A Comparative Study of Machine Learning Techniques for Aviation Applications

Maheshwari, Apoorv; Davendralingam, Navindran; DeLaurentis, Daniel A.

doi:10.2514/6.2018-3980

Cited by 27 publications

(15 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Neural networks are a technique form a mathematical representation of inter connected neurons in systems similar to that of brain. The functions are mathematically expressed as nodes, and are interrelated to produce a complex inputs and outputs web based on the sequence; functions types used and connectivity among the neural nodes of networks to dictate the effectiveness in various machine learning problem forms [20]. In the present work we propose the novel HCNN-LSTM (Health Care Neural Network-Long Short Term Memory based multimodal risk prediction) is used to classify the EHR data in the effective manner.…”

Section: Methodsmentioning

confidence: 99%

Pregnancy Period Diabetic and Blood Pressure Predictive Analysis using HCNN LSTM

Christobel,

Kumar

2020

IJEAT

View full text Add to dashboard Cite

Diabetes has transformed into the worldwide diseases and can occur for all age groups irrespective of their gender. Unlike other diseases, Diabetes needs continuous monitoring as it leads to much adverse effect on functioning of human body. Especially, the diabetes that occurs in female during the pregnancy had its impact over the mother along with their infant before its birth. Many studies showed early prediction can prevent and delimit the challenges that were posed by diabetes among pregnant women. Several health care prediction models often suffer from inconsistencies in data and feature selection that reduce the prediction performance. In the present work, we had proposed the novel Health Care Neural Network-Long Short Term Memory (HCNN-LSTM) to predict the Pregnancy Period Diabetic and Blood Pressure. The Pima Indian diabetes dataset was employed construct the proposed prediction model to predict the patient as diabetic and non-diabetic. For the purpose of comparison, the decision tree, random forest and Navies’ Bayes algorithm are implemented for classification. From the analysis, it was evident that the proposed HCNN-LSTM showed optimum values on performance metrics than the other classifiers. The proposed work can be expanded considering several features of diabetic prediction in future.

show abstract

Section: Methodsmentioning

confidence: 99%

Pregnancy Period Diabetic and Blood Pressure Predictive Analysis using HCNN LSTM

Christobel,

Kumar

2020

IJEAT

View full text Add to dashboard Cite

show abstract

“…The training datasets (D T rain ) are used to train and build the model. Then, the model is evaluated using testing datasets (D T est ) [41], [42].…”

Section: ) Stage 3 -Fdd Techniquementioning

confidence: 99%

“…To classify an observation, the attribute test condition at the root node initially decides the appropriate branch to be followed. Based on the obtained decision, the algorithm continues to another interior node with a new test condition, or to a leaf node associated with the class label to be assigned to the observation [38], [42].…”

Section: ) Decision Tree (Dt) Classifiermentioning

confidence: 99%

Fault Identification of Photovoltaic Array Based on Machine Learning Classifiers

et al. 2021

View full text Add to dashboard Cite

Fault identification in Photovoltaic (PV) array is a contemporary research topic motivated by the higher penetration levels of PV systems in recent electrical grids. Therefore, this work aims to define an optimal Machine learning (ML) structure of automatic detection and diagnosis algorithm for common PV array faults, namely, permanent (Arc Fault, Line-to-Line, Maximum Power Point Tracking unit failure, and Open-Circuit faults), and temporary (Shading) under a wide range of climate datasets, fault impedances, and shading scenarios. To achieve the best-fit ML structure, three distinct ML classifiers are compared, namely, Decision Tree (DT) based on different splitting criteria, K-Nearest Neighbors (KNN) based on the different metrics of distance and weighting functions, and Support Vector Machine (SVM) based on different Kernel functions and multi-classification approaches. Also, Bayesian Optimization is adopted to assign the optimal hyperparameters to the fault classifiers. To investigate the performance of classifiers reported, both simulation and experimental case studies are carried out and presented.INDEX TERMS Photovoltaic array faults, machine learning, decision tree, nearest neighbors, support vector machine, Bayesian optimization.

show abstract

“…Currently, deep learning algorithms are widely applied to aerospace information engineering problems. Moreover, applications involving deep Convolutional Neural Network (CNN) architectures have been demonstrated in many studies, for example, processing satellite images to detect forest-fire hazard areas [4], estimating and forecasting air travel demand [5], determining the crack length in aerospace-grade aluminum samples [6], aircraft maintenance and aircraft health management applications [7], and so on; however, their applications in pose estimation is limited compared with aerospace information applications. In this study, we apply deep learning to solve the problems involved in spacecraft pose estimation.…”

Section: Related Workmentioning

confidence: 99%

Vision-Based Spacecraft Pose Estimation via a Deep Convolutional Neural Network for Noncooperative Docking Operations

et al. 2020

View full text Add to dashboard Cite

The capture of a target spacecraft by a chaser is an on-orbit docking operation that requires an accurate, reliable, and robust object recognition algorithm. Vision-based guided spacecraft relative motion during close-proximity maneuvers has been consecutively applied using dynamic modeling as a spacecraft on-orbit service system. This research constructs a vision-based pose estimation model that performs image processing via a deep convolutional neural network. The pose estimation model was constructed by repurposing a modified pretrained GoogLeNet model with the available Unreal Engine 4 rendered dataset of the Soyuz spacecraft. In the implementation, the convolutional neural network learns from the data samples to create correlations between the images and the spacecraft’s six degrees-of-freedom parameters. The experiment has compared an exponential-based loss function and a weighted Euclidean-based loss function. Using the weighted Euclidean-based loss function, the implemented pose estimation model achieved moderately high performance with a position accuracy of 92.53 percent and an error of 1.2 m. The in-attitude prediction accuracy can reach 87.93 percent, and the errors in the three Euler angles do not exceed 7.6 degrees. This research can contribute to spacecraft detection and tracking problems. Although the finished vision-based model is specific to the environment of synthetic dataset, the model could be trained further to address actual docking operations in the future.

show abstract

A Comparative Study of Machine Learning Techniques for Aviation Applications

Cited by 27 publications

References 13 publications

Pregnancy Period Diabetic and Blood Pressure Predictive Analysis using HCNN LSTM

Pregnancy Period Diabetic and Blood Pressure Predictive Analysis using HCNN LSTM

Fault Identification of Photovoltaic Array Based on Machine Learning Classifiers

Vision-Based Spacecraft Pose Estimation via a Deep Convolutional Neural Network for Noncooperative Docking Operations

Contact Info

Product

Resources

About