1: \begin{abstract}

2: The requirements for real-world manipulation tasks are diverse and often conflicting; some tasks necessitate force constraints or collision avoidance, while others demand high-frequency feedback. Satisfying these varied requirements with a fixed state-action representation and control strategy is challenging, impeding the development of a universal robotic foundation model. In this work, we propose \MC, the first LLM-enabled automatic control synthesis approach that creates customized state representations and control strategies tailored to specific tasks. \MC leverages a generic hierarchical control framework to address a wide range of heterogeneous tasks. Our core insight is the decomposition of the state space into an abstract task space and a concrete tracking space. By harnessing LLM's extensive common sense and control knowledge, we enable the LLM to design these spaces, including states, dynamic models, and controllers, using pre-defined but abstract templates. 

3: \MC stands out for its fully model-based nature, allowing for rigorous analysis, efficient parameter tuning, and reliable execution. It not only utilizes decades of control expertise encapsulated within LLMs to facilitate heterogeneous control but also ensures formal guarantees such as safety and stability. Our method is validated both in real-world scenarios and simulations across diverse tasks with conflicting requirements, such as collision avoidance versus convergence and compliance versus high precision.

4: % \footnote{Webpage with videos and additional results: \url{https://meta-control-paper.github.io}}

5: Videos and additional results are at \url{meta-control-paper.github.io}

6: \end{abstract}

7: