Researchers at MIT are working on project that will collect real-time traffic and other road conditions from drivers and share it across a network in a bid to potentially reduce commute times.

Called CarTel, it uses a computer in the car about the size of a cellphone that connects via existing WiFi networks to a web server. Across a number of drivers, the system gathers data that can be used to predict traffic routes based on real-time and historical traffic information.

“The idea is to use sensors in cars as a platform to measure things on the road going beyond just traffic to pollution and even potholes,” says Samuel Madden, professor at MIT’s department of electrical engineering and computer science. “We call it the mobile sensor networking project.” Madden and his team have been working on this project for about three years.

MIT could have done better instead calling Dash Navigation, a Sunnyvale, California-based company that has had a similar system running for the last two years.

Earlier this year, after months of running trials, Dash launched a personal navigation device called the Dash Express that features access to real-time traffic information gathered from other Dash devices on the road and transmitted using both WiFi and cellular networks.

“We have had this built from the ground up and running 24 x 7,” says Rob Currie, president and chief operating officer of Dash, “and it has about 10 million miles logged on it.”

Dash and MIT’s answer to what most drivers would love to know before they set out on their morning commute is promising though both technologies have some way to go before users can completely rely on them, says an analyst.

MIT’s system deployed on nearly 50 Boston-area cars, most of which are taxis, tracks traffic by monitoring each vehicle’s speed at different points during the trip. This week the researchers introduced new low cost hardware, a home brew of GPS antenna and WiFi access point for about $100 a pop. A Dash navigation device costs $300.

The significant innovation here is the ability of the device to quickly transfer data as the car moves at high speeds, says MIT’s Madden.

“The car uploads data as it drives by open wireless access points and is very efficient in associating with them,” he says. “It can take about 15 seconds to connect using a regular system but we can do it in 360 milliseconds.”

That raises the question of security as open WiFi networks pose significant threats. Madden believes a commercial implementation of MIT’s system would be done using partner WiFi networks such as free citywide municipal networks or hotspots from service providers. Dash meanwhile uses cellular networks to transfer the data collected.

Important as the issue is of traffic for drivers, both Dash and MIT’s approach may be too alike to change how users view real time traffic data, says Bart van der Horst, managing director for research firm Canalys.

Currently collecting real time traffic data involves gathering information from existing users on the road, which means only major roads in big cities are likely to covered.

“That creates a trust issue with users,” says van der Horst. “You just don’t know on which roads the data will be available and when it will not.”

Plus real-time traffic data is just not the focus for most device buyers right now, says van der Horst. “The PND market now is very much focused on low price and simplicity,” he says.

Adding connectivity makes the device more complex and increases its price, something that manufacturers in the U.S. don’t want.

That’s why TomTom, for instance, has real time traffic information in partnership with Vodafone available in Europe, but not in the U.S. Neither does its rival Garmin.