Barrett, The Honors College Thesis/Creative Project Collection

One answer to the lack of general knowledge for alternative energy and integration topics is seen in the workforce development content Laboratory of Energy and Power Solutions has generated for the past 6 years. LEAPS is a world-changing organization that provides both technical and business solutions in areas of grid modernization, workforce development, and global energy access that facilitates the global transition to a resilient, low-carbon economy. This paper will aim to explain the contributions of David Hobgood, an Arizona State University senior, to LEAPS workforce development content through the course of the Spring 2022 semester. This paper goes into detail on the process of completing this educational content, amplifies key aspect, and presents the results of a two week pilot that presented the generated content.

Classification in machine learning is quite crucial to solve many problems that the world is presented with today. Therefore, it is key to understand one’s problem and develop an efficient model to achieve a solution. One technique to achieve greater model selection and thus further ease in problem solving is estimation of the Bayes Error Rate. This paper provides the development and analysis of two methods used to estimate the Bayes Error Rate on a given set of data to evaluate performance. The first method takes a “global” approach, looking at the data as a whole, and the second is more “local”—partitioning the data at the outset and then building up to a Bayes Error Estimation of the whole. It is found that one of the methods provides an accurate estimation of the true Bayes Error Rate when the dataset is at high dimension, while the other method provides accurate estimation at large sample size. This second conclusion, in particular, can have significant ramifications on “big data” problems, as one would be able to clarify the distribution with an accurate estimation of the Bayes Error Rate by using this method.

This creative project is a part of the work being done as a Senior Design Project in which an autonomous solar charge controller is being developed. The goal of this project is to design and build a prototype of an autonomous solar charge controller that can work independently of the power grid. This solar charge controller is being built for a community in Monument Valley, Arizona who live off grid. The controller is designed to step down power supplied by an array of solar panels to charge a 48V battery and supply power to an inverter. The charge controller can implement MPPT (Maximum Power Point Tracking) to charge the battery and power the inverter, it also is capable of disconnecting from the battery when the battery is fully charged and reconnecting when it detects that the battery has discharged. The charge controller can also switch from supplying power to the inverter from the panel to supplying power from the battery at low sun or night. These capabilities are not found in solar charge controllers that are on the market. This project aims to achieve all these capabilities and provide a solution for the problems being faced by the current solar charge controller

This project examines the dynamics and design of control systems for a rocket in propulsive ascent and descent using a simplified model with motion constrained to a vertical plane. The governing differential equations are analyzed. They are then linearized, after which transfer functions are derived relating controllable input variables to controlled output variables. The effect of changes in various parameters as well as other aspects of the system are examined. Methods for controller design based on the derived transfer functions are discussed. This will include the discussion of control of the final descent and landing of the rocket. Lastly, there is a brief discussion about both the successes and limitations of the model analyzed.

The idea for this thesis emerged from my senior design capstone project, A Wearable Threat Awareness System. A TFmini-S LiDAR sensor is used as one component of this system; the functionality of and signal processing behind this type of sensor are elucidated in this document. Conceptual implementations of the optical and digital stages of the signal processing is described in some detail. Following an introduction in which some general background knowledge about LiDAR is set forth, the body of the thesis is organized into two main sections. The first section focuses on optical processing to demodulate the received signal backscattered from the target object. This section describes the key steps in demodulation and illustrates them with computer simulation. A series of graphs capture the mathematical form of the signal as it progresses through the optical processing stages, ultimately yielding the baseband envelope which is converted to digital form for estimation of the leading edge of the pulse waveform using a digital algorithm. The next section is on range estimation. It describes the digital algorithm designed to estimate the arrival time of the leading edge of the optical pulse signal. This enables the pulse’s time of flight to be estimated, thus determining the distance between the LiDAR and the target. Performance of this algorithm is assessed with four different levels of noise. A calculation of the error in the leading-edge detection in terms of distance is also included to provide more insight into the algorithm’s accuracy.

This paper discusses the process of creating and testing the haptic feedback wearable that utilizes a sweeping Light Detection and Ranging sensor. This design comes as an extension to the capstone project for electrical engineers. The design works by attaching a LiDAR sensor to a sweeping servo motor, and whenever an object is detected by the sensor, a motor will vibrate to notify the user that an object is nearby. The design incorporates four motors so that the user will have a sense of where an obstacle is coming from and be able to navigate around that obstacle. The design was tested for its accuracy in distance and angle measurement, its efficiency when it came to processing the data, and the uncertainty of the sensor due to beam spreading. Plotting the results for the distance and angle accuracy showed that the design is capable of accurate measurements. The implementation of the code was also very efficient and had no issues with latency when processing the data from the sensor. There was also uncertainty at the larger ranges for the sensor.

In wireless communication systems, the process of data transmission includes the estimation of channels. Implementing machine learning in this process can reduce the amount of time it takes to estimate channels, thus, resulting in an increase of the system’s transmission throughput. This maximizes the performance of applications relating to device-to-device communications and 5G systems. However, applying machine learning algorithms to multi-base-station systems is not well understood in literature, which is the focus of this thesis.

The stability of cheerleading stunts is crucial to athlete safety and team success. Consistency in stunt technique contributes to success in stunting skills, giving a team the tools to win competitions. Increased stunt technique reduces the chances of falls and the severity of those falls. Proper technique also prevents injuries caused by improper positions that place pressure on the lower back and shoulders. Bases must maintain strong technique with proper lines of support in order to maximize stunt stability. Through exploration of the EmbeddedML system, involving a neural network implemented using a SensorTile, cheerleading motions can be successfully classified. Using this system, it is possible to identify motions that result in both weak and injurious positions almost instantly. By alerting athletes to these incorrect motions, improper stunt technique can be corrected quickly and without the involvement of a coach. This automated technique correction would be incredibly beneficial to the sport of competitive cheerleading