LP-Based Identification of general nonlinear multivariable maps: Preliminary results

Mazen Alamir (1)

1 GIPSA-SYSCO - GIPSA - Systèmes non linéaires et complexité

In this paper, a systematic way to approximate general nonlinear maps by means of linear programs (LP) is proposed. The argument of the map to be identified is a parameter of the LP problem while the LP-matrices are computed based on a learning data. Some examples are given to illustrate the proposed algorithm. Note that this is a working document for which several versions will be available. The author would be grateful for any constructive comments on his proposal.

Mots clés: nonlinear identification.; parametric Linear programming

Domaines: Automatique / Robotique

Mazen Alamir. LP-Based Identification of general nonlinear multivariable maps: Preliminary results. [Research Report] GIPSA-lab. 2015. ⟨hal-01131522⟩