In the End-to-End Era, Is Scenario-Based Safety Evaluation for Autonomous Driving Still Valid?

Abstract: End-to-end architectures are moving from research papers to mass production, yet the cornerstone of global autonomous driving safety evaluation — scenario-based development and testing — still rests on the assumption that systems can be decomposed into perception, planning, and control modules. This article systematically analyzes five structural challenges that scenario methods face in the end-to-end era, argues that they remain valid but are no longer sufficient, and proposes an evolutionary path that supplements scenario methods with large-scale aerial naturalistic driving data within a three-layer collaborative framework. ...

April 6, 2026 · 14 min · 2910 words · 张玉新 Yuxin Zhang · 0

Value and Challenges of Japan's SAKURA Automated Driving Safety Evaluation Framework V4.0

Abstract: In March 2026, JAMA released the fourth edition (Ver.4.0) of the SAKURA Automated Driving Safety Evaluation Framework — a 344-page national-level safety evaluation technical document. This article systematically examines this safety evaluation system jointly developed by Toyota, Honda, Nissan, and other major Japanese automakers, covering its corporate value, engineering perspectives, core methodology, and the frontier challenges posed by the end-to-end AI era, while exploring its implications for China’s standardization efforts. ...

April 6, 2026 · 11 min · 2329 words · 张玉新 Yuxin Zhang · 0