Masterarbeit (w/m/d): Comparative Analysis of Multi-Sensor Integration in SENSE-Rai System

Overview

The DLR Institute of Software Technology sees software as a catalyst for research and innovation. The institute’s staff, currently numbering around 200, make significant contributions to advancements in aviation, space, energy, transportation, and security through the development of state‑of‑the‑art software solutions and innovative research.

Our areas of competence include reliable and safety‑critical software systems, artificial intelligence, high‑performance computing and quantum computing, human‑system interaction and visualisation, software and systems engineering, as well as digital platforms and digital twins.

What To Expect

This thesis investigates the integration of multiple sensors within SENSE‑Rai systems, utilizing the Robot Operating System 2 (ROS 2) framework. It performs a comparative analysis of performance, scalability, and overhead when using ROS 2 with various Data Distribution Services (DDS), covering aspects such as cost, licensing, efficiency, and optimization capabilities. The study also explores the potential for sensor technology transfer to market applications.

SENSE‑Rai aims to develop new technologies and methods to enhance data processing with artificial intelligence (AI) through the use of real and virtual sensors (e.g., cameras, lidars, etc.), standardized interfaces, and various software and hardware. It focuses on combining physical models with data‑driven models and emphasizes edge computing.

The project establishes methodological and technological foundations leading to specific system developments for selected use cases, including the creation of a middleware layer with ROS 2. This middleware supports seamless integration of diverse sensor systems, managing data acquisition, calibration, data referencing, and quality measurement.

Your tasks

• Literature Review
  
  • Overview of current research on sensor integration in embedded systems.
  • DDS in ROS 2, focusing on cost, licensing, efficiency, optimization capabilities, and advantages for embedded systems such as real‑time communication, modularity, and scalability.
• Objectives
  
  • Compare the performance and scalability of multi‑sensor systems with different DDS integration (e.g., eProsima Fast DDS, Eclipse Cyclone DDS, RTI Connext DDS, Zetascale Zenoh DDS).
  • Quantify the network overhead introduced by using default DDS settings in comparison to others.
  • Evaluate the potential for technology transfer of sensor integration techniques to similar SENSE‑Rai‑embedded systems, and determine the most suitable DDS based on comparative results.
• Expected Contributions
  
  • Quantified metrics for the performance, scalability, and overhead of different DDS configurations for the SENSE‑Rai framework.
  • Recommendations for optimizing sensor integration and data processing in SENSE‑Rai embedded systems.

Your profile

• Enrolled in a relevant degree program (mechatronics, computer engineering, electrical engineering, robotics, informatics, or closely related field) and looking for a thesis topic.
• Experience with programming languages and embedded Linux (e.g., C/C++, Python, Bash).
• Familiarity with ROS 2 (nodes, topics, services, parameters, launch files).
• Use and configuration of DDS middleware (e.g., Fast DDS, Cyclone DDS, Connext, Zenoh).
• Performance measurement and benchmarking (latency, throughput, CPU, memory, bandwidth).
• Familiar with Git, unit testing, CI/CD basics, and documentation.
• Ability to communicate in English, as it is the supervision language.

We offer

DLR stands for diversity, appreciation and equality for all people. We promote independent work and the individual development of our employees both personally and professionally. To this end, we offer numerous training and development opportunities. Equal opportunities are of particular importance to us, which is why we want to increase the proportion of women in science and management in particular. Applicants with severe disabilities will be given preference if they are qualified.