Research Article
Terrain Referenced Navigation Using a Multilayer Radial Basis Function-Based Extreme Learning Machine
Algorithm 1
Proposed ML-RBF-ELM algorithm.
(1) Prepare a training set where is the input data with the normalized longitude and latitude and is the normalized terrain height in the DEM | (2) Set the activation function of the 1st hidden layer, , by equation (2) and assign the center and width of the RBF by K-means clustering | (3) Calculate the 1st hidden layer output matrix by equation (3) | (4) Calculate the output weight matrix of the 1st hidden layer, where | (5) Calculate the new input data of the 2nd hidden layer: | (6) Set the activation function of the 2nd hidden layer, , by equation (2) and assign the center and width of the RBF by K-means clustering | (7) Calculate the 2nd hidden layer output matrix by equation (3) | (8) Calculate the output weight matrix of the 2nd hidden layer, where | (9) Calculate the new input data of the 3rd hidden layer: | (10) Set the activation function of the 3rd hidden layer, , by equation (2) and assign the center and width of the RBF by K-means clustering | (11) Calculate the 3rd hidden layer output matrix by equation (3) | (12) Calculate the output weight matrix of the 3rd hidden layer, | (13) Compute and verify the regression results, |
|