stub Microscale Concave Interfaces Could Help Autonomous Vehicles - Unite.AI
Connect with us

Artificial Intelligence

Microscale Concave Interfaces Could Help Autonomous Vehicles

Updated on

A new study details how microscale concave interfaces (MCI) could help autonomous vehicles read signs. MCIs are structures that reflect light to produce optical phenomena.

Qiaoqiang Gan is an engineering researcher at the University of Buffalo. According to Gan, these effects could be used in the future to aid autonomous vehicles in identifying traffic signs.

“It is vital to be able to explain how a technology works to someone before you attempt to adopt it. Our new paper defines how light interacts with microscale concave interfaces,” Gan says.

The research was published in Applied Materials Today on August 15.

The collaborative study was led by Gan, PhD, professor of electrical engineering in the UB School of Engineering and Applied Sciences. It involved a team from UB, the University of Shanghai for Science and Technology, Fuda University, Texas Tech University and Hubei University. The first authors of the paper include Jacob Rada, UB PhD student in electrical engineering, and Haifeng Hu, PhD, professor of optical-electrical and computer engineering at the University of Shanghai for Science and Technology.

Retroreflective Material

The study was heavily focused on a retroreflective material, or a thin film consisting of polymer microspheres placed on the sticky side of transparent tape. Those microspheres are partially embedded in tape. 

When white light is shined on this film, it is reflected and causes the light to create concentric rainbow rings. On the other hand, a single-colored laser generates a pattern of bright and dark rings. Infrared lasers also produce distinctive signals consisting of concentric rings when reflected.

The Experiments

The research also involved experiments that used the thin film in a stop sign. According to Rada, when the patterns were formed, they showed up on a visual camera that detects visible light, as well as a LIDAR camera that detects infrared signals. 

“Currently, autopilot systems face many challenges in recognizing traffic signs, especially in real-world conditions,” Gan says. “Smart traffic signs made from our material could provide more signals for future systems that use LIDAR and visible pattern recognition together to identify important traffic signs. This may be helpful to improve the traffic safety for autonomous cars.”

“We demonstrated a new combined strategy to enhance the LIDAR signal and visible pattern recognition that are currently performed by both visible and infrared cameras,” Rada says. “Our work showed that the MCI is an ideal target for LIDAR cameras, due to the constantly strong signals that are produced.”

The technology is available for licensing, and a U.S. patent for the retroreflective material has been issued. A counterpart has also been issued in China. The patent holders are Fudan University and UB.

According to Gan, the film will be tested with different wavelengths of light, as well as different materials for the microspheres. The goal here is to enhance performance for applications like traffic signs designed for autonomous systems. 

Alex McFarland is an AI journalist and writer exploring the latest developments in artificial intelligence. He has collaborated with numerous AI startups and publications worldwide.