Student job, internship Testing and development of advanced imitation learning algorithms

Most vehicles in intralogistics are driven manually, as human drivers are superior to automated systems in many respects from today's perspective. In order to harness human capabilities for automated systems, human driving behavior is to be simulated in a logistics environment and used to generate synthetic data sets. Based on this, an industrial truck (FFZ) is to be enabled via imitation learning to autonomously execute driving movements based on the implicit knowledge of experienced drivers. We are investigating this in the LernFFZ project using various methods of imitation learning in a simulation environment.

Your tasks

As part of your internship or part-time job, you will have the opportunity to familiarize yourself with the exciting field of imitation learning (IL) and actively participate in the LernFFZ project! You will independently test already known IL algorithms in Python and check their applicability. Furthermore, you will actively develop the algorithms further and adapt them to the use case at hand. The following activities await you in detail:

• Testing of already known imitation learning (IL) algorithms
• Further development of known system architectures
• Adaptation of the algorithms to the present use case
• Optimization of the hyperparameters

Your profile

You are studying one of the following subjects:

• Computer science
• Data science

You are interested in developing and implementing challenging and educational activities in the field of Imitation Learning (IL) and have a sound knowledge of Python and Pytorch. Ideally, you already have previous experience in the field of machine learning and have already trained your first neural network.

A final thesis following the internship or part-time job is also possible.

We offer

• Appropriate remuneration for student job
• independent work
• flexible working hours
• well-equipped workplaces
• test implementation
• possibly long-term cooperation