Undergrad Research Project - Road surface roughness measurements with smartphones

Fall 2015

Nandini Ramakrishnan
Martial Hebert
Project description

The goal of this project is to measure the roughness of road surfaces with data from smartphones, which have been mounted in vehicles. Through the course of the project, we will develop an algorithm, with which detection of poor roads and infrastructure inventory will become more accurate and efficient. The relevant data are the readings from the accelerometer, gyro, and GPS. Images that have been taken at the same time will be used for visual verification, of which a large database already exists. With the use of Android programming, we will in particular learn how to record sensor readings, investigate the sensor properties, i.e. their update rates, uncertainties, errors etc. With this knowledge we will develop algorithms to determine surface roughness of roads. The algorithm should be able to distinguish between smooth (newly paved) roads, rough roads, potholes, speed bumps, and possibly roads with shoving. The algorithm should work at different speeds.

Return to project list