FuzzSense: Towards A Modular Fuzzing Framework for Autonomous Driving Software

2025-04-14

AI Summary

To address the isolation and lack of coordination among fuzzing tools targeting heterogeneous environment components—such as scenarios, sensors, and vehicle dynamics—in autonomous driving (AD) software testing, this paper proposes the first modular black-box fuzzing framework tailored for next-generation AD platforms like Autoware.Universe. The framework adopts a plug-and-play integration architecture, enabling dynamic orchestration of heterogeneous fuzzers and cross-component concurrent fuzzing. It introduces the first LiDAR-specific fuzzer and features deep integration with the AWSIM simulation platform. Experimental evaluation on Autoware.Universe successfully uncovered multiple semantic-level control vulnerabilities, demonstrating the superiority of coordinated fuzzing in detecting deep logical flaws. The framework is open-sourced, establishing a standardized, extensible technical foundation for safety-critical testing of AD systems.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Application Domains: Mobility, Driving & Flight

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Fuzz testing to find semantic control vulnerabilities is an essential activity to evaluate the robustness of autonomous driving (AD) software. Whilst there is a preponderance of disparate fuzzing tools that target different parts of the test environment, such as the scenario, sensors, and vehicle dynamics, there is a lack of fuzzing strategies that ensemble these fuzzers to enable concurrent fuzzing, utilizing diverse techniques and targets. This research proposes FuzzSense, a modular, black-box, mutation-based fuzzing framework that is architected to ensemble diverse AD fuzzing tools. To validate the utility of FuzzSense, a LiDAR sensor fuzzer was developed as a plug-in, and the fuzzer was implemented in the new AD simulation platform AWSIM and Autoware.Universe AD software platform. The results demonstrated that FuzzSense was able to find vulnerabilities in the new Autoware.Universe software. We contribute to FuzzSense open-source with the aim of initiating a conversation in the community on the design of AD-specific fuzzers and the establishment of a community fuzzing framework to better target the diverse technology base of autonomous vehicles.

Problem

• Lack of integrated fuzzing tools for autonomous driving software
• Need for concurrent fuzzing across diverse AD components
• Absence of community-driven AD-specific fuzzing framework

Innovation

• Modular black-box mutation-based fuzzing framework
• Ensembles diverse AD fuzzing tools concurrently
• LiDAR sensor fuzzer plugin for AD platforms