Closed-loop simulation for the OpenADStack automated driving stack with CARLA or SUMO, incl. OpenSCENARIO scenario-based testing
-
Updated
Nov 19, 2026 - Python
Closed-loop simulation for the OpenADStack automated driving stack with CARLA or SUMO, incl. OpenSCENARIO scenario-based testing
OpenSCENARIO 2.1 to CARLA compiler: ANTLR frontend, typed semantics, py_trees behavior trees, deterministic mock adapter and CARLA cut-in validation.
Final thesis
AFALS: how far can the assumptions behind a proven-safe cruise controller be stretched before it crashes? Search-based falsification with CMA-ES, TPE, annealing and random search.
Uncertainty-aware safety validation of a learned ACC / car-following policy: closed-loop simulation, safety shield, and plausibility-constrained scenario search with OpenSCENARIO export.
Critical-event screening and tail fidelity of generated cut-in scenarios, on synthetic data
Parameterizing, clustering and efficiently simulating 2,322 real highway cut-ins from highD, replayed against an ACC + AEB system.
To associate your repository with the scenario-based-testing topic, visit your repo's landing page and select "manage topics."