Clemson University researchers have been awarded a $1.16 million grant from the U.S. Department of Energy to study the use of automated and connected vehicle technology to boost energy efficiency and create a cleaner transportation system.
The DOE recently announced the investment as part of a $19.4 million package for more than a dozen projects nationwide to accelerate research of advanced battery, lightweight materials, engine technologies, and energy-efficient mobility systems.
Clemson’s research will be led by Dr. Ardalan Vahidi, an associate professor of mechanical engineering on the university’s main campus. Collaborating on the two-year project will also be Dr. Beshah Ayalew, an automotive engineering professor at Clemson University’s International Center for Automotive Research in Greenville, and Dr. Yunyi Jia, director of the collaborative robotics and automation lab at ICAR.
“Connected and automated vehicles are marketed for their increased safety, driving comfort, and time-saving potential,” Vahidi said. “With much easier access to information, increased processing power, and precision control, they also offer unprecedented opportunities for energy-efficient driving.”
According to Vahidi, some vehicles are already outfitted with automated and connected technologies, including sensors that use video, radar, and Lidar, a laser-based technology that continually and accurately scans and maps the environment around the vehicle.
The technologies, which could one day allow vehicles to communicate with each other and surrounding transportation infrastructure, are routinely touted for their potential to prevent automobile accidents and save human lives. In fact, the tech could eliminate 90 percent of U.S. traffic fatalities per year, according to a 2016 report from McKinsey & Company, a consulting firm.
But Vahidi and his colleagues hope to demonstrate the energy-saving benefits of connected and automated vehicle technologies. More specifically, the researchers plan to use cellular communication and dedicated short range radar to reduce unnecessary braking and idling, which waste fuel.
“Every time you engage your brakes, precious energy of the moving car is wasted as heat in the brake pads. To speed up again, energy has to be used. This is more so in conventional vehicles that do not have regenerative brakes,” Vahidi said.
He added that many autonomous and connected vehicles are designed to behave like human drivers, meaning they are typically reactive to their immediate surrounding, which results in more braking and similar “short-sighted decisions” that consume more energy.
That’s why Vahidi and his colleagues plan to design speed and lane selection algorithms “that are anticipative in nature rather than reactive.”
The researchers plan to collaborate with the DOE’s Argonne National Lab in Illinois and the International Transportation Innovation Center (ITIC) in Greenville to test their algorithms.
“We plan to do this first via detailed computer simulations and then in isolated experiments on a test track at ITIC. In computer simulations, we will emulate the motion of thousands of vehicles on realistically modeled roads and evaluate impact of a limited number of connected and automated vehicles on energy efficiency of mixed traffic. … In experiments, we will drive one or two autonomous vehicles on the ITIC test track and create realistic interactions between them and with our simulation environment,” Vahidi said.
If successful, the group’s algorithms could not only boost gas mileage for autonomous cars but also reduce their environmental impact.
In 2015, the last year data is available, transportation accounted for 27 percent of total U.S. greenhouse gas emissions, which can cause multiple health issues such as asthma and heart attacks, according to the Environmental Protection Agency. The transportation sector is the second largest contributor of the country’s emissions after the electricity sector.
Autonomous vehicles could reduce energy consumption in transportation by as much as 90 percent, according to research from the Department of Energy. But they could also promote longer commutes and faster driving, increasing consumption by more than 200 percent.
Vahidi’s findings could reduce energy consumption in autonomous and connected vehicles by as much as 20 percent. “Our proposed solutions enable energy savings and reduction of emissions relying mostly on software, information, and connectivity and with minimal hardware investments beyond what is expected and is standard for vehicle connectivity and automation,” Vahidi said.
Q&A: The Future of Driverless Cars with Clemson University Researcher Ardalan Vahidi
While driverless cars won’t become commonplace overnight, they’re quickly becoming more of a reality as Audi, BMW, Ford Motor Co., and other automakers continue developing their own versions of cars that require little to no human input.
We sat down with Clemson University researcher Ardalan Vahidi, an associate professor of mechanical engineering, to discuss the impending future for safe, hands-free transportation and its implications.
Are autonomous vehicles really better drivers than humans?
AV: Human drivers are often reactive when following other cars, as their view is often blocked by the preceding car and therefore their event horizon is very limited. In sudden slowdowns, they often fail to consider the cars approaching from behind. This is not only disruptive to traffic flow and is unsafe, but it can result in inefficient slow-down of multiple vehicles. Balancing the position proactively with respect to the cars in the front and back is cognitively demanding for humans. Most autonomous cars do not necessarily do better.
How will autonomous vehicles improve peoples’ lives?
AV: They will make transportation safer, more comfortable, and hassle free. If done right, the technology can save energy and reduce the environmental impact of transportation sector. Autonomous vehicles save people considerable time by not requiring them to be at the steering wheel. People can be more productive during daily commutes or avoid unnecessary commutes for delivering goods. Less parking and narrower roads can support transportation of the same number of people, and so our future cities may have more friendly and greener landscapes instead of sprawling roads and parking lots.
How long will it be before autonomous vehicles are on the road?
AV: There are different projections. Major auto manufacturers, technology firms, and startup companies have started a race toward building fully automated cars. Many automated functions such as adaptive cruise control and lane-keeping assist are already available on several production vehicles. It is expected that the first fully automated vehicles be available for sale before 2020. A projection is that 20 to 40 percent of vehicle sales be automated by 2030 and full penetration could happen in several stages over the next few decades.
What are some challenges that still need to be overcome?
AV: Reliable sensing of the environment is a main challenge. Current autonomous cars use expensive means to sense what is surrounding them and still cannot identify all situations that a trained/experienced human driver would. The reliability needs to increase followed by public acceptance, and the cost has to drop for widespread adoption. Legal and institutional issues are other barriers to overcome, but in my opinion, those will not be major hurdles when reliable operation of a fleet of autonomous cars is demonstrated.