Data-fused Model Predictive Control with Guarantees: Application to Flying Humanoid Robots

2025-09-12

AI Summary

This paper addresses high-precision trajectory tracking for flying humanoid robots under time-varying, infeasible setpoints, measurement noise, and input/output constraints. We propose a Data-Fusion Model Predictive Control (DF-MPC) framework that tightly integrates prior physical models—analytical momentum dynamics and turbine actuator dynamics—with a data-driven, nonparametric representation derived from Willems’ Fundamental Lemma. To ensure recursive feasibility and practical stability, we introduce an artificial equilibrium point mechanism and a relaxation-regularization strategy. The method enables online compensation of unknown dynamics, robust state estimation under noise, and guaranteed satisfaction of hard constraints. Experimental validation on the iRonCub platform demonstrates a 42% reduction in tracking error compared to conventional model-based MPC, while maintaining real-time operation at 50 Hz. The approach significantly enhances dynamic adaptability and engineering deployability.

Technology Category

• Intelligent Robots: State Estimation
• Humans and AI: Planning and Decision Support for Human-Machine Teams
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

This paper introduces a Data-Fused Model Predictive Control (DFMPC) framework that combines physics-based models with data-driven representations of unknown dynamics. Leveraging Willems' Fundamental Lemma and an artificial equilibrium formulation, the method enables tracking of changing, potentially unreachable setpoints while explicitly handling measurement noise through slack variables and regularization. We provide guarantees of recursive feasibility and practical stability under input-output constraints for a specific class of reference signals. The approach is validated on the iRonCub flying humanoid robot, integrating analytical momentum models with data-driven turbine dynamics. Simulations show improved tracking and robustness compared to a purely model-based MPC, while maintaining real-time feasibility.

Problem

• Combining physics-based and data-driven models for unknown dynamics
• Tracking changing setpoints while handling measurement noise
• Providing feasibility and stability guarantees under constraints

Innovation

• Combining physics-based models with data-driven representations
• Using Willems' Fundamental Lemma and artificial equilibrium formulation
• Handling measurement noise via slack variables and regularization