Grasping Causality for the Explanation of Criticality for Automated Driving

TL;DR

This paper formalizes causal reasoning for understanding and reducing critical scenarios in automated driving, addressing limitations of purely statistical approaches by integrating causal knowledge based on Judea Pearl's theory.

Contribution

It introduces a formal framework for causal queries in automated driving safety, enabling the modeling of influencing factors and their effects on criticality, with evaluation methods and data requirements.

Findings

Formal causal framework for automated driving safety

Evaluation of causal modeling quality on a small example

Discussion on data requirements for causal inference

Abstract

The verification and validation of automated driving systems at SAE levels 4 and 5 is a multi-faceted challenge for which classical statistical considerations become infeasible. For this, contemporary approaches suggest a decomposition into scenario classes combined with statistical analysis thereof regarding the emergence of criticality. Unfortunately, these associational approaches may yield spurious inferences, or worse, fail to recognize the causalities leading to critical scenarios, which are, in turn, prerequisite for the development and safeguarding of automated driving systems. As to incorporate causal knowledge within these processes, this work introduces a formalization of causal queries whose answers facilitate a causal understanding of safety-relevant influencing factors for automated driving. This formalized causal knowledge can be used to specify and implement abstract safety principles that provably reduce the criticality associated with these influencing factors. Based on Judea Pearl's causal theory, we define a causal relation as a causal structure together with a context, both related to a domain ontology, where the focus lies on modeling the effect of such influencing factors on criticality as measured by a suitable metric. As to assess modeling quality, we suggest various quantities and evaluate them on a small example. As availability and quality of data are imperative for validly estimating answers to the causal queries, we also discuss requirements on real-world and synthetic data acquisition. We thereby contribute to establishing causal considerations at the heart of the safety processes that are urgently needed as to ensure the safe operation of automated driving systems.