Expert-LaSTS: Expert-Knowledge Guided Latent Space for Traffic Scenarios
This work addresses scenario-based testing for autonomous vehicles, offering an incremental improvement with automated mining and expert integration.
The paper tackled the problem of clustering traffic scenarios and detecting novel types for autonomous vehicle testing by developing an expert-knowledge guided latent space representation, which showed performance advantages over baseline methods.
Clustering traffic scenarios and detecting novel scenario types are required for scenario-based testing of autonomous vehicles. These tasks benefit from either good similarity measures or good representations for the traffic scenarios. In this work, an expert-knowledge aided representation learning for traffic scenarios is presented. The latent space so formed is used for successful clustering and novel scenario type detection. Expert-knowledge is used to define objectives that the latent representations of traffic scenarios shall fulfill. It is presented, how the network architecture and loss is designed from these objectives, thereby incorporating expert-knowledge. An automatic mining strategy for traffic scenarios is presented, such that no manual labeling is required. Results show the performance advantage compared to baseline methods. Additionally, extensive analysis of the latent space is performed.