A replica of the world would be incomplete without simulating naturally moving traffic. For the AAI intelligent traffic simulation module, we are training individual traffic agents with independent decision-making capabilities and thus non-deterministic individual driving behavior - just like we as humans all have our own driving style. Mixed in the overall traffic flow, the agents help to generate natural looking traffic.
It is a complicated journey for our traffic agents to learn human-like behavior. Like student drivers, our driver models first have to learn the basics of driving and applicable traffic rules. But the journey does not end here, since in the real world you can observe much more complex driving styles than the model behavior taught in driving schools. Many unwritten rules and local habits exist and people often need to negotiate with each other in order to solve complicated situations. Therefore just like ourselves, the AAI traffic agents never stop learning and are being continuously trained using state of the art machine learning algorithms and real-world traffic data teaching cultural and regional values.
Starting with reinforcement learning, traffic agents are taught to drive by being rewarded or penalized for wanted or unwanted behavior respectively. Once the driver model has developed solid driving skills, the next level of learning starts with feeding the training pipeline with local naturalistic traffic data - introducing it to the full array of human driving behavior.
Based on our training pipeline and availability of traffic recordings from the real world, it is possible to generate distinguishable and localized driver profiles fitting simulations for all required countries and areas. Contact us to learn more about our solution!
The benefit of simulated endurance testing can only be fully exploited with naturally behaving traffic surrounding the ego vehicle.
Our driver profiles contain a range of agent types such as rather conservative and defensive characters, which prefer toabide by the traffic rules and travel at steady speeds, or very dynamic and even aggressive characters, which have a tendency to occasionally break the law and make their way with more extreme acceleration and speeds.
All incidents are integrally documented and re-playable for analysis within AAI ReplicaR®. With the AAI Scenario Cloning software, interesting maneuvers from simulated Endurance Tests can be transferred into deterministic scenario templates and stored in the scenario database for repeated testing.
The AAI Intelligent Traffic is an independent and highly compatible software module, which can be integrated into various simulation environments. Interfaces supported include Functional Mock-up Unit (FMU/FMI) , middleware software and more. Let us know about the setup of your simulation environment - we will support the integration of our traffic module.