TY - JOUR
T1 - CausalAF
T2 - 6th Conference on Robot Learning, CoRL 2022
AU - Ding, Wenhao
AU - Lin, Haohong
AU - Li, Bo
AU - Zhao, Ding
N1 - Funding Information:
We gratefully acknowledge support from the National Science Foundation under grant CAREER CNS-2047454 and support from the Manufacturing Futures Initiative at Carnegie Mellon University made possible by the Richard King Mellon Foundation.
Publisher Copyright:
© 2023 Proceedings of Machine Learning Research. All rights reserved.
PY - 2023
Y1 - 2023
N2 - Generating safety-critical scenarios, which are crucial yet difficult to collect, provides an effective way to evaluate the robustness of autonomous driving systems. However, the diversity of scenarios and efficiency of generation methods are heavily restricted by the rareness and structure of safety-critical scenarios. Therefore, existing generative models that only estimate distributions from observational data are not satisfying to solve this problem. In this paper, we integrate causality as a prior into the scenario generation and propose a flow-based generative framework, Causal Autoregressive Flow (CausalAF). CausalAF encourages the generative model to uncover and follow the causal relationship among generated objects via novel causal masking operations instead of searching the sample only from observational data. By learning the cause-and-effect mechanism of how the generated scenario causes risk situations rather than just learning correlations from data, CausalAF significantly improves learning efficiency. Extensive experiments on three heterogeneous traffic scenarios illustrate that CausalAF requires much fewer optimization resources to effectively generate safety-critical scenarios. We also show that using generated scenarios as additional training samples empirically improves the robustness of autonomous driving algorithms.
AB - Generating safety-critical scenarios, which are crucial yet difficult to collect, provides an effective way to evaluate the robustness of autonomous driving systems. However, the diversity of scenarios and efficiency of generation methods are heavily restricted by the rareness and structure of safety-critical scenarios. Therefore, existing generative models that only estimate distributions from observational data are not satisfying to solve this problem. In this paper, we integrate causality as a prior into the scenario generation and propose a flow-based generative framework, Causal Autoregressive Flow (CausalAF). CausalAF encourages the generative model to uncover and follow the causal relationship among generated objects via novel causal masking operations instead of searching the sample only from observational data. By learning the cause-and-effect mechanism of how the generated scenario causes risk situations rather than just learning correlations from data, CausalAF significantly improves learning efficiency. Extensive experiments on three heterogeneous traffic scenarios illustrate that CausalAF requires much fewer optimization resources to effectively generate safety-critical scenarios. We also show that using generated scenarios as additional training samples empirically improves the robustness of autonomous driving algorithms.
KW - Autonomous Driving
KW - Causal Generative Models
KW - Scenario Generation
UR - http://www.scopus.com/inward/record.url?scp=85164933190&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85164933190&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:85164933190
SN - 2640-3498
VL - 205
SP - 812
EP - 823
JO - Proceedings of Machine Learning Research
JF - Proceedings of Machine Learning Research
Y2 - 14 December 2022 through 18 December 2022
ER -