With the Traffic Virtualizer you can analyze and virtualize real-world traffic recordings for simulation-based verification and validation of ADAS/AD.
What is the Traffic Virtualizer?
The Traffic Virtualizer is a tool for automated generation of simulation scenarios and extraction of ODD related knowledge from recorded traffic data for verification and validation of ADAS/AD. It automatically processes vehicle trajectories and object-list data.
Providing a safety rating for autonomous vehicles is very time-consuming and complex due to the almost limitless number of possible traffic scenarios. With the Traffic Virtualizer you can transfer recordings of real test drives into the simulation and thus carry out thousands of tests of safety-critical and realistic driving scenarios with special hardware and software comfortably as a simulation, while at the same time collecting knowledge about your ODD.
Application Areas for Traffic Virtualizer
The Traffic Virtualizer is used in verification and validation of advanced driver assistance functions and autonomous vehicles, as well as in data driven development of automated driving functions to close the gap between real-world data and simulation.
Elevate your Test and Validation Activities with Real-World Scenarios
- Exact and consistent replica of real-world scenarios in simulation
- Create thousands of critical variations from just one recorded real-world scenario
- Real-world scenarios for physics-based sensor simulation in AURELION
- OpenSCENARIO and OpenDRIVE
Understand and Quantify the Challenges of your ODD
- Detect and categorize traffic situations in recorded data according to abstract scenarios
- Extraction of scenario parameters
- Collect suitable data for coverage- and residual-risk evaluations
Explore our Complete Range of Solutions
- Road Generation from recordings: We provide a service to create the OpenDRIVE road model based on the recorded sensor data.
- Digital Twin 3D Environments for AURELION: A realistic reproduction of the sensor view in physics-based sensor simulation requires the generation of complete digital twins including material properties.