Skip to content
Book Open access

MoRE: Traffic Rule Modeling and Runtime Evaluation for Model-Based Testing of Automated Driving Systems in CARLA

Oct 2026 · Proceedings of the ACM/IEEE 29th International Conference on Model Driven Engineering Languages and Systems · 0 citations · 14 references

Abstract

Testing autonomous driving systems (ADSs) requires evaluating not only whether an ADS completes a scenario, but also whether its behavior satisfies traffic rules throughout the simulation. Model-based testing (MBT) can support repeatable and traceable ADS testing, but it still needs executable test oracles that connect traffic rules with observable simulation facts. This paper presents MoRE, a tool for modeling and runtime checking of traffic rules in model-based ADS testing using CARLA, an open-source autonomous driving simulator. MoRE has MetaMoRE, a metamodel that represents road semantics, traffic participants, frame-level observations, temporal information, and derived behavioral events. Traffic rules are specified as queries over MetaMoRE instances instead of being hard-coded against low-level simulator APIs. The implemented workflow consists of an adapter that lifts CARLA runtime data into model instances, a recorder that serializes scenario records for offline reuse, and an evaluator that detects behaviors, interprets user-defined rules, and reports their evaluation outcomes with links to concrete simulation frames. The demonstration illustrates how MoRE supports both live and offline traffic-rule checking over CARLA scenarios and provides reusable scenario records for testing ADS behavior. The source code of MoRE is available at https://github.com/WSE-Lab/MoRE-ADS, while a demonstration video can be found at https://www.modelcopilot.org/more.html.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.