|'Ghost' AV research may put the brakes on stop-and-go traffic|
The best way to maximize the energy efficiency of autonomous vehicles may be to put them in traffic with other AVs.
Researchers at Clemson University have developed algorithms to help autonomous, connected cars better anticipate what other vehicles will do, thus preventing the cars from braking too frequently and wasting energy through heat.
"There are a lot of groups focusing on autonomous vehicles, but the focus on how they can be energy efficient is not as mainstream. That's our niche," Ardalan Vahidi, the Clemson mechanical engineering professor whose team created the algorithms, said in a statement.
The team tested its algorithms on a closed track, using vehicles rigged to operate autonomously.
The cars were connected wirelessly to allow them to send and receive messages, such as speed and heading. The researchers then used computer simulations to add "ghost" vehicles in front of and behind the real AVs to mimic road traffic. Some of these ghost cars were autonomous, others were operated by simulated human drivers.
Researchers found that the real AVs saved the most energy when they followed autonomous ghost vehicles. The ghost AVs were able to share their intentions with the real AVs several seconds in advance, thus allowing the vehicles to coordinate braking.
"The simulated human drivers, like real-life human drivers, were less predictable, giving the vehicles less time to work together," Clemson said in the news release.
Using the algorithms to optimize the way AVs operate yielded energy savings from 8 to 23 percent.
In a video describing his team's research, Vahidi says networked automated vehicles can ease traffic for other cars on the road.
"Automated vehicles, by driving more energy efficiently, help smooth and harmonize the traffic behind them, reduce the stop-and-go behavior. Phantom traffic jams, as they say."
-- Leslie J. Allen