Dual AI models improve testing for safer autonomous driving simulations
Teach-to-Crash: A Closed-Loop Student-Teacher LLM Framework for Collision-Inducing Test Scenario Generation
Software EngineeringArtificial IntelligenceMultiagent Systems
Summary
Testing self-driving cars in simulation is hard because dangerous crashes are rare but important to find. The authors created a system called Teach-to-Crash that uses two AI language models working together: one guides the search for crash scenarios, and the other creates detailed driving scenarios. This teamwork leads to more frequent, varied, and useful crash examples. Their tests showed it finds crashes faster and more reliably than other methods.
What this means in practice
- •For automotive testing teams: Generate diverse and frequent collision scenarios to improve validation and safety analysis of self-driving systems before real-world deployment.
- •For simulation software developers: Integrate adaptive dual-AI control to better steer scenario generation towards critical failures within driving simulators.
Authors
Zaid Ghazal, Khouloud Gaaloul, Bruce Maxim
Abstract
Validating Autonomous Driving Systems (ADS) in simulation requires testing architectures that can discover rare, safety-critical failures while generating scenarios that are executable, diverse, and useful for downstream failure analysis. We introduce Teach-to-Crash, a closed-loop testing framework that combines a constrained ego-centric scenario representation, stagnation-aware search control, and a dual-LLM architecture for adaptive failure discovery. A high-reasoning Teacher LLM acts as an adaptive search controller, while a low-reasoning Student LLM emits simulator-executable scenarios in a strict JSON schema. The Teacher intervenes only when rolling collision rate and time-to-collision metrics stagnate, providing strategic guidance to redirect the search. In a CARLA case study with two experimental setups that vary the ego vehicle's speed policy, Teach-to-Crash achieves the highest Collision Hit Rate (90.79%), the shortest mean Time-to-Collision (18.31 s), and a competitive Collision Discovery Rate (136.21). PAFOT attains a higher mean CDR (179.44), but with substantially larger variance. Teach-to-Crash also yields the highest diversity (0.547) and, averaged across both setups on the CARLA Traffic Manager controller, the highest avoidability-based usefulness proxy (60.04%) among the compared methods. These results, within the evaluated CARLA scope, provide evidence that closed-loop dual-LLM reasoning can steer adversarial simulation-based testing over a constrained executable program space, generating failures that are frequent, structurally diverse, and assessed as more frequently avoidable.