Active Learning of Driving Scenario Trajectories
This work addresses the cost and error-prone nature of annotating driving trajectories for autonomous vehicle validation, offering a more efficient method, though it is incremental as it builds on existing active learning techniques.
The study tackled the problem of annotating driving scenario trajectories for autonomous vehicle verification by developing a generic active learning framework that embeds trajectories into a latent space, enabling task-agnostic annotation and detection of unknown classes. The framework was evaluated on real-world datasets from Volvo Cars, showing effectiveness in labeling and detecting unknown trajectory types, with embedding quality identified as a key factor.
Annotated driving scenario trajectories are crucial for verification and validation of autonomous vehicles. However, annotation of such trajectories based only on explicit rules (i.e. knowledge-based methods) may be prone to errors, such as false positive/negative classification of scenarios that lie on the border of two scenario classes, missing unknown scenario classes, or even failing to detect anomalies. On the other hand, verification of labels by annotators is not cost-efficient. For this purpose, active learning (AL) could potentially improve the annotation procedure by including an annotator/expert in an efficient way. In this study, we develop a generic active learning framework to annotate driving trajectory time series data. We first compute an embedding of the trajectories into a latent space in order to extract the temporal nature of the data. Given such an embedding, the framework becomes task agnostic since active learning can be performed using any classification method and any query strategy, regardless of the structure of the original time series data. Furthermore, we utilize our active learning framework to discover unknown driving scenario trajectories. This will ensure that previously unknown trajectory types can be effectively detected and included in the labeled dataset. We evaluate our proposed framework in different settings on novel real-world datasets consisting of driving trajectories collected by Volvo Cars Corporation. We observe that active learning constitutes an effective tool for labelling driving trajectories as well as for detecting unknown classes. Expectedly, the quality of the embedding plays an important role in the success of the proposed framework.