TY - GEN
T1 - Expert-LaSTS
T2 - 2022 IEEE Intelligent Vehicles Symposium, IV 2022
AU - Wurst, Jonas
AU - Balasubramanian, Lakshman
AU - Botsch, Michael
AU - Utschick, Wolfgang
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - 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.
AB - 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.
KW - Clustering
KW - Deep Learning
KW - Novelty Detection
KW - Scenario-Based Testing
UR - http://www.scopus.com/inward/record.url?scp=85132288201&partnerID=8YFLogxK
U2 - 10.1109/IV51971.2022.9827187
DO - 10.1109/IV51971.2022.9827187
M3 - Conference contribution
AN - SCOPUS:85132288201
T3 - IEEE Intelligent Vehicles Symposium, Proceedings
SP - 484
EP - 491
BT - 2022 IEEE Intelligent Vehicles Symposium, IV 2022
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 5 June 2022 through 9 June 2022
ER -