Accession Number : ADA558410


Title :   Using Soft Computing Technologies for the Simulation of LCAC Dynamics


Descriptive Note : Conference paper


Corporate Author : NAVAL SURFACE WARFARE CENTER CARDEROCK DIV BETHESDA MD


Personal Author(s) : Fu, Thomas K ; Faller, William E ; Hess, David E


Full Text : http://www.dtic.mil/dtic/tr/fulltext/u2/a558410.pdf


Report Date : Sep 2011


Pagination or Media Count : 9


Abstract : Data acquired from experiments with a 1/6th scale, free-running model of an air-cushioned, amphibious vehicle (LCAC) in waves and calm water were used to train a recursive neural network (RNN). This network is used to simulate the six degree-of-freedom motion of the LCAC providing faster than real-time, time-domain predictions of the vehicle's dynamics as a function of the control signals given by the driver. Results are presented comparing the time-series predictions of the RNN simulation with experimental data. Two error measures are used to quantify the results, an average angle measure and a correlation coefficient, and they indicate good solutions in every case. The intent is to use this time domain simulation of LCAC to gain further insight into the vehicle's dynamics in calm water and irregular waves.


Descriptors :   *AIR CUSHION VEHICLES , AMPHIBIOUS VEHICLES , DYNAMICS , NEURAL NETS , SIMULATION , SYMPOSIA , TIME DOMAIN , TIME SERIES ANALYSIS


Subject Categories : Cybernetics
      Surface Effect Vehicles & Amphibious Vehicles


Distribution Statement : APPROVED FOR PUBLIC RELEASE