Tiny Learning-Based MPC for Multirotors: Solver-Aware Learning for Efficient Embedded Predictive Control

2024-10-31

arXiv.org

AI Summary

High-precision, high-frequency (100 Hz) model predictive control (MPC) remains infeasible for 53-g micro multirotors (e.g., Crazyflie 2.1) under severe computational constraints (Teensy 4.0 microcontroller). Method: This paper introduces the first solver-aware learning-based MPC framework tailored for microcontrollers. It integrates structured dynamical modeling, lightweight Gaussian process residual learning, and a custom, highly efficient quadratic programming (QP) solver, enabling full onboard deployment. Contribution/Results: The system achieves stable real-time operation at 100 Hz on the target hardware. Experimental evaluation demonstrates a 23% average reduction in trajectory tracking error compared to state-of-the-art embedded MPC approaches. To our knowledge, this is the first demonstration of real-time learning-based MPC on micro aerial vehicles, establishing a new paradigm for intelligent autonomous control on resource-constrained platforms.

Technology Category

• Machine Learning: Hardware-aware ML
• Intelligent Robots: Learning & Optimization for ROB
• Multiagent Systems: Multiagent Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• 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

Abstract

Tiny aerial robots show promise for applications like environmental monitoring and search-and-rescue but face challenges in control due to their limited computing power and complex dynamics. Model Predictive Control (MPC) can achieve agile trajectory tracking and handle constraints. Although current learning-based MPC methods, such as Gaussian Process (GP) MPC, improve control performance by learning residual dynamics, they are computationally demanding, limiting their onboard application on tiny robots. This paper introduces Tiny Learning-Based Model Predictive Control (LB MPC), a novel framework for resource-constrained micro multirotor platforms. By exploiting multirotor dynamics' structure and developing an efficient solver, our approach enables high-rate control at 100 Hz on a Crazyflie 2.1 with a Teensy 4.0 microcontroller. We demonstrate a 23% average improvement in tracking performance over existing embedded MPC methods, achieving the first onboard implementation of learning-based MPC on a tiny multirotor (53 g).

Problem

• Developing efficient embedded predictive control for tiny robots
• Overcoming computational constraints of learning-based MPC methods
• Enabling onboard model learning for agile trajectory tracking

Innovation

• Solver-aware learning for embedded predictive control
• Co-designed MPC framework for resource-constrained platforms
• Onboard implementation achieving 100 Hz control frequency