We provide a stability analysis tool for controllers acting on dynamics represented by Gaussian processes (GPs).
GPs infer a distribution over all plausible models given the observed data instead of one (possibly erroneous) model and, thus, avoid severe modeling errors.
We present a tool to analyze stability for a given con- troller and Gaussian process (GP) dynamics model. Our tool analytically derives a stability region where ...
This work provides a stability analysis tool for controllers acting on dynamics represented by Gaussian processes, and considers arbitrary Markovian control ...
Employing Gaussian processes as forward models for learning control is particularly appealing as they incorporate uncertainty about the system dynamics. GPs ...
We provide stability analysis tools for controllers acting on dynamics represented by Gaussian processes (GPs).
Missing: Forward Models.
Abstract—Gaussian Process State Space Models (GP-SSMs) are a non-parametric model class suitable to represent non- linear dynamics.
Abstract—This paper introduces a learning-based robust control algorithm that provides robust stability and perfor- mance guarantees during learning.
People also ask
What are the limitations of Gaussian processes?
Is Gaussian process regression non-linear?
Is the Gaussian process a kernel method?
What are the advantages of Gaussian processes?
The aim of this paper is to develop a novel synthesis technique for repetitive control (RC) based on a new more general internal model.
This adaptability is crucial in aerial robotics, where the capacity to quickly adjust to new conditions is key to maintaining both stability and optimal ...