Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enabling On-Chip High-Frequency Adaptive Linear Optimal Control via Linearized Gaussian Process

Yuan Gao, Yinyi Lai, Jun Wang, Yini Fang

TL;DR

Gaussian process is incorporated to model the adaptive external aerodynamics with linear model predictive control to enable real-time high-frequency solutions and end-to-end Bayesian optimization during sample collection stages to improve the control performance.

Abstract

Unpredictable and complex aerodynamic effects pose significant challenges to achieving precise flight control, such as the downwash effect from upper vehicles to lower ones. Conventional methods often struggle to accurately model these interactions, leading to controllers that require large safety margins between vehicles. Moreover, the controller on real drones usually requires high-frequency and has limited on-chip computation, making the adaptive control design more difficult to implement. To address these challenges, we incorporate Gaussian process (GP) to model the adaptive external aerodynamics with linear model predictive control. The GP is linearized to enable real-time high-frequency solutions. Moreover, to handle the error caused by linearization, we integrate end-to-end Bayesian optimization during sample collection stages to improve the control performance. Experimental results on both simulations and real quadrotors show that we can achieve real-time solvable computation speed with acceptable tracking errors.

Enabling On-Chip High-Frequency Adaptive Linear Optimal Control via Linearized Gaussian Process

TL;DR

Gaussian process is incorporated to model the adaptive external aerodynamics with linear model predictive control to enable real-time high-frequency solutions and end-to-end Bayesian optimization during sample collection stages to improve the control performance.

Abstract

Unpredictable and complex aerodynamic effects pose significant challenges to achieving precise flight control, such as the downwash effect from upper vehicles to lower ones. Conventional methods often struggle to accurately model these interactions, leading to controllers that require large safety margins between vehicles. Moreover, the controller on real drones usually requires high-frequency and has limited on-chip computation, making the adaptive control design more difficult to implement. To address these challenges, we incorporate Gaussian process (GP) to model the adaptive external aerodynamics with linear model predictive control. The GP is linearized to enable real-time high-frequency solutions. Moreover, to handle the error caused by linearization, we integrate end-to-end Bayesian optimization during sample collection stages to improve the control performance. Experimental results on both simulations and real quadrotors show that we can achieve real-time solvable computation speed with acceptable tracking errors.
Paper Structure (22 sections, 16 equations, 5 figures, 1 table, 1 algorithm)

This paper contains 22 sections, 16 equations, 5 figures, 1 table, 1 algorithm.

Table of Contents

  1. INTRODUCTION
  2. Related Works
  3. Problem Statement
  4. Linearized dynamics of the quadrotors
  5. Dynamic Behavior of Swarm Systems
  6. Problem Statement and Proposed Approach
  7. Preliminaries
  8. Gaussian Process and Bayesian Optimization
  9. Linear-Quadratic Regulator and Linear MPC
  10. controller design
  11. Linearized Gaussian Processes
  12. Adaptive Linear MPC
  13. Sampling Scheme using Bayesian Optimization
  14. Experimental Results
  15. Simulation
  16. ...and 7 more sections

Figures (5)

  • Figure 1: Exploring complex interactions between multirotors, we developed a GP-based controller for close-proximity flight, using the position difference between the two drones as input.
  • Figure 2: Screenshots of LinGP-LinMPC controller on real drones, which is able to predict the downwash force and keep stable.
  • Figure 4: The Gaussian process model is used to capture the main trends of the data and make force predictions based on the changes in the horizontal and vertical distances.
  • Figure 5: The tracking performance of various controllers on a swapping task across different initial positions in the simulator.
  • Figure 6: Crazyflie and the experiment environment.