Data-Driven Feedback Linearization with Manifold Optimization – code and simulation data
收藏DataCite Commons2026-04-08 更新2026-05-04 收录
下载链接：
https://repod.icm.edu.pl/citation?persistentId=doi:10.18150/YEGX8D
下载链接
链接失效反馈官方服务：
资源简介：
This dataset contains the code used to reproduce the experiments from the paper:"Data-Driven Feedback Linearization with Manifold Optimization"(J. Piasek-Skupna, 2025). https://doi.org/10.1016/j.ifacol.2026.03.012The code implements a manifold-constrained learning framework for feedback linearization using PyTorch and Pymanopt.Main components:- polynomial + Fourier feature map for φ(x)- manifold parameterization of β(x) using matrix exponentials- tangent matching loss- Riemannian conjugate gradient optimizationThe system used in the experiments is the nonlinear control-affine system:x1' = x2x2' = -x1 + (x3)2 + u1x3' = x4x4' = -x3 + sin(x1) + u2Training data are generated synthetically by sampling states from a Gaussian distribution and inputs from a bounded random distribution.To reproduce the experiments run: manifolds.py
提供机构：
RepOD
创建时间：
2026-04-02
