Effects of LiDAR Faults on the Perception of Autonomous Vehicles

Institute

Lehrstuhl für Fahrzeugtechnik (TUM-ED)

Type

Bachelor's Thesis / Semester Thesis / Master's Thesis /

Content

experimental /  

Description

MOTIVATION

Reliable perception is a key requirement for safe autonomous driving. However, real-world conditions such as fog, dirt on sensors, or partial occlusions can significantly degrade LiDAR measurements and affect downstream perception algorithms. 

This thesis investigates how different types of LiDAR degradation influence the performance of object detection and tracking systems. Using simulated degradation models (e.g., fog, sensor contamination, and reduced field-of-view) and real world data, you will evaluate the robustness of state-of-the-art perception algorithms on degraded point cloud data.

Requirements

YOUR ROLE

This thesis contributes to the analysis of how LiDAR faults affects the perception component of an Autonomous vehicle. The work packages of this thesis contains:

• Literature research on "LiDAR Faults", AV Perception” and "Perception Metrics”
• Analyze state of the art perception/detection frameworks, e. g. MMDetection3D, Autoware Perception 
• Conceptualize a pipeline to modify point clouds systematically to simulate real world sensor degradation
• Define metrics to evaluate the performance of perception/detection algorithms under degradation
• Integrate this pipeline into perception algorithms or a detection framework
• Analysis of data and documentation of results

WHAT YOU SHOULD BRING ALONG

• Strong Motivation and Interest for AVs
• Basic Knowledge in Programming, e. g. Python, C++
• Structured and independent way of working

If you are interested in joining this project, feel free to send me an application with your CV and transcript of records. I look forward to receive your application.

Possible start

sofort

Contact

Nijinshan Karunainayagam, M. Eng.
Room: MW3507
Phone: +49 89 289 15386
nijinshan.karunainayagamtum.de