Neural networks review for performance enhancement in GPS/INS integration

Global Positioning System (GPS) and Inertial Navigation System (INS) are the most widespread technologies used for navigation information. Each once possesses unique characteristics and boundaries. Therefore, the integration of the two systems defeats both system shortages. In this paper we investigate various neural networks like - the Constructive network (Cascade Correlation Network (CCN) and Feedback Cascade Correlation Network (FBCCN)), Associative memory network with Hebbian rule and Delta rule (Hetero Associative Memory Neural Network (HAM-NN) and Bidirectional Associative Memory Neural Network (BAM-NN)), Higher order network (Sigma-Pi neural network and Pi-Sigma neural network), and the Feed forward network (Back Propagation Neural network (BPN) and Radial Basis Function Neural network (RBFN)) in language of Root Mean Square Error (RMSE), Performance Index (PI), Architecture complexity, Algorithm complexity, Hardware complexity and the number of epochs for GPS/INS data integration.