goal of this project is to eliminate the use of external hardware; the developed
application uses the two internal smartphone sensors, which includes location sensor and camera sensor. Both these sensors are only operated when a user uses the application to report an incident. The application establishes a warning system to the commuters within a locality, this is achieved by the integration of the pothole data with Google Maps. The maps feature could be used to notify vehicular users about the possible cavities on the road.
Most of the factors determining the priority are collected autonomously using the smartphone’s location sensors and require minimal data collection from the user. The application also includes the development of a low load security algorithm for transmitting sensitive information from mobile devices. This research builds up a model for determining the priority of the pothole based on contexts, goals, context and confidence level in the collected data. The prototype could be used as a guide to further help the authorities to tackle one of the major concerns of road maintenance.

Author: Sudhish Thazhakasseril, Igor Khokhlov, Leon Reznik