Gradient Descent Feed Forward Neural Networks for Forecasting the Trajectories
暂无分享,去创建一个
The paper demonstrates the forecasting of an aircraft trajectory in the vertical plane using gradient descent method for training a feed forward neural network system. For prediction of trajectory a neural networks system has been trained using a set of some arbitrary trajectories and then used to forecast for the new ones. Sliding Window method is being used for predictions, which is able to consider real points during flight to improve the precision in prediction. The results show that neural network can successfully be applied for such predictions.
[1] John J. Burken. Reconfigurable Control with Neural Network Augmentation for a Modified F-15 Aircraft , 2013 .
[2] Carlo L. Bottasso,et al. Neural-Augmented Planning and Tracking Pilots for Maneuvering Multibody Dynamics , 2007 .
[3] Carlo L. Bottasso,et al. Adaptive planning and tracking of trajectories for the simulation of maneuvers with multibody models , 2006 .
[4] Geoffrey E. Hinton,et al. Learning representations by back-propagating errors , 1986, Nature.