Deep learning based state estimation and modeling dynamic system using Kalman filter

摘要

In recent trends predicting and modeling a time varying system is necessary in order to reduce the processing time. State estimation and prediction of the dynamic system is successfully implemented using Kalman filter. Deep learning is one of the promising technologies which produce an improvement in accuracy, reduction in processing time after sufficient training. Hence the performance of the state estimation and modeling the system can be improved by applying deep learning along with existing method. In our paper, we are proposing a method for implementation of existing state estimation along with deep learning architecture. In general our proposed can be applied in time series forecasting, network routing, communication and image processing applications.