Transferring the Complexity of the Real World to Simulation
dSPACE and its group company understand.ai, which specializes in automated, AI-based data annotation, offer a new service that generates simulation scenarios from recorded measurement data used to validate functions for autonomous driving and driver assistance. With this offer, dSPACE supports its customers in the automotive industry in developing autonomous vehicles quickly and efficiently using realistic simulation.
To reliably validate autonomous or semi-autonomous vehicles, thousands of near-realistic scenarios are required — including rare events. Manually creating these rare events in special editors is extremely time-consuming. “With the Scenario Generation Service, we bring the complexity of the real world into a simulation and enable validation with thousands of relevant and critical simulation scenarios,” says product manager Thorsten Püschl, summarizing the advantages of the new service.
The Scenario Generation Service from understand.ai and dSPACE uses existing sets of data recorded during measurement runs. In a highly-automated process, AI-based annotation solutions from understand.ai extract the relevant information from the raw data of the vehicle sensors. This creates realistic and consistent simulation scenarios. Optionally, data from object lists can also be used for scenario generation.
The generated scenarios are used to create exact reproductions of real driving situations in the simulation that help simulate events from test drives in the laboratory or compare simulations of sensor models with measurement data for sensor model validation. Generating logical simulation scenarios helps create many new, previously unknown corner cases as simulations via scenario-based testing, thus enabling the testing of driving functions with a large number of relevant and critical situations. The Scenario Generation Service can also create the road model required for the simulation based on the sensor data. Alternatively, HD maps can also be used.
Additionally, detailed 3D models of the vehicle environment can be generated for physical sensor simulation.
The generated scenarios can be immediately used in the dSPACE ASM simulation environment as well as the existing dSPACE infrastructure for powerful software-in-the-loop (SIL) and hardware-in-the-loop (HIL) testing across multiple developmental stages. Additionally, the scenarios are provided in OpenSCENARIO and OpenDRIVE, so that they can be transferred to other simulators.