Reproduce “A Comprehensive Study of Bug-Fix Patterns in Autonomous Driving Systems” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuntianyi Chen, Yuqi Huai, Yirui He, Shilong Li, Changnam Hong, Qi Alfred Chen, Joshua Garcia
This empirical study provides the first comprehensive analysis of bug-fix patterns in autonomous driving systems by examining 1,331 real-world bug fixes from Baidu Apollo and Autoware. The research introduces a novel hierarchical framework to categorize bug fixes, distinguishing between syntactic and domain-specific semantic patterns, and details common repair strategies for enhancing ADS reliability.
Key insights
- The study introduces a "Modularization Granularity" hierarchical framework for ADS bug analysis, distinguishing between syntactic (code-level) and semantic (algorithmic/logic) bug-fix patterns.
- Method Call and If-related issues are the most frequent syntactic bug-fix patterns, while Path Planning Optimization and Data Flow Correction dominate semantic patterns.
- Algorithm-related root causes are the most diverse, leading to a wide range of semantic bug-fix patterns, and Data Flow Correction is pervasive across various bug symptoms.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.