Engineering Commons^™

Using Convolutional Neural Networks For Autonomous Drone Navigation, Joshua Jowers

Industrial Engineering Undergraduate Honors Theses

Unmanned Aerial Vehicles (UAVs), more commonly known as drones, serve various purposes, notably in military applications. Consequently, there arises a need for navigation methods impervious to intercepted signals [1]. Previous research has explored numerous solutions, including machine learning. This paper delves into a specific machine learning approach employing a Convolutional Neural Network (CNN) to discern image locations [2]. It elucidates the conversion of a CNN model between two machine learning libraries and presents results from multiple experiments examining parameters and factors influencing the approach's efficacy. These experiments encompass testing different data sources, image quantities, and processing pipelines to gauge their …

Artificial Intelligence-Assisted Inertial Geomagnetic Passive Navigation, Andrei Cuenca

Doctoral Dissertations and Master's Theses

In recent years, the integration of machine learning techniques into navigation systems has garnered significant interest due to their potential to improve estimation accuracy and system robustness. This doctoral dissertation investigates the use of Deep Learning combined with a Rao-Blackwellized Particle Filter for enhancing geomagnetic navigation in airborne simulated missions.

A simulation framework is developed to facilitate the evaluation of the proposed navigation system. This framework includes a detailed aircraft model, a mathematical representation of the Earth's magnetic field, and the incorporation of real-world magnetic field data obtained from online databases. The setup allows an accurate assessment of the performance …

Predicting Dynamic Fragmentation Characteristics From High-Impact Energy Events Utilizing Terrestrial Static Arena Test Data And Machine Learning, Katharine Larsen, Riccardo Bevilacqua, Omkar S. Mulekar, Elisabetta L. Jerome, Thomas J. Hatch-Aguilar

Student Works

To continue space operations with the increasing space debris, accurate characterization of fragment fly-out properties from hypervelocity impacts is essential. However, with limited realistic experimentation and the need for data, available static arena test data, collected utilizing a novel stereoscopic imaging technique, is the primary dataset for this paper. This research leverages machine learning methodologies to predict fragmentation characteristics using combined data from this imaging technique and simulations, produced considering dynamic impact conditions. Gaussian mixture models (GMMs), fit via expectation maximization (EM), are used to model fragment track intersections on a defined surface of intersection. After modeling the fragment distributions, …

Neural Network Models For Generating Synthetic Flight Data, Nathaniel Sisson

Doctoral Dissertations and Master's Theses

Flight test data is a valuable resource used in many aerospace applications. However, procuring a sufficiently large database of flight test data poses several challenges. Nominal flight tests can be expensive and time-consuming and require much post-processing depending on the availability of sensors and the quality of the sensor output. Flight test performed outside of nominal flight conditions, or flight tests in which failures are introduced, add to the inherent risk and danger associated with flight tests. The most popular alternative to flight test, numerical simulations, may fail to fully capture all non-linear behavior. While flight tests will always be …

Machine Learning To Predict Warhead Fragmentation In-Flight Behavior From Static Data, Katharine Larsen

Doctoral Dissertations and Master's Theses

Accurate characterization of fragment fly-out properties from high-speed warhead detonations is essential for estimation of collateral damage and lethality for a given weapon. Real warhead dynamic detonation tests are rare, costly, and often unrealizable with current technology, leaving fragmentation experiments limited to static arena tests and numerical simulations. Stereoscopic imaging techniques can now provide static arena tests with time-dependent tracks of individual fragments, each with characteristics such as fragment IDs and their respective position vector. Simulation methods can account for the dynamic case but can exclude relevant dynamics experienced in real-life warhead detonations. This research leverages machine learning methodologies to …

Machine Learning In Aerodynamic Shape Optimization, Jichao Li, Xiaosong Du, Joaquim R.R.A. Martins

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Machine learning (ML) has been increasingly used to aid aerodynamic shape optimization (ASO), thanks to the availability of aerodynamic data and continued developments in deep learning. We review the applications of ML in ASO to date and provide a perspective on the state-of-the-art and future directions. We first introduce conventional ASO and current challenges. Next, we introduce ML fundamentals and detail ML algorithms that have been successful in ASO. Then, we review ML applications to ASO addressing three aspects: compact geometric design space, fast aerodynamic analysis, and efficient optimization architecture. In addition to providing a comprehensive summary of the research, …

Air Passenger Demand Forecast Through The Use Of Artificial Neural Network Algorithms, Juan Gerardo Muros Anguita, Oscar Díaz Olariaga

International Journal of Aviation, Aeronautics, and Aerospace

Airport planning depends to a large extent on the levels of activity that are anticipated. In order to plan facilities and infrastructures of an airport system and to be able to satisfy future needs, it is essential to predict the level and distribution of demand. This document presents a short- and medium-term forecast of the demand for air passengers carried out through a specific case study (Colombia), in which the impact of the pandemic period due to COVID-19 on air traffic was taken into account. To make the forecast, an algorithm that implements techniques based on Artificial Neural Networks (ANN) …

Air Passenger Demand Forecast Through The Use Of Artificial Neural Network Algorithms, Juan Gerardo Muros Anguita, Oscar Díaz Olariaga

International Journal of Aviation, Aeronautics, and Aerospace

Airport planning depends to a large extent on the levels of activity that are anticipated. To plan the facilities and infrastructures of an airport system and to be able to satisfy future needs, it is essential to predict the level and distribution of demand. This document presents a short- and medium-term forecast of the demand for air passengers carried out through a specific case study (Colombia), in which the impact of the pandemic period due to COVID-19 on air traffic was taken into account. To make the forecast, an algorithm that implements techniques based on Artificial Neural Networks (ANN) (Machine …

Numerical Modeling Of Advanced Propulsion Systems, Peetak P. Mitra

Doctoral Dissertations

Numerical modeling of advanced propulsion systems such as the Internal Combustion Engine (ICE) is of great interest to the community due to the magnitude of compute/algorithmic challenges. Fuel spray atomization, which determines the rate of fuel-air mixing, is a critical limiting process for the phenomena of combustion within ICEs. Fuel spray atomization has proven to be a formidable challenge for the state-of-the-art numerical models due to its highly transient, multi-scale, and multi-phase nature. Current models for primary atomization employ a high degree of empiricism in the form of model constants. This level of empiricism often reduces the art of predictive …

Reaction Wheels Fault Isolation Onboard 3-Axis Controlled Satellite Using Enhanced Random Forest With Multidomain Features, Mofiyinoluwa Oluwatobi Folami

Electronic Theses and Dissertations

With the increasing number of satellite launches throughout the years, it is only natural that an interest in the safety and monitoring of these systems would increase as well. However, as a system becomes more complex it becomes difficult to generate a high-fidelity model that accurately describes all the system components. With such constraints using data-driven approaches becomes a more feasible option. One of the most commonly used actuators in spacecraft is known as the reaction wheel. If these reaction wheels are not maintained or monitored, it could result in mission failure and unwarranted costs. That is why fault detection …

Soarnet, Deep Learning Thermal Detection For Free Flight, Jake T. Tallman

Master's Theses

Thermals are regions of rising hot air formed on the ground through the warming of the surface by the sun. Thermals are commonly used by birds and glider pilots to extend flight duration, increase cross-country distance, and conserve energy. This kind of powerless flight using natural sources of lift is called soaring. Once a thermal is encountered, the pilot flies in circles to keep within the thermal, so gaining altitude before flying off to the next thermal and towards the destination. A single thermal can net a pilot thousands of feet of elevation gain, however estimating thermal locations is not …

Deep Reinforcement Learning Applied To Spacecraft Attitude Control And Moment Of Inertia Estimation Via Recurrent Neural Networks, Nathaniel A. Enders

Theses and Dissertations

This study investigated two distinct problems related to unknown spacecraft inertia. The first problem explored the use of a recurrent neural network to estimate spacecraft moments of inertia using angular velocity measurements. Initial results showed that, for the configuration examined, the neural network can estimate the moments of inertia when there is a known external torque. The second problem trained a reinforcement learning agent, via proximal policy optimization, to control the attitude of a spacecraft. The results demonstrated that reinforcement learning may be a viable option for guidance and control solutions where the spacecraft model may be unknown. The trained …

Next-Generation Re-Entry Aerothermodynamic Modeling Of Space Debris Using Machine Learning, Nicholas Sia

Graduate Theses, Dissertations, and Problem Reports

The number of resident space objects re-entering the atmosphere is expected to rise with increased space activity over recent years and future projections. Predicting the survival and impact location of the medium to large sized re-entering objects becomes important as they can cause on ground casualties and damage to property. Uncertainties associated with the re-entry process makes necessary a probabilistic approach, which can be computationally expensive when using high-fidelity numerical methods for estimating aerothermodynamic properties. To date, object-oriented analysis is the dominant tool used for atmospheric re-entry modeling and simulation, where aerothermodynamic coefficients are used to determine the risk a …

Predictability Improvement Of Scheduled Flights Departure Time Variation Using Supervised Machine Learning, Deepudev Sahadevan, Palanisamy Ponnusamy Dr, Manjunath K. Nelli Mr, Varun P. Gopi Dr

International Journal of Aviation, Aeronautics, and Aerospace

The departure time uncertainty exacerbates the inaccuracy of arrival time estimation and demand for arrival slots, particularly for movements to capacity constrained airports. The Estimated Take-Off Time (ETOT) or Estimated Departure Time(ETD) for each individual flight is currently derived from Air Traffic Flow Management System (ATFMS), which are solely determined based on individual flight plan Estimated Off Block Time(EOBT) or subsequent delays updated by Airline. Even if normal weather conditions prevail, aircraft departure times will differ from ETOTs determined by the ATFMS due to a number of factors such as congestion, early/delayed inbound flight (linked flights), reactionary delays and air …

Terramechanics And Machine Learning For The Characterization Of Terrain, Bryan W. Southwell

Electronic Thesis and Dissertation Repository

An instrumented rover wheel can collect vast amounts of data about a planetary surface. Planetary surfaces are changed by complex geological processes which can be better understood with an abundance of surface data and the use of terramechanics. Identifying terrain parameters such as cohesion and angle of friction hold importance for both the rover driver and the planetary scientist. Knowledge of terrain characteristics can warn of unsafe terrain and flag potential interesting scientific sites. The instrumented wheel in this research utilizes a pressure pad to sense load and sinkage, a string potentiometer to measure slip, and records motor current draw. …

Machine Learning Classification Of Interplanetary Coronal Mass Ejections Using Satellite Accelerometers, Kelsey Doerksen

Electronic Thesis and Dissertation Repository

Space weather phenomena is a complex area of research as there are many different variables and signatures that are used to identify the occurrence of solar storms and Interplanetary Coronal Mass Ejections (ICMEs), with inconsistencies between databases and solar storm catalogues. The identification of space weather events is important from a satellite operation point of view, as strong geomagnetic storms can cause orbit perturbations to satellites in low-earth orbit. The Disturbance storm time (Dst) and the Planetary K-index (Kp) are common indices used to identify the occurrence of geomagnetic storms caused by ICMEs, among several other signatures that are not …

Visualization And Machine Learning Techniques For Nasa’S Em-1 Big Data Problem, Antonio P. Garza Iii, Jose Quinonez, Misael Santana, Nibhrat Lohia

SMU Data Science Review

In this paper, we help NASA solve three Exploration Mission-1 (EM-1) challenges: data storage, computation time, and visualization of complex data. NASA is studying one year of trajectory data to determine available launch opportunities (about 90TBs of data). We improve data storage by introducing a cloud-based solution that provides elasticity and server upgrades. This migration will save $120k in infrastructure costs every four years, and potentially avoid schedule slips. Additionally, it increases computational efficiency by 125%. We further enhance computation via machine learning techniques that use the classic orbital elements to predict valid trajectories. Our machine learning model decreases trajectory …