PhD Position – Hybrid Learning Control for Agent-Based Intralogistics Systems

Organisation/Company cellumation GmbH Department HR Research Field Engineering » Computer engineering Engineering » Mechanical engineering Researcher Profile First Stage Researcher (R1) Positions PhD Positions Application Deadline 31 Mar 2026 - 23:59 (Europe/Berlin) Country Germany Type of Contract Temporary Job Status Full-time Hours Per Week 40 Offer Starting Date 1 May 2026 Is the job funded through the EU Research Framework Programme? Horizon Europe - MSCA Marie Curie Grant Agreement Number Is the Job related to staff position within a Research Infrastructure? No

Offer Description

About cellumation

cellumation is a Bremen-based deep-tech company and spin-off from the Bremen Institute of Production and Logistics (BIBA). We develop the celluveyor – a modular, software-defined material flow platform that is transforming intralogistics. Our patented technology replaces rigid conveyor systems with intelligent hexagonal cells equipped with omnidirectional wheels, enabling dynamic routing, sorting, buffering, and positioning. From e-commerce fulfillment to manufacturing and warehousing, the celluveyor adapts to any material flow challenge with maximum flexibility and minimal space.

About the CAVECORE Doctoral Network

CAVECORE (Continuous, Automated Validation, and Evaluation of Cognitive Robots in Open-Ended Environments) is a Marie Skłodowska-Curie Doctoral Network dedicated to advancing cognitive robotics. The network trains early-stage researchers to develop robots that can interact, learn, and adapt in real-world environments while ensuring quality, safety, and reliability through systematic validation methods. CAVECORE combines cutting-edge research in AI-enabled robotics with novel evaluation frameworks aligned with EU AI Act requirements.

You will work at the intersection of machine learning, control theory, and autonomous multi-agent systems to develop hybrid learning-based control strategies for cellumation's intelligent material flow platform.

Research Objectives:

• Combine machine learning and classical control theory, planning theory or multi agent systems to improve the performance of the celluveyor system. Examples of that could be:
  • Develop a trajectory generation system that can provide near optimal trajectories including speed and accelerations constraints at the extreme points of the parcels under extreme time constraints.
  • Develop a path generation system that globally optimizes the throughput of the system. The system must be able to generate paths in a tight time-window.
• The exact focus of the PhD thesis is not pre-defined and should be defined by the candidate over the course of the first year. The domain of the PhD thesis must be machine learning and either control theory, path planning, or multi agent systems.
• Usage of the design philosophy of hybrid controllers that combine the adaptability of machine learning with the safety guarantees of classical control methods. E.g., the system must be able to detect invalid output and provide fall back to allow for safe operation in real world applications.

Expected Results:

• The results should improve the performance of the celluveyor system in real world applications.
• The result must be reliable enough to apply to production machines.
• The result should be general enough to enable verification on 'Beckhoff XPlanar’ system.
• The developed result must meet the criteria to obtain a PhD.

Planned Secondments:

• University of Bielefeld (Prof. Neumann): 2 months in M14–15
• University of York (Prof. Gerasimou): 2 months in M26–27

You will gain hands-on experience with cellumation's cutting-edge technology, collaborate with leading European research institutions, and contribute to shaping the future of trustworthy autonomous systems in logistics.

What We Offer

• 36-month employment contract with cellumation GmbH
• Competitive salary according to Marie Skłodowska-Curie Actions regulations (EU Living Allowance + Mobility Allowance)
• Interdisciplinary training program combining robotics, AI, and validation methods
• International secondments at top-tier academic institutions
• Access to real-world intralogistics systems and state-of-the-art hardware/software platforms
• Collaboration with industry and academic partners across Europe
• Career development in a fast-growing deep-tech startup environment
• Enrollment in a doctoral program at a partner university

Your Profile

• Master's degree (or equivalent) in Computer Science, Engineering, Robotics, or a related field
• Excellent programming skills in C++ and Python
• Strong motivation to conduct innovative research at the intersection of machine learning and control systems
• Fluent in English (written and spoken)
• Practical experience with (multi-agent) path planning strategies
• Basic knowledge of ROS2 (Robot Operating System) and simulation techniques
• Hands-on experience with machine learning algorithms and developing/testing complex software systems
• Basic understanding of hardware/embedded software development

Additional Requirements (MSCA Eligibility):

• Early-stage researcher : at the time of recruitment, you must be in the first four years of your research career and have not yet been awarded a doctoral degree
• Mobility rule : you must not have resided or carried out your main activity (work, studies, etc.) in Germany for more than 12 months in the 3 years immediately before the recruitment date
• For applicants outside the EU : You need to apply for a work visa in Germany/EU, so additional requirements might be necessary (language certificates etc.)

Where to apply

E-mail

Requirements

Research Field Computer science » Programming Education Level Master Degree or equivalent

Skills/Qualifications

• Master's degree (or equivalent) in Computer Science, Engineering, Robotics, or a related field
• Excellent programming skills in C++ and Python
• Strong motivation to conduct innovative research at the intersection of machine learning and control systems
• Fluent in English (written and spoken)
• Practical experience with (multi-agent) path planning strategies