Artificial Intelligence Solutions for Transportation

Artificial Intelligence Solutions for Transportation

This piece was originally published in the January 2018 issue of electroindustry.

Caption: Police can track criminals using technology that recognizes faces and license plates. Photo by Shuyu Zheng

Shuyu Zheng, Research Intern, NEMA

The latest applications of artificial intelligence (AI) technology in the fields of defense intelligence, medical imaging, smart cities, finance, and cybersecurity were on display in Washington, D.C., in November. Hosted by NVIDIA, an American technology company that designs graphics processing units (GPUs) for gaming, cryptocurrency, and other markets, NVIDIA’s GPU Technology Conference attracted more than 1,500 attendees from government, Congress, the military, research institutes, and the private sector.

One platform for accelerating the adoption of autonomous driving combines deep learning, sensor fusion, and surround vision technologies to capture real-time surroundings, precisely locate a vehicle on a highly detailed map, and guide the vehicle on a safe path. The platform captures the data from multiple camera, lidar, radar, and ultrasonic sensors and then uses algorithms to understand and present the environment around the vehicles, including static and dynamic objectives. Deep neural networks that detect and classify objects can make the fused sensor data even more accurate. The platform is designed to support ASIL-D, the highest level of automotive functional safety as defined by ISO 26262 Functional Safety for Road Vehicles.

Other systems promote safety and smart cities. Police can use the technology to recognize faces and license plates to identify criminals on the loose. The platform in the car, coupled with GPUs in the data center, can create highly detailed maps and keep them updated for automakers and map companies. The platform can also send back traffic condition data to help alleviate congestion. Cooperating with Huawei Technologies, NVIDIA used this technology to ease the congestion in Shenzhen, China, by 30 percent.

With its futuristic possibilities, AI technology creates abundant opportunities for electrical manufacturers while setting higher requirements for electrical equipment and devices.


Leave a Reply

Your email address will not be published. Required fields are marked *