Two University of California graduate students have proposed a system to allow tighly packed clusters of self-driving cars to travel at speeds up to 120 miles per hour, using exclusive lanes on existing roadways. The “Hyperlane” plan works similarly to high speed toll lanes, but in this case, a central computer would coordinate the traffic.

Fully autonomous vehicles are not yet allowed on most roads, but some companies, such as Tesla and Volvo, offer features like adaptive cruise control and steering assistance that requires little input from the driver.

The graduate students, Anthony Barrs and Baiyu Chen, created their Hyperlane plan in response to the projected expense of high-speed rail systems for high traffic routes, such as the overburdened road system from San Francisco to Los Angeles. High speed rail for that route is projected to cost 139 million dollars per mile. For this reason, the rail project has been postponed repeatedly. The plan Barrs and Chen are proposing would cost just 12 million per mile.

“Long story short: high-speed rail didn’t pencil out over here,” said Barrs. “We’re primarily looking at cities with major nodes that need to be connected, where high-speed would create high value.”

They said the Hyperlane plan would best serve metro areas like the San Francisco Bay Area, Dallas-Ft Worth, and Baltimore-DC. High density cities like New York lack the space to expand current roads, and medium and low density cities lack the volume of traffic to make construction worthwhile.

“If you’re in DC and you have this Hyperlane connecting the DC metro area, you can now fly out of the Baltimore-Washington international airport much more easily,” said Barrs, discussing the high traffic and limited public transit options on the routes to the area’s airport. “This creates additional transportation options.”

The Hyperlane takes the idea of lower congestion, higher speed lanes, and uses it to allow autonomous vehicles to travel at up to 120 miles per hour, in computer controlled clusters of vehicles. A centralized computer would be in constant communication with every vehicle using 5G technology, allowing tightly packed traffic to move at high speeds.

According to Barrs, “We liken the Hyperlane network to an air traffic control system.”

Sensors placed in the road would receive information about traffic congestion, weather, accidents, and other variables, adjusting vehicle speed in response. The lanes would charge fees based on the level of demand, similarly to Uber’s pricing structure.

Some lanes would be taken from existing highways, while in other areas, new lanes would be built next to highways.

“Berkeley to Palo Alto would be a 40-minute trip, and by splitting among five or six people, would cut down the cost,” according to Chen. During rush hour, that trip can take up to 2 hours. However, the idea is still in the planning stages. The pair are looking for investors to back physical testing of the idea.
“There’ s a step that exists between theoretical and physical testing, and that’s getting a coalition of partners to fund it,” according to Barrs.

“We’re looking at state transportation officials and transportation companies like Uber and Lyft.”

Leave a Reply

Your email address will not be published.