Masterthesis - Reinforcement Learning approach for path following and base control

Advertisement for the field of study such as: Robotics, Cybernetics, Computer Science, Mechanical Engineering, Mechatronics or comparable.

In the “Mobile Robot Navigation” research group, we develop autonomous mobile robots for outdoor applications, such as in agriculture and forestry, the municipal sector, and logistics. The focus is on precise, robust navigation in outdoor environments.

Classic local controllers (e.g., Regulated Pure Pursuit, MPPI) calculate the target speeds for a given path, which the base controller translates into wheel commands. This modular architecture often requires a high level of parameterization and shows limited adaptability to different vehicle types and operating conditions. The aim of the master's thesis is to investigate an end-to-end reinforcement learning approach for path tracking that learns to map the target path and current sensor data directly to motor control variables. The focus is on evaluating the extent to which such an approach can improve portability to new vehicle types and increase path following accuracy compared to classic architectures.

What you will do

• Analysis of the existing navigation stack (local controller/path-following controller, base controller)
• Development and training of a reinforcement learning approach that maps the classic separation of path-following controller and base controller in a common policy
• Investigation of model-free and/or model-based RL methods in terms of stability, data requirements, and transferability
• Integration into an existing ROS2 environment
• Validation in simulation in an existing ROS2 environment and on our own robots
• Comparison with existing controllers in terms of path error, stability, and commissioning time
• Documentation, evaluation, and scientific processing of the results

What you bring to the table

• Valid enrollment at a German university/Hochschule in robotics, cybernetics, computer science, mechanical engineering, mechatronics, or similar fields
• Experience with reinforcement learning
• Experience with ROS is an advantage
• Analytical thinking skills
• Enthusiasm for mobile robotics
• Fluent in English or German

What you can expect

• Cutting-edge technology in the field of mobile outdoor robotics
• Practical work with our robots in Stuttgart
• Responsibility and freedom to implement your own ideas
• Collaboration with the best students in their field

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

+49 711 970-1415

[email protected]

Fraunhofer Institute for Manufacturing Engineering and Automation IPA

Requisition Number: 82688Application Deadline: