How Will Next-Gen Vehicles Drive Our Future?

When Dr. Mehrdad “Mark” Ehsani first heard about automakers investing in electric vehicle propulsion in the late 1980s, he was skeptical. “At first, I didn’t think it would be a viable alternative or support for conventional propulsion,” Ehsani said. “But the more I looked at it, the more it seemed promising, at least in certain areas.” He and his team have since conducted landmark research to make electric and hybrid vehicle propulsion viable, and his research has been cited more than 24,000 times.

Though Ehsani and his team have helped advance today’s electric and hybrid vehicles, his views on their implementation differ from most. For one, he is skeptical about their ability to single-handedly turn the tide on or significantly impact climate change. “If we could eliminate all carbon emissions from surface transportation in the U.S. tomorrow, its impact on global temperatures over 50 years would be roughly 1/40 of a degree Celsius,” he stated.

Instead, he sees more potential in electric vehicles making for better budget cars in the developed world and revitalizing vehicle infrastructure in the developing world. For example, he pointed to quality electric cars potentially replacing the noisy, inefficient rickshaws that populate India’s busy streets.
 

On Oct. 31, 2018, Chancellor John Sharp ’72, Dr. M. Katherine Banks (then dean of the College of Engineering) and the mayor of Bryan, Texas, boarded a trolley that took them in a simple loop around Bryan’s historic downtown. Dr. Srikanth Saripalli was behind the wheel, but the trolley drove itself—literally. It was one of two self-driving vehicles operating that day, making Bryan one of the first U.S. cities to have autonomous vehicles roving its public streets. “We worked closely with the city to ask the question: Are these vehicles ready to be deployed in dense pedestrian areas?” Saripalli said.

As director of Texas A&M’s Center for Autonomous Vehicles and Sensor Systems, Saripalli has worked at the forefront of driverless technology. His current research focuses on autonomous off-road vehicles for the military, which present unique obstacles since their sensors cannot rely on street markings for direction. “It’s a completely different domain,” Saripalli said. “The sensor may detect that the vehicle is on grass, but it can’t tell if there’s a puddle or anything beneath it that can bog it down.”

As for driverless cars on the road, he remains hopeful for further implementation and believes deep learning technology can help autonomous vehicles safely share the road with their human counterparts.
 

When car manufacturers first tested their high-end driverless vehicles in California, they prioritized safety and programmed the cars to follow state traffic laws to the T. They soon realized this was a mistake when the vehicles were involved in a series of low-speed collisions with human drivers. This inspired Dr. Dominique Lord, a traffic safety researcher, to investigate.

“We ran a study where students in a driving simulator followed a vehicle that mimicked the patterns of these driverless cars,” Lord said. Many of the student drivers nearly crashed into the virtual vehicles. “What happens is that automated vehicles are designed to follow the rules of the road almost too closely.” For example, a self-driving car stopping for three seconds at a stop-controlled intersection—as mandated by law—can throw off human drivers’ rhythm when they are used to other drivers doing a “rolling stop,” causing an inadvertent collision.

If autonomous vehicles eventually overcome obstacles like this, though, they have the power to make us rethink our traffic safety infrastructure. “Think about the rumble strips that shake your car and grab your attention when you’re veering off the road,” Lord explained. “This does nothing for a driverless car, as it uses cameras and sensors to maintain lane control. So, if we continue to invest in autonomous technology in the coming decades, we’ll have to rethink how we keep the roads safe for everyone.”