[Paper Review] Building a Credible Case for Safety: Waymo's Approach for the Determination of Absence of Unreasonable Risk

Francesca Favarò, Laura Fraade-Blanar+7 more|arXiv (Cornell University)|Jun 2, 2023

Safety Systems Engineering in Autonomy15 citations

TL;DR

The paper presents Waymo's blueprint for constructing a credible safety case for fully autonomous driving, detailing a layered, dynamic, and credible safety framework that is methodology-agnostic and useful for any AV developer.

ABSTRACT

This paper presents an overview of Waymo's approach to building a reliable case for safety - a novel and thorough blueprint for use by any company building fully autonomous driving systems. A safety case for fully autonomous operations is a formal way to explain how a company determines that an AV system is safe enough to be deployed on public roads without a human driver, and it includes evidence to support that determination. It involves an explanation of the system, the methodologies used to develop it, the metrics used to validate it and the actual results of validation tests. Yet, in order to develop a worthwhile safety case, it is first important to understand what makes one credible and well crafted, and align on evaluation criteria. This paper helps enabling such alignment by providing foundational thinking into not only how a system is determined to be ready for deployment but also into justifying that the set of acceptance criteria employed in such determination is sufficient and that their evaluation (and associated methods) is credible. The publication is structured around three complementary perspectives on safety that build upon content published by Waymo since 2020: a layered approach to safety; a dynamic approach to safety; and a credible approach to safety. The proposed approach is methodology-agnostic, so that anyone in the space could employ portions or all of it.

Motivation & Objective

• Explain what makes a safety case credible for fully autonomous driving.
• Describe Waymo's three complementary safety perspectives that underpin the safety case.
• Provide a methodology-agnostic framework that others can adopt to demonstrate absence of unreasonable risk.

Proposed method

• Outline a layered approach to safety that combines multiple perspectives.
• Present a dynamic approach to safety that accounts for changing conditions and system evolution.
• Propose a credible approach to safety focusing on justification and evidence quality.
• Offer guidance on aligning evaluation criteria and ensuring credibility of validation methods.

Experimental results

Research questions

• RQ1 How can a safety case for fully autonomous operations be structured to be credible and credible to stakeholders?
• RQ2 What are the complementary safety perspectives that support a credible safety case for autonomous driving?
• RQ3 How can acceptance criteria and validation methods be justified as sufficient and credible for deployment decisions?

Key findings

• Waymo proposes a three-perspective framework (layered, dynamic, credible) to build a safety case.
• The framework is described as methodology-agnostic, enabling broad uptake by different organizations.
• The paper emphasizes alignment on evaluation criteria and the credibility of methodologies used to validate safety.
• A formal safety case explains system details, methodologies, metrics, and validation results to support deployment decisions.