This work is funded by a NASA science and technology graduate research opportunity. It involves developing advanced non-linear estimation and optimization techniques for determining attitude from light curves. In particular, I developed a novel global optimization approache to attitude estimation using particle swarm optimization, and have also considered evolutionary algorithms, simulated annealing, and other global optimization algorithms. In addition, I have investigated a variety of non-linear attitude filtering approaches, including conventional MEKF and MUKF filtering, as well as more advanced techniques such as gaussian mixture filtering, koopman operator based filtering, and particle filtering.

This work was funded by the Air Force Office of Scientific Research. Helped develop a conceptual spaces approach to fusing hard and soft data for space threat assessment. This involved combining information about capabilities and state (orbit, maneuvering capabilities, payload etc.) with information about the political and global context of the situation (intelligence reports, news, analyst opinion etc.) to create a wholistic assessment of the current and future threat levels associated with particular objects in space. This work was done collaboratively in a multi-disciple team.

Surveyed a number of topics at the lab in order to assess the applicability of data fusion techniques to those fields, and worked on analyzing solar wind data for a space science group. Data analysis done using Fourier analysis, as well as correlation analysis. (Re)discovered a 2 day lag in the correlation between solar wind and high energy electrons in the upper atmosphere.

Interned at the Aerothermal / Aero-Optic Evaluation Center run by CUBRC. Developed software tools for analysis of pressure and temperature data from hypersonic flows. Created a graphic user interface in MATLAB that performed Short Time Fourier Transforms on data in order to display the energy variation over both frequency and time. This GUI included user-friendly IO capabilities for reading the appropriate data files and saving the desired plots, as well as features for generating several different plots (including full FFT plots and frequency response graphs at selected times), data cleaning capabilities (outlier removal, manual data pruning etc.), and batch processing of multiple data sets. This tool replaced an aged C script that required substantial user knowledge to operate, limiting its usefulness. The development of this tool allows for anyone to process data with minimal training.Designed experiments to investigate the used of LED lights as a replacement for the incandescent flash bulbs used to illuminate test sections in the shock tunnels. Created a test setup, and a testing procedure, and then ran experiments and recorded data. Determined the available LEDs were not capable of producing equivalent lighting conditions to the conventional incandescent lighting system.

Designed and built a test rig for dye visualization of flow structures around a morphing wing geometry. This involved designing and actuating the morphing wing, implementing a dye injection system, and selecting appropriate locations for dye tubes to effectively visualize the critical flow structures.

Analyzed the frequency response of a periodically structured one dimensional rod, and represented the results visually. Wrote code in MATLAB to reproduce and confirm theoretical results generated by Dr. Nouh and his graduate students, and helped create animations in MATLAB to visually display these results. Simulated a two dimension plate with periodic locally resonant structures in ANSYS subject to periodic forcing in order to asses the displacement response. Confirmed analytical work demonstrating that periodically distributed piezo- electric cells embedded in a structure can be used to actively cancel vibrations.