Neurorobotik
Current Topic
Neurorobotik 

Investigating the Integration of Large Language Models for Natural Language-Based Robot Navigation using ROS2 and NAV2

Illustration of a mobile robot that plans a path based on a voice command. (AI generated)Illustration, AI generated (Google AI)

This master thesis explores the potential of leveraging Large Language Models (LLMs) and intelligent agents to enable mobile robot navigation through natural language commands. Building upon the established ROS2 and NAV2 frameworks, the research will investigate how LLMs can interpret and translate complex human instructions into executable navigation tasks. The project will involve a comprehensive literature review on the current state-of-the-art, particularly focusing on developments like ros-llm and similar agent-based architectures. A key component of the work will be the design and implementation of a system that integrates an LLM with the robot's control stack. The thesis will evaluate the system's performance, robustness, and adaptability in various real-world scenarios.

Advisor:

Requirements:

  • Basic knowledge in Linux/Ubuntu would be good.
  • Basic knowledge in ROS/Python programming.
  • Basic knowledge in LLMs.