Control-oriented meta-learning
- URL: http://arxiv.org/abs/2204.06716v1
- Date: Thu, 14 Apr 2022 03:02:27 GMT
- Title: Control-oriented meta-learning
- Authors: Spencer M. Richards, Navid Azizan, Jean-Jacques Slotine, Marco Pavone
- Abstract summary: 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.
- Score: 25.316358215670274
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Real-time adaptation is imperative to the control of robots operating in
complex, dynamic environments. Adaptive control laws can endow even nonlinear
systems with good trajectory tracking performance, provided that any uncertain
dynamics terms are linearly parameterizable with known nonlinear features.
However, it is often difficult to specify such features a priori, such as for
aerodynamic disturbances on rotorcraft or interaction forces between a
manipulator arm and various objects. In this paper, we turn to data-driven
modeling with neural networks to learn, offline from past data, an adaptive
controller with an internal parametric model of these nonlinear features. Our
key insight is that we can better prepare the controller for deployment with
control-oriented meta-learning of features in closed-loop simulation, rather
than regression-oriented meta-learning of features to fit input-output data.
Specifically, we meta-learn the adaptive controller with closed-loop tracking
simulation as the base-learner and the average tracking error as the
meta-objective. With both fully-actuated and underactuated nonlinear planar
rotorcraft subject to wind, we demonstrate that our adaptive controller
outperforms other controllers trained with regression-oriented meta-learning
when deployed in closed-loop for trajectory tracking control.
Related papers
- DATT: Deep Adaptive Trajectory Tracking for Quadrotor Control [62.24301794794304]
Deep Adaptive Trajectory Tracking (DATT) is a learning-based approach that can precisely track arbitrary, potentially infeasible trajectories in the presence of large disturbances in the real world.
DATT significantly outperforms competitive adaptive nonlinear and model predictive controllers for both feasible smooth and infeasible trajectories in unsteady wind fields.
It can efficiently run online with an inference time less than 3.2 ms, less than 1/4 of the adaptive nonlinear model predictive control baseline.
arXiv Detail & Related papers (2023-10-13T12:22:31Z) - Active Learning of Discrete-Time Dynamics for Uncertainty-Aware Model
Predictive Control [49.60520501097199]
We present a self-supervised learning approach that actively models the dynamics of nonlinear robotic systems.
Our approach showcases high resilience and generalization capabilities by consistently adapting to unseen flight conditions.
arXiv Detail & Related papers (2022-10-23T00:45:05Z) - 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) - Data-Efficient Deep Reinforcement Learning for Attitude Control of
Fixed-Wing UAVs: Field Experiments [0.37798600249187286]
We show that DRL can successfully learn to perform attitude control of a fixed-wing UAV operating directly on the original nonlinear dynamics.
We deploy the learned controller on the UAV in flight tests, demonstrating comparable performance to the state-of-the-art ArduPlane proportional-integral-derivative (PID) attitude controller.
arXiv Detail & Related papers (2021-11-07T19:07:46Z) - Learning Adaptive Control for SE(3) Hamiltonian Dynamics [15.26733033527393]
This paper develops adaptive geometric control for rigid-body systems, such as ground, aerial, and underwater vehicles.
We learn a Hamiltonian model of the system dynamics using a neural ordinary differential equation network trained from state-control trajectory data.
In the second stage, we design a trajectory tracking controller with disturbance compensation from an energy-based perspective.
arXiv Detail & Related papers (2021-09-21T05:54:28Z) - Adaptive-Control-Oriented Meta-Learning for Nonlinear Systems [29.579737941918022]
We 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.
With a nonlinear planar rotorcraft subject to wind, we demonstrate that our adaptive controller outperforms other controllers trained with regression-oriented meta-learning.
arXiv Detail & Related papers (2021-03-07T23:49:59Z) - Meta-Learning-Based Robust Adaptive Flight Control Under Uncertain Wind
Conditions [13.00214468719929]
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.
arXiv Detail & Related papers (2021-03-02T18:43:59Z) - Anticipating the Long-Term Effect of Online Learning in Control [75.6527644813815]
AntLer is a design algorithm for learning-based control laws that anticipates learning.
We show that AntLer approximates an optimal solution arbitrarily accurately with probability one.
arXiv Detail & Related papers (2020-07-24T07:00:14Z) - Tracking Performance of Online Stochastic Learners [57.14673504239551]
Online algorithms are popular in large-scale learning settings due to their ability to compute updates on the fly, without the need to store and process data in large batches.
When a constant step-size is used, these algorithms also have the ability to adapt to drifts in problem parameters, such as data or model properties, and track the optimal solution with reasonable accuracy.
We establish a link between steady-state performance derived under stationarity assumptions and the tracking performance of online learners under random walk models.
arXiv Detail & Related papers (2020-04-04T14:16:27Z) - 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.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.