Accelerometer data can reveal as yet unreported hazards
The DriveScore app uses phone accelerometer data to provide personal driving style feedback to the owner on harsh breaking, swerving and sharp vehicle movements. But what if many vehicles are behaving the same way at same place on the road? It could well be a strong indicator of an unreported incident and the cars behind should be warned.
If Waze could anonymously combine the same accelerometer data from multiple phones, it could be used to predict the hidden hazard. Astronomers use the same logic to identify the presence of black holes in space by observing the changes in acceleration of nearby particles, only in this case it would be looking for patterns in how multiple cars are behaving.
As an example, if every driver in lane 3 of a motorway is slowing, changing lanes then speeding up again at the same static location, there will likely be a common cause for the acceleration changes, such as a stranded vehicle, accident or other obstruction. This was the case for me at Christmas when on a tight bend connecting 2 motorways I came across at accident in lane 2 and had to swerve to avoid it, as did everyone behind me. Had the 'DriveScore event' on my journey been shared by the multiple drivers who experienced the same sudden acceleration change, it could have prevented a further pile up.
Even in good traffic, by combining the same detailed GPS and accelerometer data from multiple cars, Waze would be able to identify common patterns in speed and acceleration, such as that 80% of vehicles are changing lanes at the same a relative point along a motorway which would indicate that there is a slow moving vehicle being overtaken by all. Drivers behind could then be warned accordingly.
It must also be possible to detect from a cars speed when a driver is using cruise-control and when that speed changes or the car changes lane, it might prove useful to those behind. Even cars that have adaptive cruise-control can only react based on the speed of the
bumper ahead, with help from Waze, patterns in acceleration and deceleration could be shared from much further ahead to guide the speed of the approaching traffic.