MIT App Would Help Cars Avoid Traffic Jams

According to the researchers who built the program:

RoadRunner is an in-vehicle app for traffic congestion control without costly roadside infrastructure, instead judiciously harnessing vehicle-to-vehicle communications, cellular connectivity, and onboard computation and sensing to enable large-scale traffic congestion control at higher penetration and finer granularity than previously possible. RoadRunner limits the number of vehicles in a congested region or road by requiring each to possess a token for entry. Tokens can circulate and be reused among multiple vehicles as vehicles move between regions. We built RoadRunner as an Android app utilizing LTE, 802.11p, and 802.11n radios, deployed it on 10 vehicles, and measured cellular access reductions of up to 84% and response time improvements of up to 80%. In a microscopic agent-based traffic simulator, RoadRunner achieved travel speed improvements of up to 7.7% over an industry-strength electronic road pricing system.

Of course, the improvement was only based on a simulation. It is too early to tell whether the service would hold up in actual traffic jams.

The project is one of what have become dozens of “self-driving” car plans. The most prominent of these is Google Inc.’s (NASDAQ: GOOG) “driverless” car, which the search company has been testing for over a year. Granted, the goals of this program are much more ambitious than avoiding traffic jams. However, Google recently admitted that its driverless car has a number of drawbacks. An MIT Technology Review analysis of the Google product showed that it has trouble parking and does not work well in heavy rain.

The failures of driverless cars to actually be “driverless” have not kept most of the world’s major manufacturers from experiments of their own. For the time being, these companies may do no better than selling cars that keep drivers out of traffic jams — maybe.

ALSO READ: Tesla to Move Into Driverless Car Business