317. Evaluation V-Suite introduction

The Evaluation V-Suite scenarios are used for passive analysis of simulation logs or recorded driving data. These scenarios defines patterns, behaviors, or conditions to look for in the data to support automated scenario matching, KPI extraction, and coverage tracking for scenario-specific insights. These scenarios prescribe test content and actor behavior.

317.1 Key features of Evaluation scenarios

• Passive data analysis

  Designed for post-simulation or drive log analysis, not for controlling tests. Scenarios detect specific behaviors, events, or conditions in recorded data.

• Structured, abstract scenario library

  Built with OpenSCENARIO DSL, the library includes configurable templates focused on behavioral patterns (e.g., cut-in, junction approach) rather than fixed events.

• Customizable and extensible

  Users can tune scenario parameters or add new scenarios to meet project or domain-specific needs. The library is continuously updated and refined.

• Detection rules and metrics

  Each scenario defines entry and exit criteria, and detection logic, and generates outputs such as scenario intervals, KPIs, and coverage metrics for in-depth analysis.

• Independent from Generation V-Suite

  Evaluation scenarios are optimized for passive analysis, while Generation scenarios are used to drive simulations or tests.

• Core of the Evaluation Pipeline

  Powers automated detection, coverage tracking, and reporting in the Foretify evaluation workflow.