We’ve saddled production vehicles with high-quality sensors in the interest of safety — or what is currently referred to as ADAS or advanced driver assistance systems. But while most of these sensors are on 100% of the time the vehicle is running, they’re primarily doing the job of a quiet watchdog. They’re active and sensing, but accidents are infrequent and things like advanced cruise control aren’t running but a portion of a duty cycle.
Instead, we need to consider sensors in our vehicles as our subjugates — that we can control and can use for safety, data collection and profit. During a 30-minute commute, a driver should be able to produce meaningful data from her vehicle’s sensor suite such as:
Updating out-of-date road and map data Lane marking and map matching Traffic information Localized weather, road surface and visibility information Accident reconstruction of nearby vehicle incidents The modern vehicle is a sensor workhorse, but has a very narrow idea of work. We need to broaden it as a networked sensor workhorse, with the owner / operator benefitting by letting her vehicle’s sensors add to a corpus of sensor data. All of the things listed above are services and markets that measure in the billions.
I call this act of selling back vehicle sensor data ‘machine earning’ and I believe that over the coming decade we will see a number of markets emerge to serve this need, some coming from within the vehicle and some perhaps operating on top. My friend Colin O’Donnell calls this ‘captcha for cars.’
As vehicles get more sensors, the driver gets more ‘workers.’ Now we need to unleash the power of these sensors for a greater, more valuable good. Machine earning will be a new way to ‘work’ and earn for miles traveled.