A geometric FDI method for nonlinear control-affine systems combines subspace angles for isolability with mirror descent adaptation of neural network layers to achieve uniformly ultimately bounded state and fault estimation errors.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SY 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Geometric Fault Identification via Mirror Descent Learning
A geometric FDI method for nonlinear control-affine systems combines subspace angles for isolability with mirror descent adaptation of neural network layers to achieve uniformly ultimately bounded state and fault estimation errors.