Flight data analysis of the HyShot 2 scramjet flight experiment

The development of scramjet propulsion for alternative launch and payload delivery capabilities has been composed largely of ground experiments for the last 40 years. With the goal of validating the use of short duration ground test facilities, a ballistic reentry vehicle experiment called HyShot was devised to achieve supersonic combustion in flight above Mach 7.5. It consisted of a double wedge intake and two back-to-back constant area combustors; one supplied with hydrogen fuel at an equivalence ratio of 0.34 and the other unfueled. Of the two flights conducted, HyShot 1 failed to reach the desired altitude due to booster failure, whereas HyShot 2 successfully accomplished both the desired trajectory and satisfactory scramjet operation. Postflight data analysis of HyShot 2 confirmed the presence of supersonic combustion during the approximately 3 s test window at altitudes between 35 and 29 km. Reasonable correlation between flight and some preflight shock tunnel tests was observed.

Tracking the states of a nonlinear and nonstationary system in the weight-space of artificial neural networks

We propose a novel interpretation and usage of Neural Network (NN) in modeling physiological signals, which are allowed to be nonlinear and/or nonstationary. The method consists of training a NN for the k-step prediction of a physiological signal, and then examining the connection-weight-space (CWS) of the NN to extract information about the signal generator mechanism. We de. ne a novel feature, Normalized Vector Separation (gamma(ij)), to measure the separation of two arbitrary states i and j in the CWS and use it to track the state changes of the generating system. The performance of the method is examined via synthetic signals and clinical EEG. Synthetic data indicates that gamma(ij) can track the system down to a SNR of 3.5 dB. Clinical data obtained from three patients undergoing carotid endarterectomy of the brain showed that EEG could be modeled (within a root-means-squared-error of 0.01) by the proposed method, and the blood perfusion state of the brain could be monitored via gamma(ij), with small NNs having no more than 21 connection weight altogether.

Training selective use of a complex categorisation strategy

The present study aims to encourage selective use of a complex categorisation strategy. More specifically, participants will be trained to use a two dimensional strategy in one region of category space and a more complex three-dimensional strategy in another region of category space. In the 2–3 conditions, participants will be presented with stimuli requiring the two-dimensional strategy in the first phase of training and the three-dimensional strategy in the second phase of training. In the 3-2 conditions, participants will be presented with stimuli requiring the three-dimensional strategy in the first phase of training and the two-dimensional strategy in the second phase of training. The main dependent measure will be performance on exceptions to the two-dimensional strategy. If participants learn to selectively use the three-dimensional strategy, then we expect them to correctly classify novel exceptions that occur in the three-dimensional region of the category space and incorrectly classify novel exceptions that occur in the two-dimensional region of the category space.

Improving support vector solutions by selecting a sequence of training subsets

In this paper we demonstrate that it is possible to gradually improve the performance of support vector machine (SVM) classifiers by using a genetic algorithm to select a sequence of training subsets from the available data. Performance improvement is possible because the SVM solution generally lies some distance away from the Bayes optimal in the space of learning parameters. We illustrate performance improvements on a number of benchmark data sets.

Tracking the states of a nonlinear system in the weight-space of a feed-forward neural network

Nonlinear, non-stationary signals are commonly found in a variety of disciplines such as biology, medicine, geology and financial modeling. The complexity (e.g. nonlinearity and non-stationarity) of such signals and their low signal to noise ratios often make it a challenging task to use them in critical applications. In this paper we propose a new neural network based technique to address those problems. We show that a feed forward, multi-layered neural network can conveniently capture the states of a nonlinear system in its connection weight-space, after a process of supervised training. The performance of the proposed method is investigated via computer simulations.