Related papers: Meta-Learning-Based Robust Adaptive Flight Control Under Uncertain Wind Conditions

Meta-Learning-Based Robust Adaptive Flight Control Under Uncertain Wind Conditions

URL: http://arxiv.org/abs/2103.01932v1
Date: Tue, 2 Mar 2021 18:43:59 GMT
Title: Meta-Learning-Based Robust Adaptive Flight Control Under Uncertain Wind Conditions
Authors: Michael O'Connell, Guanya Shi, Xichen Shi, Soon-Jo Chung
Abstract summary: Realtime model learning is challenging for complex dynamical systems, such as drones flying in variable wind conditions. We propose an online composite adaptation method that treats outputs from a deep neural network as a set of basis functions. We validate our approach by flying a drone in an open air wind tunnel under varying wind conditions and along challenging trajectories.
Score: 13.00214468719929
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Realtime model learning proves challenging for complex dynamical systems, such as drones flying in variable wind conditions. Machine learning technique such as deep neural networks have high representation power but is often too slow to update onboard. On the other hand, adaptive control relies on simple linear parameter models can update as fast as the feedback control loop. We propose an online composite adaptation method that treats outputs from a deep neural network as a set of basis functions capable of representing different wind conditions. To help with training, meta-learning techniques are used to optimize the network output useful for adaptation. We validate our approach by flying a drone in an open air wind tunnel under varying wind conditions and along challenging trajectories. We compare the result with other adaptive controller with different basis function sets and show improvement over tracking and prediction errors.

Related papers

Physics-Inspired Temporal Learning of Quadrotor Dynamics for Accurate Model Predictive Trajectory Tracking [76.27433308688592]
Accurately modeling quadrotor's system dynamics is critical for guaranteeing agile, safe, and stable navigation. We present a novel Physics-Inspired Temporal Convolutional Network (PI-TCN) approach to learning quadrotor's system dynamics purely from robot experience. Our approach combines the expressive power of sparse temporal convolutions and dense feed-forward connections to make accurate system predictions.
arXiv Detail & Related papers (2022-06-07T13:51:35Z)
Neural-Fly Enables Rapid Learning for Agile Flight in Strong Winds [96.74836678572582]
We present a learning-based approach that allows rapid online adaptation by incorporating pretrained representations through deep learning. Neural-Fly achieves precise flight control with substantially smaller tracking error than state-of-the-art nonlinear and adaptive controllers.
arXiv Detail & Related papers (2022-05-13T21:55:28Z)
Control-oriented meta-learning [25.316358215670274]
We use data-driven modeling with neural networks to learn, offline from past data, an adaptive controller with an internal parametric model of nonlinear features. We meta-learn the adaptive controller with closed-loop tracking simulation as the base-learner and the average tracking error as the meta-objective.
arXiv Detail & Related papers (2022-04-14T03:02:27Z)
Optimizing Airborne Wind Energy with Reinforcement Learning [0.0]
Reinforcement Learning is a technique that learns to associate observations with profitable actions without requiring prior knowledge of the system. We show that in a simulated environment Reinforcement Learning finds an efficient way to control a kite so that it can tow a vehicle for long distances.
arXiv Detail & Related papers (2022-03-27T10:28:16Z)
Learning Fast and Slow for Online Time Series Forecasting [76.50127663309604]
Fast and Slow learning Networks (FSNet) is a holistic framework for online time-series forecasting. FSNet balances fast adaptation to recent changes and retrieving similar old knowledge. Our code will be made publicly available.
arXiv Detail & Related papers (2022-02-23T18:23:07Z)
Learning-based vs Model-free Adaptive Control of a MAV under Wind Gust [0.2770822269241973]
Navigation problems under unknown varying conditions are among the most important and well-studied problems in the control field. Recent model-free adaptive control methods aim at removing this dependency by learning the physical characteristics of the plant directly from sensor feedback. We propose a conceptually simple learning-based approach composed of a full state feedback controller, tuned robustly by a deep reinforcement learning framework.
arXiv Detail & Related papers (2021-01-29T10:13:56Z)
Learning a Contact-Adaptive Controller for Robust, Efficient Legged Locomotion [95.1825179206694]
We present a framework that synthesizes robust controllers for a quadruped robot. A high-level controller learns to choose from a set of primitives in response to changes in the environment. A low-level controller that utilizes an established control method to robustly execute the primitives.
arXiv Detail & Related papers (2020-09-21T16:49:26Z)
Model-Based Meta-Reinforcement Learning for Flight with Suspended Payloads [69.21503033239985]
Transporting suspended payloads is challenging for autonomous aerial vehicles. We propose a meta-learning approach that "learns how to learn" models of altered dynamics within seconds of post-connection flight data.
arXiv Detail & Related papers (2020-04-23T17:43:56Z)
Logarithmic Regret Bound in Partially Observable Linear Dynamical Systems [91.43582419264763]
We study the problem of system identification and adaptive control in partially observable linear dynamical systems. We present the first model estimation method with finite-time guarantees in both open and closed-loop system identification. We show that AdaptOn is the first algorithm that achieves $textpolylogleft(Tright)$ regret in adaptive control of unknown partially observable linear dynamical systems.
arXiv Detail & Related papers (2020-03-25T06:00:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.