Research Article

GNSS/Low-Cost MEMS-INS Integration Using Variational Bayesian Adaptive Cubature Kalman Smoother and Ensemble Regularized ELM

Algorithm 1

RELM.
Input: Samples , number of hidden
   node
Output:
(1)  Step  1. Randomly generate the input
   weights and the bias value
(2) Step  2. Calculate the hidden output matrix
   using (28)–(30)
(3) Step  3. Calculate the output weight value ,
   
   where is the Moore-Penrose generalized
   inverse of matrix and .
(4) Step  4. Apply regularization of ELM using
   (31)–(34).