Going without a Jerk

This article is a republished version of the Technical Article Board at SSET

Abstract

Potholes on the road are a great threat to motorists using the highway. Currently there is very little infrastructure to locate these and repair them in a timely manner. This article presents a method to report the exact location of the portholes without any manual step. You just need to put a smartphone in your pocket.

Introduction

The article here discusses the use of the accelerometer in the phone for detecting the potholes and reporting them to the civil authorities through a database. The database can be used to locate the potholes on a map and take remedial measures quickly. The work explained here has been an extension of that developed in Nericell by Microsoft Research Bangalore. The Nericell gives a map view of the locations that and provides extension to accident detection due to abnormal braking.

Prelims of Accelerometer

The accelerometer is a device used in every smartphone and is capable of determining the acceleration along each perpendicular coordinates – X, Y and Z. The measurements are taken relative to the acceleration due to gravity, \(g (= 9.8 m/s^2).\) The directions are illustrated below. Using the X, Y components accelerometer can measure the abnormalities in the road plane. Along the Z direction, abnormalities due to level shifts can be detected- potholes or bumps for instance.

X, Y and Z directions of accelerometer

How It Works

It is possible to determine the following from the 3 reported acceleration graphs.

• Breaking Detection – accident reporting
• Pothole Detection
• Speed detection

If \(x\) is the distance, then \(\frac{dx}{dt}\) is the velocity, and \(\frac{d^2x}{dt^2}\) is the acceleration. If we look into the higher derivatives of \(x\), it is possible to detect the changes is acceleration corresponding to sudden changes in the accelerometer signal. It is possible to identify the sudden spiking of the signal along the \(x\) axis and translate it a jerk. Now the intensity of the jerk is dependent on the speed of the moving vehicle. At high speeds (ie > 25 kmph) , the jerk is prominent, while the jerk is less noticed for lower speeds (ie <25 kmph).

References

Mohan, Prashanth, Venkata N. Padmanabhan, and Ramachandran Ramjee. "Nericell: rich monitoring of road and traffic conditions using mobile smartphones." Proceedings of the 6th ACM conference on Embedded network sensor systems. ACM, 2008.