AI-Driven Reporting, Alerting and Autonomous Error Management for Roadside Sensor Systems

02.03.2026, Abschlussarbeiten, Bachelor- und Masterarbeiten

Design an AI-based system that transforms detected perception anomalies into structured reports, alerts, action recommendations, and automated recordings for scalable infrastructure monitoring.

Motivation & Relevance

Detecting anomalies is only the first step. For scalable roadside infrastructure operation, perception failures must trigger meaningful actions: alerts, structured reports, recommended interventions, and documentation for later analysis. Today, most infrastructure monitoring lacks intelligent orchestration between anomaly detection and operational response. An AI-driven error management layer can significantly improve reliability and maintainability.

This thesis focuses on operationalization and autonomous system supervision.

Project Description

In this thesis, you will design a reporting and error-management framework that builds on detected perception anomalies.

The system will:

• aggregate anomaly events
• classify error types and severity levels
• generate structured monitoring reports
• trigger real-time alerts
• create automated failure-aware scene recordings
• generate action recommendations (e.g., recalibration, sensor check)
• design explainable outputs for operators
• evaluate alert quality and false alarm rate

The focus is not detection itself, but intelligent orchestration and decision support.

Your Tasks

• Design event aggregation and severity models
• Implement alerting and logging pipeline
• Develop rule-based and AI-based action recommendations
• Design explainability concepts
• Evaluate system usability and robustness

Your Profile

• Master’s student in CS, Robotics, EE or related field
• Strong programming skills
• Interest in AI agents and system supervision
• Experience with distributed systems is a plus

What you will gain

• Experience in AI system orchestration
• Knowledge of automated infrastructure monitoring
• Insight into real-world operational AI systems
• Hands-on work on scalable supervision architectures

I am looking forward to receiving your application. Please include a very brief motivational statement and your grade transcript

Kontakt: erik-leo.hass@tum.de