Master’s Thesis - Conflict Detection and Resolution in Roadside-Vehicle Cooperative Perception
10.03.2026, Abschlussarbeiten, Bachelor- und Masterarbeiten
The Chair of Robotics, Artificial Intelligence, and Real-Time Systems offers a Master’s thesis focusing on reliability and redundancy management in cooperative perception systems.
Motivation & Relevance
Cooperative perception systems combine information from multiple sensors across vehicles and roadside infrastructure. While redundancy improves environmental awareness, it also introduces challenges when sensors report conflicting or inconsistent information. Examples include phantom objects, duplicated detections, inconsistent trajectories, or mismatching object classifications. Without proper conflict management, these inconsistencies can degrade perception reliability and compromise safety. Detecting and resolving such conflicts is essential for scalable cooperative perception deployments.
Project Description
In this thesis, you will develop algorithms for detecting and resolving conflicts between vehicle and roadside perception systems. Your system will:
- Identify mismatches between object detections from multiple sensors
- Detect phantom objects and duplicated tracks
- Analyze conflicting object states such as position or velocity
- Develop conflict resolution strategies using confidence scores or rules
- Evaluate improvements in perception consistency and reliability
The result will be a framework that improves robustness of cooperative perception through conflict-aware redundancy management.
Your tasks
- Design algorithms for conflict detection in multi-sensor perception
- Implement methods for resolving inconsistent object states
- Analyze error cases such as phantom objects or ID switches
- Evaluate performance improvements using cooperative perception datasets
Your Profile
- Master’s student in Computer Science, Robotics, Electrical Engineering or related field
- Proficiency in Python or C++
- Interest in sensor fusion and perception systems
- Experience with object tracking or multi-sensor systems is beneficial
What you will gain
- Expertise in cooperative perception reliability
- Hands-on experience with multi-sensor perception pipelines
- Insight into redundancy management for safety-critical systems
How to apply
Please send your CV and a transcript of your grades with your application.
Kontakt: erik-leo.hass@tum.de