Autonomous Driving System Configuration Defect Detection Method Based on Fuzzing

Jinzhao Liu, Xiao Yu*, Li Zhang, Yanqiu Zhang, Yuanzhang Li, Kun Tan, Yuan Tan

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The configuration item values of each component within the autonomous driving system (ADS) are mutually constrained and complexly associated, which leads to problems such as configuration logic faults and memory overflow easily occurring during the configuration process, thus triggering the frequent occurrence of system configuration faults. This paper proposes a dynamic fuzzing method for configuration parameters based on static analysis of source code, through mapping and extracting configuration constraints within the system, realizing dynamic fuzzing for configuration parameters based on source code characterization, and completing the diagnosis of configuration defects and assistive repair functions. The experimental results show that the fuzzing method is effective in detecting and localizing configuration parameter defects under a variety of ADS anomalies, which not only dramatically improves the coverage rate of system configuration defect detection, but also achieves a configuration defect detection rate of 89.9%.

Original languageEnglish
Title of host publicationDigital Forensics and Cyber Crime - 15th EAI International Conference, ICDF2C 2024, Proceedings
EditorsSanjay Goel, Ersin Uzun, Mengjun Xie, Sumantra Sarkar
PublisherSpringer Science and Business Media Deutschland GmbH
Pages296-312
Number of pages17
ISBN (Print)9783031893599
DOIs
Publication statusPublished - 2025
Externally publishedYes
Event15th EAI International Conference on Digital Forensics and Cyber Crime, ICDF2C 2024 - Dubrovnik, Croatia
Duration: 9 Oct 202410 Oct 2024

Publication series

NameLecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST
Volume614 LNICST
ISSN (Print)1867-8211
ISSN (Electronic)1867-822X

Conference

Conference15th EAI International Conference on Digital Forensics and Cyber Crime, ICDF2C 2024
Country/TerritoryCroatia
CityDubrovnik
Period9/10/2410/10/24

Keywords

  • Autonomous Driving
  • Configuration Defects
  • Fuzzing
  • Static Analysis

Fingerprint

Dive into the research topics of 'Autonomous Driving System Configuration Defect Detection Method Based on Fuzzing'. Together they form a unique fingerprint.

Cite this