In this study, CMAC (Cerebellar ModelArticulated Controller) neural architectures are shown to be viable for the purposes of real-time learning and control. An adaptive critic neuro-control design has been implemented that learns in real-time how to back up a trailer truck along a fixed straight line trajectory. The truck backer-upper experiment is a standard performance measure in the neural network literature, but previously the training of the controllers was done off-line. With the CMAC neural architectures, it was possible to train the neuro-controllers on-line in real-time on a MS-DOS PC 386.
A neural network based approach to model the seismic response of multi-story frame buildings is presented. The seismic response of frames is emulated using multi-layer feedforward neural networks with a backpropagation learning algorithm. Actual earthquake accelerograms and corresponding structural response obtained from analytical models of buildings are used in training the neural networks. The application of the neural network model is demonstrated by studying one to six story high building frames subjected to seismic base excitation. Furthermore, the learning ability of the network is examined for the case of multiple inputs where lateral forces at floor levels are included simultaneously with the base excitation. The effects of the network parameters on learning and accuracy of predictions are discussed. Based on this study, it is found that appropriately configured neural network models can successfully learn and simulate the linear elastic dynamic behavior of multi-story buildings.
This follow-on paper describes the principal methods of implementing, and documents the results of exercising, a set of six-degree-of-freedom rigid-body equations of motion and planetary geodetic, gravitation and atmospheric models for simple vehicles in a variety of endo-and exo-atmospheric conditions with various NASA, and one popular open-source, engineering simulation tools. This effort is intended to provide an additional means of verification of flight simulations. The models used in this comparison, as well as the resulting time-history trajectory data, are available electronically for persons and organizations wishing to compare their flight simulation implementations of the same models.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.