There is a newer version of the record available.

Published January 15, 2026 | Version v1
Software Open

DeFT: Maintaining Determinism and Extracting Unit Tests for Autonomous Driving Planning

Authors/Creators

  • 1. EDMO icon University of California, Irvine

Description

This repository corresponds to the ICSE 2026 Research Track paper and its accompanying artifact, DeFT, a tool and methodology designed to improve testing reliability in autonomous driving systems by addressing non-determinism in planning tests. Traditional system-level scenario tests often produce varying outcomes, making failure reproduction and debugging challenging. DeFT is a methodology that converts non-deterministic system-level scenario tests into deterministic module-level tests by extracting and reconstructing module inputs.

Files

DeFT-main.zip

Files (239.1 MB)

Name Size Download all
md5:7f7d37bb04aeb7be82cd7544ef954e71
239.1 MB Preview Download

Additional details

Funding

U.S. National Science Foundation
CAREER: Securing the AI Stack in Autonomous CPS under Physical-Layer Attacks: A Systems Perspective 2145493
U.S. National Science Foundation
Research Infrastructure: Planning and Prototyping a Community-Wide Autonomous Driving Software Testing Infrastructure 2346561
U.S. National Science Foundation
Collaborative Research: SaTC: CORE: Small: NSF-DST: Towards Secure and Resilient Collaborative Autonomous Driving (CoAD) 2413877
U.S. National Science Foundation
CAREER: Enhancing Software Testing and Debugging for Autonomous Driving 2443763

Software

Repository URL
https://github.com/YuqiHuai/DeFT/
Programming language
Python , C++