Abstract: Multilayered Perceptrons Neural Networks are used for data assimilation in two non-linear dynamic systems: the H\'enon and Lorenz systems in chaotic state. This aproach "emulates" here the Kalman Filter data assimilation method but does not require recalculation of the gain matrix at each instant of assimilation. In the case of H\'enon system an Adaptive Extended Kalman Filter was used to provide examples for network training. In the case of Lorenz system the Extended Kalman Filter was used for network training. The preliminary results obtained were promising.