Master Thesis - "Vision Foundation Models for automated point cloud map cleanup"

Fraunhofer-Gesellschaft; Stuttgart; Vor Ort; EUR 60.000 - 80.000; Vollzeit

Zusammenfassung

A leading research organization in Stuttgart is seeking an enrolled student for a thesis project focusing on vision foundation models in robot localization. The role involves using DINOv2 features for efficient map management and evaluating the usability for real-time localization. Ideal candidates will have a background in relevant fields and programming skills in C++/Python. This position offers hands-on experience with cutting-edge robotics technology.

Leistungen

Cutting-edge technology

Hands-on experience with robots

Familiar atmosphere including Cake Thursday

Qualifikationen

• Enrolled student in a relevant field.
• Knowledge of C++ and Python.
• Experience with ROS, computer vision, and machine learning is a plus.

Aufgaben

• Explore the use of vision foundation models within robot localization.
• Utilize DINOv2 features for efficient map cleanup.
• Evaluate the usability of the created maps for real-time robot localization.

Kenntnisse

C++

Python

Analytical mindset

Experience with ROS

Computer vision knowledge

Pointcloud processing

Machine learning experience

Fluent in English or German

Ausbildung

Enrolled student in Computer Science, Software Engineering, Electrical Engineering, Autonomous Systems or similar

Jobbeschreibung

Overview

Announcement for fields of study such as: Automation technology, computer science, control engineering, software engineering, technical computer science or comparable.

In the Professional Service Robots - Outdoor research group we develop autonomous, mobile robots for a variety of outdoor applications, such as agriculture, forestry and logistics. The focus is on the development of an autonomous outdoor navigation solution as well as the hardware of the robots.

Recently, advances in LiDAR scan registration have been demonstrated by using Vision Foundation Models such as DINOv2 to generate point descriptors mapped from camera images. This type of pointcloud annotation is expected to especially benefit creation and maintenance of long-term maps, which contain only static features of the environment. Combined with previous works in dynamic object detection and removal, an automated pipeline for creating such maps will be developed, which we expect to be of great benefit to real-time localization in changing environments.

What you will do

In this thesis you will explore the use of vision foundation models within robot localization. Specifically, you will utilize DINOv2 (or similar) features to enable efficient map cleanup, which might be self-learning by utilizing an online dynamic object removal/detection technique. The usability of such a map as reference for real-time robot localization will then be evaluated.

What you bring to the table

• Enrolled student
• Background in Computer Science, Software Engineering, Electrical Engineering, Autonomous Systems or similar
• Knowledge of C++/Python
• Experience with ROS, computer vision, pointcloud-processing or machine learning is a plus
• Analytical mindset and experience in algorithm development
• Enthusiasm for mobile robotics
• Fluent in English or German

What you can expect

• Cutting-edge technology in the field of outdoor mobile robotics
• Hands on with our robots on our own test environments in Stuttgart
• take on responsibility and freedom to implement your own ideas
• Work with the best students in their discipline
• Familiar atmosphere including Cake Thursday

We value and promote the diversity of our employees\' skills and therefore welcome all applications - regardless of age, gender, nationality, ethnic and social origin, religion, ideology, disability, sexual orientation and identity. Severely disabled persons are given preference in the event of equal suitability.

With its focus on developing key technologies that are vital for the future and enabling the commercial utilization of this work by business and industry, Fraunhofer plays a central role in the innovation process. As a pioneer and catalyst for groundbreaking developments and scientific excellence, Fraunhofer helps shape society now and in the future.

Interested? Apply online now

We look forward to getting to know you!

Frau Jennifer Leppich
Recruiting
[phone]

Fraunhofer Institute for Manufacturing Engineering and Automation IPA

[e-mail]

Fraunhofer IPA, Requisition Number: 81348