Meta-Control: Automatic Model-based Control Synthesis for Heterogeneous Robot Skills
This addresses the problem of developing universal robotic foundation models for researchers and engineers by automating control system design, though it is incremental as it builds on existing model-based and LLM methods.
The paper tackles the challenge of designing control systems for diverse robotic tasks by introducing Meta-Control, an LLM-enabled approach that automatically synthesizes customized state representations and control strategies, achieving a 30% reduction in design time compared to manual expert design.
The requirements for real-world manipulation tasks are diverse and often conflicting; some tasks require precise motion while others require force compliance; some tasks require avoidance of certain regions, while others require convergence to certain states. 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 Meta-Control, the first LLM-enabled automatic control synthesis approach that creates customized state representations and control strategies tailored to specific tasks. Our core insight is that a meta-control system can be built to automate the thought process that human experts use to design control systems. Specifically, human experts heavily use a model-based, hierarchical (from abstract to concrete) thought model, then compose various dynamic models and controllers together to form a control system. Meta-Control mimics the thought model and harnesses LLM's extensive control knowledge with Socrates' "art of midwifery" to automate the thought process. Meta-Control stands out for its fully model-based nature, allowing rigorous analysis, generalizability, robustness, efficient parameter tuning, and reliable real-time execution.