3D Game Tech for Self-driving
Hyundai Mobis will introduce three-dimensional (3D) game video and deep-running technology to improve the accuracy of an autonomous navigation sensor under development.
Hyundai Mobis announced on September 16 that the company started to develop video technology for autonomous driving simulations based on high-definition software for 3D game development. The technique is to conduct autonomous driving tests in various 3D virtual environments used in computer games.
This technology can enhance camera performance as it tests and implements a desired environment regardless of various constraints that may appear in the real world. Using high-definition 3D videos, the technology can produce various driving scenarios such as roads at rainy nights, complex urban areas, puddles and road construction sites. Hyundai Mobis explained that the technology could greatly improve the object recognition accuracy of a camera for autonomous vehicles under development.
The Hyundai Mobis India Institute has recently signed a contract with Tata Electronics, an Indian software company that provides solutions in the areas of artificial intelligence (AI), Internet of Things (IoT), big data and autonomous navigation.
Hyundai Mobis established a research institute in Hyderabad, which is called “Silicon Valley in India,” in 2007. The lab employs a large number of local talent to develop and verify the driving assistant system (DAS), autonomous navigation systems, and multimedia software among others.
"We will complete the development of the technology for autonomous navigation simulations by the end of next year," said Yang Seung-wook, director of the Hyundai Mobis ICT Research Center and vice president of Hyundai Mobis. “We plan to make use of specialized companies at home and abroad in order to develop core technology for future cars."
Hyundai Mobis is also developing automatic driving video classification technology using deep learning technology to improve the recognition performance of autonomous driving car cameras by the end of the first half of next year. The front camera on an autonomous vehicle captures a number of objects such as other vehicles, lanes, pedestrians and traffic lights like human eyes.