What Did I Learn? Operational Competence Assessment for AI-Based Trajectory Planners
This addresses the need for trustworthy AI deployment in automated driving by improving safety and explainability, though it is incremental as it builds on existing knowledge graph and dataset analysis techniques.
The paper tackles the problem of assessing the operational competence of AI-based trajectory planners in automated driving by developing a method to identify situations not sufficiently covered in training data, applied to the NuPlan dataset to monitor model competence.
Automated driving functions increasingly rely on machine learning for tasks like perception and trajectory planning, requiring large, relevant datasets. The performance of these algorithms depends on how closely the training data matches the task. To ensure reliable functioning, it is crucial to know what is included in the dataset to assess the trained model's operational risk. We aim to enhance the safe use of machine learning in automated driving by developing a method to recognize situations that an automated vehicle has not been sufficiently trained on. This method also improves explainability by describing the dataset at a human-understandable level. We propose modeling driving data as knowledge graphs, representing driving scenes with entities and their relationships. These graphs are queried for specific sub-scene configurations to check their occurrence in the dataset. We estimate a vehicle's competence in a driving scene by considering the coverage and complexity of sub-scene configurations in the training set. Higher complexity scenes require greater coverage for high competence. We apply this method to the NuPlan dataset, modeling it with knowledge graphs and analyzing the coverage of specific driving scenes. This approach helps monitor the competence of machine learning models trained on the dataset, which is essential for trustworthy AI to be deployed in automated driving.