As part of my MS project, I have been working on an android app that can detect and subsequently predict speed bumps. Although it is still a work in progress, I was able to get the detection part to work. Following image shows the speed bumps when travelling from Bangalore to Mysore:
Based on GPS data and accelerometer data from my android app, I could detect using R scripts (run offline) if a bump has occurred at any point in time. These were then plotted over a lat-long map. Google maps integration is still not done. So, I have just overlaid my image over google maps using gimp.
Next target is to upload the data points to a server running Python scripts to create a database of speed bumps.
For me, the motivation to do this is simple. Speed bumps on highways are just frustrating. If a database of such speed bumps exists, then it is easy to get warnings on the phone based on GPS location. Building the database is the tough part.