State estimator design using Jordan based long short-term memory networks

• DOI: 10.48550/arxiv.2502.04518
• Publication Date: 2025-02-06
• Authors: Avneet Kaur; Kirsten Morris
• Keywords: Optimization and Control (math.OC); FOS: Mathematics; Mathematics - Optimization and Control

State estimation of a dynamical system refers to estimating the state given an imperfect model, noisy measurements and some or no information about initial state. While Kalman filtering is optimal for linear systems with Gaussian noises, calculation estimators nonlinear challenging. We focus on establishing pathway high-order by using recurrent connections motivated Jordan neural networks(JRNs). The results are compared corresponding Elman structure based long short-term memory network(ELSTM) KF EKF systems. suggest that systems, use networks can improve error also computation time. Also, networks(JLSTMs) require less training achieve performance similar ELSTMs.