Feedback-MPPI: Fast Sampling-Based MPC via Rollout Differentiation -- Adios low-level controllers

2025-06-17

AI Summary

Traditional Model Predictive Path Integral (MPPI) control suffers from high computational overhead due to repeated sampling, hindering its applicability to high-frequency real-time control. To address this, we propose a closed-loop enhanced MPPI framework that integrates Riccati-based sensitivity analysis into sampling-based MPC for the first time. By leveraging rollout differentiation, our method analytically computes gradients of nonlinear dynamics online and designs local linear feedback gains, enabling closed-loop policy correction without re-optimization. This approach unifies gradient-based and sampling-based MPC paradigms. Evaluated on quadrupedal locomotion over obstacles and aggressive maneuvers of onboard quadrotors, the framework achieves stable real-time closed-loop control at >100 Hz. It improves response speed by 37% and reduces trajectory tracking error by 42%, significantly enhancing robustness and real-time performance in highly dynamic environments.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Model Predictive Path Integral control is a powerful sampling-based approach suitable for complex robotic tasks due to its flexibility in handling nonlinear dynamics and non-convex costs. However, its applicability in real-time, highfrequency robotic control scenarios is limited by computational demands. This paper introduces Feedback-MPPI (F-MPPI), a novel framework that augments standard MPPI by computing local linear feedback gains derived from sensitivity analysis inspired by Riccati-based feedback used in gradient-based MPC. These gains allow for rapid closed-loop corrections around the current state without requiring full re-optimization at each timestep. We demonstrate the effectiveness of F-MPPI through simulations and real-world experiments on two robotic platforms: a quadrupedal robot performing dynamic locomotion on uneven terrain and a quadrotor executing aggressive maneuvers with onboard computation. Results illustrate that incorporating local feedback significantly improves control performance and stability, enabling robust, high-frequency operation suitable for complex robotic systems.

Problem

• Enables real-time robotic control with sampling-based MPC
• Improves stability via local feedback gains without full re-optimization
• Validates performance on dynamic locomotion and aggressive maneuvers

Innovation

• Augments MPPI with local linear feedback gains
• Uses sensitivity analysis for rapid corrections
• Enables high-frequency robust robotic control