A Comprehensive Study of Bug-Fix Patterns in Autonomous Driving Systems

2025-02-04

AI Summary

Software defects in autonomous driving systems (ADS) pose significant challenges due to their complex, safety-critical nature, yet systematic characterization of their properties, repair difficulties, and debugging obstacles remains lacking. Method: We conduct an empirical study of 1,331 real-world defect fixes from Apollo and Autoware, jointly analyzing commit histories and issue reports via qualitative coding, statistical analysis, and pattern induction. Contribution/Results: We propose (1) the first hierarchical ADS defect model; (2) the first dual-track classification framework integrating 15 syntactic and 27 semantic repair patterns; and (3) the first large-scale ADS defect repair benchmark dataset. Our analysis identifies path planning, data flow, and configuration management as high-frequency defect domains and demonstrates statistically significant heterogeneity in repair pattern distribution—providing reusable pattern guidelines and an evaluation foundation for ADS defect detection, localization, and automated repair.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Application Domains: Mobility, Driving & Flight
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search

Abstract

As autonomous driving systems (ADSes) become increasingly complex and integral to daily life, the importance of understanding the nature and mitigation of software bugs in these systems has grown correspondingly. Addressing the challenges of software maintenance in autonomous driving systems (e.g., handling real-time system decisions and ensuring safety-critical reliability) is crucial due to the unique combination of real-time decision-making requirements and the high stakes of operational failures in ADSes. The potential of automated tools in this domain is promising, yet there remains a gap in our comprehension of the challenges faced and the strategies employed during manual debugging and repair of such systems. In this paper, we present an empirical study that investigates bug-fix patterns in ADSes, with the aim of improving reliability and safety. We have analyzed the commit histories and bug reports of two major autonomous driving projects, Apollo and Autoware, from 1,331 bug fixes with the study of bug symptoms, root causes, and bug-fix patterns. Our study reveals several dominant bug-fix patterns, including those related to path planning, data flow, and configuration management. Additionally, we find that the frequency distribution of bug-fix patterns varies significantly depending on their nature and types and that certain categories of bugs are recurrent and more challenging to exterminate. Based on our findings, we propose a hierarchy of ADS bugs and two taxonomies of 15 syntactic bug-fix patterns and 27 semantic bug-fix patterns that offer guidance for bug identification and resolution. We also contribute a benchmark of 1,331 ADS bug-fix instances.

Problem

• Analyzes bug-fix patterns in autonomous driving systems
• Improves reliability and safety of autonomous driving systems
• Proposes taxonomies for bug identification and resolution

Innovation

• Empirical study of bug-fix patterns
• Hierarchy and taxonomies for bug resolution
• Benchmark of 1,331 ADS bug-fix instances