SCOUT: A Lightweight Framework for Scenario Coverage Assessment in Autonomous Driving
This work addresses the need for efficient and scalable scenario coverage assessment in autonomous driving, offering a practical alternative to expensive human annotations or large vision-language models, though it is incremental as it builds on existing distillation techniques.
The paper tackles the problem of assessing scenario coverage for autonomous driving agents by proposing SCOUT, a lightweight surrogate model that predicts coverage labels from sensor representations, achieving high accuracy while significantly reducing computational costs compared to existing methods.
Assessing scenario coverage is crucial for evaluating the robustness of autonomous agents, yet existing methods rely on expensive human annotations or computationally intensive Large Vision-Language Models (LVLMs). These approaches are impractical for large-scale deployment due to cost and efficiency constraints. To address these shortcomings, we propose SCOUT (Scenario Coverage Oversight and Understanding Tool), a lightweight surrogate model designed to predict scenario coverage labels directly from an agent's latent sensor representations. SCOUT is trained through a distillation process, learning to approximate LVLM-generated coverage labels while eliminating the need for continuous LVLM inference or human annotation. By leveraging precomputed perception features, SCOUT avoids redundant computations and enables fast, scalable scenario coverage estimation. We evaluate our method across a large dataset of real-life autonomous navigation scenarios, demonstrating that it maintains high accuracy while significantly reducing computational cost. Our results show that SCOUT provides an effective and practical alternative for large-scale coverage analysis. While its performance depends on the quality of LVLM-generated training labels, SCOUT represents a major step toward efficient scenario coverage oversight in autonomous systems.