Most of the systems performing open-vocabulary navigation use closed LLMs accessible via an API. If this kind of systems are to be adopted for practical robotic applications, it is important that the information is not sent to third-party systems. The goal of the project is to evaluate fine-tuning of local open-source models for processing natural language commands for robotic navigation.

Recent breakthroughs in natural language processing are enabling robots to understand human language like never before. The thesis will address the problem of translating human language to executable robot skills.

The goal of this master thesis is to integrate a large vision-language model (VLM) with a manipulation policy in order to control a robotic hand for predefined manipulation tasks, such as grasping or pushing.

The goal of this master thesis is to explore existing approaches, datasets and models that provide textual explanations of driving situations, to implement a state-of-the-art model and to validate it on predefined driving conflict situations.