Automated Driving Systems must safely share the road with many different users - vehicles ranging from motorcycles to cars, trucks, and buses, pedestrians, cyclists, and even animals. Training and testing these Systems in simulated environments requires exposing them to virtual traffic that is both highly realistic, and highly scalable. Our proprietary traffic model technology uses a variety of machine learning techniques to understand the behavior of real road users and create geo-specific AI agents that interact with the environment, the vehicle under test, and one another, just as real road users would. Deploying configurable, physically-accurate, large-scale AI-based traffic has never been easier!
Cognata generates automotive training and validation data for machine learning and deep neural networks (DNN) to complement existing data, bridge gaps, and provide non-biased, accurate, diverse, and realistic datasets on demand.