Master/Bachelor - Survey of Vehicle AD - Software Architectures (f/m/d)

Salary: 24.500 - 44.500 € per year

Requirements:

Academic studies in a relevant technical or engineering field such as Industrial Engineering, Mechanical Engineering, Automotive Engineering, Computer Science, Robotics, Electrical Engineering, Mechatronics, Control Engineering, Aerospace Engineering, or Technical Business Administration or similar

Interest in automated driving, software architectures, and agentic AD/LLM-based concepts.

Basic knowledge of literature research and academic referencing (e.g., BibTeX, Zotero, systematic screening).

Structured, analytical working style with precise documentation (EN/DE).

Beneficial: Basic skills in programming (e.g., Python/C++) and experience in open-source communities.

Remuneration is according to the German TVöD 05.

Responsibilities:

refine the topic and formulate precise research questions.

conduct a systematic literature search (IEEE, ACM, arXiv, Google Scholar), screen the studies, and extract relevant data.

develop a taxonomy of architectures: end-to-end vs. modular vs. agentic (roles, orchestration, tool use, safety/runtime aspects).

compare open-source AD stacks, analyzing architecture layers, middleware (e.g., ROS 2/CyberRT), APIs, roadmaps, releases, licensing/governance, community metrics, and maturity levels.

map agentic AD concepts, including agent roles (planner, safety critic/verifier, perception narrator, controller designer), communication patterns, knowledge integration (e.g., RAG), and integration points within existing stacks.

compile the landscape of relevant benchmarks and datasets, especially for scenarios such as parking or docking.

create evidence syntheses, comparison tables, and visualizations, deriving guidelines and research gaps.

contribute to scientific writing (DE/EN) and prepare the final presentation.

Technologies:

Apollo

LLM

Python

RAG

ROS

Robotics

Architect

More:

The Institute of Vehicle Concepts (FK) of the German Aerospace Centre (DLR) is internationally recognised for the design of future road and rail vehicles that enable climate and environmentally friendly mobility while being affordable and user-friendly at the same time. We research and demonstrate the required key technologies and maintain close cooperation with other scientific institutions as well as industrial and political bodies. You will work with our team on a systematic survey of software architectures for automated driving, focusing on end-to-end approaches, modular pipelines, and agentic AD systems (LLM-based, tool-using agents and orchestration). We develop a taxonomy of the different architectures, analyze relevant open-source stacks such as Autoware, Baidu Apollo, and potentially S-Core/Red Hat ecosystems based on documentation, releases, and repositories, and derive technological trends, gaps, and actionable recommendations.

last updated 39 week of 2026