Organization of machine learning based product development as per ISO 26262 and ISO/PAS 21448

TL;DR

This paper proposes a structured approach to organizing the development of machine learning-based systems in automotive safety standards, aligning with ISO 26262 and ISO/PAS 21448 to address ML-specific challenges.

Contribution

It introduces a framework for integrating ML development aspects into ISO 26262 processes, considering ML-specific properties and safety requirements.

Findings

Organizes ML development processes within ISO 26262 phases.

Aligns ML safety considerations with ISO/PAS 21448 for edge case handling.

Provides guidance for documentation and assessment of ML safety properties.

Abstract

Machine learning (ML) algorithms generate a continuous stream of success stories from various domains and enable many novel applications in safety-critical systems. With the advent of autonomous driving, ML algorithms are being used in the automotive domain, where the applicable functional safety standard is ISO 26262. However, requirements and recommendations provided by ISO 26262 do not cover specific properties of machine learning algorithms. Therefore, specific aspects of ML (e.g., dataset requirements, performance evaluation metrics, lack of interpretability) must be addressed within some work products, which collect documentation resulting from one or more associated requirements and recommendations of ISO 26262. In this paper, we propose how key technical aspects and supporting processes related to development of ML-based systems can be organized according to ISO 26262 phases, sub-phases, and work products. We follow the same approach as in the ISO/PAS 21448 standard, which complements ISO 26262, in order to account for edge cases that can lead to hazards not directly caused by system failure.%, but resulting from functional insufficiencies of the intended functionality or by reasonably foreseeable misuse by persons.