{"id":"arxiv/2610.05939","url":"https://arxiv.org/abs/2610.05939","tags":["math.OC","cs.CV","cs.RO"],"score":0,"title":"Structural Foundations of Nonlinear Systems with Unknown Inputs: The UID-Induced Normal Form and Minimal-Sensing Structure-from-Motion","domain":"math.OC","source":"arxiv","authors":["Agostino Martinelli"],"reviews":0,"subject":"math.OC","summary":"This paper establishes the first general structural solution to the problem of state estimation for nonlinear systems driven by unknown inputs. Building upon nonlinear unknown-input observability theory, we show that every such system admits a structurally equivalent representation, referred to as the UID-induced normal form. The proposed representation decomposes the information carried by the unknown inputs into two complementary components: unknown-input directions that are structurally decoupled from the observable dynamics and observable quantities that completely represent the…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-05","reactions":{}}