Thesis work AI-Powered Fault Analysis System for Software-Defined Vehicle CCU

About this opportunity - Background

The central compute unit of modern software-defined vehicles generates complex fault reports that require extensive manual analysis. Current root cause analysis for these faults can range from hours to months creating significant bottlenecks in the development and validation process. With approximately 200 fault reports per quarter requiring analysis and each report containing megabytes to gigabytes of heterogeneous log data (DLT logs Network traffic coredumps ...etc) engineers spend considerable time manually correlating information across multiple log formats to identify issues.
Additionally the rapid evolution of the software platform means that behavioral patterns learned from historical data may become obsolete requiring adaptive analysis approaches.

Scope of the thesis work

Primary Goal:
Develop an AI-powered fault analysis system that reduces the time-to-diagnosis for fault reports from the vehicles central compute unit.

Specific Objectives:
1. Create specialized parsing and tokenization methods for automotive log formats (DLT Pcap coredump blf)
2. Build a Large Language Model-based system capable of analyzing multi-gigabyte fault report datasets
3. Implement anomaly detection that identifies suspicious patterns when root cause analysis isnt possible
4. Develop a confidence-aware system that distinguishes between identified root causes and observed anomalies
5. Create a knowledge update mechanism to handle evolving platform behavior without full model retraining

Expected Deliverables:
A fine-tuned LLM specialized for automotive fault analysis
Custom tokenizers for DLT Pcap coredump and blf formats
Tool-augmented AI agent with access to debugging utilities (GDB integration for coredumps)
Knowledge update mechanism with retrieval support
Usable deployment with basic interface (stretch goal: cloud-deployed web service)
Evaluation framework with metrics on historical fault reports

What youll bring

Essential Skills:
Strong Python programming (PyTorch/TensorFlow experience)
Understanding of transformer architectures and LLM fine-tuning
Linux/Unix systems programming knowledge
Version control (Git) and collaborative development
Preferred Skills:
Familiarity with automotive software development (QNX)
Experience with NVIDIA Drive platform or embedded GPU systems
Knowledge of debugging tools (GDB core dump analysis)
Soft Skills:
Ability to work with domain experts to understand fault patterns
Strong analytical and problem-solving capabilities
Clear technical communication for cross-functional collaboration

Duration

The work will start in January 2026
The duration for this thesis work is 20 weeks.
30 ECTS (Master academic credits) in agreement with your Thesis Advisor in University
This thesis is to be conducted by 2 Students working in pair.