Cognitive, Hyundai Mobis develop vision software to detect traffic objects
Cognitive Pilot and Hyundai Mobis announced the completion of a software module for active safety and autonomous driving. The computer vision-based software module can reportedly recognize various classes of objects moving along the roads such as different types of cars, buses, motorcyclists, cyclists, and pedestrians.
The core technologies used in the development of the module are deep-learning neural networks that provide the system with the required recognition accuracy. The module will be integrated in the control systems of luxury vehicles.
Olga Uskova, President of Cognitive Technologies, said, “The main goal of our joint work was to create one of most reliable and safe autonomous driving systems in the world. We have successfully solved the most difficult task of an accurate detection of traffic participants: pedestrians, cyclists, motorcyclists, and other objects that don't have a constant shape. The system works in almost any conditions: rain, snow, fog, bright sun, night time, etc. In the nearest future, the company expects to present more exclusive agreements with the world's leading car manufacturers and OEMs. Those are the developments in the field of creating ADS for ground transportation and 4D imaging radar sensors.”
Cognitive Pilot's multi-platform C-Pilot system can be installed on trains, urban electric trains, and agricultural machinery.