All news
Digital Twin

Topological Feasibility Guarantees for Differentiable Predictive Control

US & West ·

The paper establishes deterministic feasibility guarantees for differentiable predictive control (DPC), a self-supervised approach for approximating explicit MPC policies. Through topological analysis of the induced reachable safe set, it shows DPC can guarantee feasibility without online safety filters.

Why it matters

Addressing feasibility guarantees for learned control policies extends the applicability of machine learning methods in safety-critical control systems.

Relevant to DT products

Original headline
Topological Feasibility Guarantees for Differentiable Predictive Control
Read the original: arXiv — Systems and Control (eess.SY)

Translation and summary are machine-generated from the source. The full article is not republished; its rights belong to the source.

Related