A three-year research project has led to algorithms that help autonomous, wirelessly-connected vehicles anticipate the behavior of other vehicles to reduce braking and save energy.
The technology was developed by Clemson University mechanical engineering professor Ardalan Vahidi, and his team, according to a news release.
The less a vehicle brakes, the less energy it wastes through heat and the more energy efficient it becomes, the news release said. The team found its algorithms resulted in energy savings ranging from 8 percent to 23 percent, depending on the scenario.
“The big picture is that we’ll have more opportunities to save energy when autonomous cars that are connected to the internet and other wireless networks start talking to each other,” Vahidi said in the news release. “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.”
The team involved in the research tested its algorithms on two separate autonomous cars, a gas-powered Mazda and an electric Nissan, both connected to the same wireless network, allowing them to send and receive data, such as speed and heading, the release said. The cars took turns travelling a closed track in southern Greenville County so that only one real car was on the track at any given time.
Researchers used computer simulations to create “ghost” vehicles in front of and behind the Mazda and Nissan, making them think they were in traffic. It allowed researchers to be more aggressive and try difficult scenarios because any collision would be with a ghost car that caused no damage or injury.
Some of the ghost vehicles were autonomous, and some were driven by computer-simulated human drivers. Each test consisted of seven laps around the track with U-turns at both ends that caused slow-downs and often traffic jams, the release said.
The Mazda and Nissan saved more energy when following autonomous ghost vehicles than the ghost vehicles driven by simulated human drivers.
The autonomous ghost vehicles shared their intentions with the Mazda and Nissan several seconds ahead of time, giving the ghost vehicle and the real-life vehicle a chance to coordinate braking. The simulated human drivers, like real-life human drivers, were less predictable, giving the vehicles less time to work together, the release said.
“The experimental vehicles — the Mazda and the Nissan — saved 20-23 percent energy when following a simulated vehicle that was automated and connected to a wireless network,” Vahidi said in the release. “When following a simulated vehicle driven by a simulated human, we measured 8-12 percent energy savings compared to human driver baselines.”
While it wasn’t part of the research, the team also observed that the technology may alleviate traffic jams, the release said. With the cars anticipating what the preceding ghost vehicles would do, it smoothed out traffic flow, helping alleviate stop-and-go congestion.
Vahidi and his team are writing a paper that will detail more of the results. The paper brings to an end three years of research funded with $1.16 million by the US Department of Energy.