A recent post on the OpenStreetMap Talk mailing list pointed to an article in MIT news about an interesting experiment called “Wiki City Rome“. The project, developed at MIT, started two days ago and uses anonymous data from mobile phones and other wireless devices to “illustrate the city’s pulse in real time”. The MIT news article explains:
By looking at a city using a “real-time control system” as a working analogy, the Wiki City project studies tools that enable people to become prime actors themselves in improving the efficiency of urban systems. In coming years, the Wiki City project will develop as an open platform where anybody can download and upload data that are location and time sensitive.
This is a superbly exciting experiment, and it echoes a dream that I have had ever since I began contributing to the OpenStreetMap (OSM). Almost everywhere that I drive now I have my Garmin GPS sitting on the dashboard collecting traces. To actually add this data to the map, I have to transfer the GPS trace onto my computer and tag the features with meaningful descriptions and names before uploading it.
But imagine if my car had a GPS receiver built-in (increasingly common), and had a wireless connection to the Internet. It could automatically upload its path and position to a sort of super-OSM project. If every car was thus equipped, then routing systems could become fairly intelligent. Whenever a new section of road was opened, the first car to travel along it would add it to the map. As more and more cars travelled the route, the map could learn properties of the road by examining the typical speed of various vehicle types. This sort of information could be time-dependent, and the map would learn which roads formed the best routes for specific times of the day or week.
Your car would know where other cars were along possible routes, and things like accidents or road-works would appear as congestion. Each car could optimise its own route, allowing for almost instant compensation for traffic jams. Radio traffic reports would become quaint memories.
As was mentioned in the MIT News article, such a system could dramatically improve the efficiency of city systems. Optimising the travel times of each vehicle would reduce our fuel demand and emission level. In essence each vehicle on the road would be an individual node in a collective “road intelligence”.