Automatic Traffic Scenario Conversion from OpenSCENARIO to CommonRoad

TL;DR

This work addresses the interoperability gap between OpenSCENARIO and CommonRoad by introducing an open-source converter that converts OpenSCENARIO dynamics into CommonRoad representations. The method uses OpenDRIVE-to-lanelet conversion, a simulation core to realize dynamic elements, and then constructs a full CommonRoad scenario with planning problems, outputting an XML file for evaluation. The authors demonstrate the tool on dozens of public OpenSCENARIO cases, validating conversions with CommonRoad tools for collision checking, motion planning, criticality assessment, safety verification, and trajectory repair. The converter, paired with existing CommonRoad tooling, enables robust benchmarking and exchange of scenarios across formats, promoting broader collaboration between research and industry.

Abstract

Scenarios are a crucial element for developing, testing, and verifying autonomous driving systems. However, open-source scenarios are often formulated using different terminologies. This limits their usage across different applications as many scenario representation formats are not directly compatible with each other. To address this problem, we present the first open-source converter from the OpenSCENARIO format to the CommonRoad format, which are two of the most popular scenario formats used in autonomous driving. Our converter employs a simulation tool to execute the dynamic elements defined by OpenSCENARIO. The converter is available at commonroad.in.tum.de and we demonstrate its usefulness by converting publicly available scenarios in the OpenSCENARIO format and evaluating them using CommonRoad tools.

Figures (4)

• Figure 1: UML class diagram of the OpenSCENARIO format. For brevity, we omit the nonessential classes such as those related to parameters.
• Figure 2: Exemplary overtaking scenario.
• Figure 3: UML class diagram of the CommonRoad format. Details of child elements are omitted for clarity.
• Figure 4: Evaluation results with CommonRoad tools. We only display the scenario information between $2.6s$ and $5.6s$. (a) shows the configuration of the converted CommonRoad scenario, with snapshots captured from the inside view of the ego vehicle during esmini simulation at three time steps. Collision checking and motion planning results are presented in (b) and (c), respectively. To provide clear insights, the criticality of the scenario is plotted on the vertical axis of the graph, with an upward trend indicating increasing criticality, as shown in (d). Finally, (e) displays the safety verification results at both $2.6s$ and $3.8s$, where the trajectory is repaired if the intended trajectory is not legally safe.