Driver distraction is a major contributor to car crashes, and visual distraction caused by using invehicle infotainment systems (IVIS) degrades driving performance and increases crash risk. Air gesture interfaces were developed to mitigate for driver distraction, and using auditory displays showed a decrease in off-road glances and an improved perceived workload. However, the design of auditory displays was not fully investigated. This thesis presents directional research in the design of auditory displays for air-gesture IVIS through two dual-task experiments of driving a simulator and air-gesture menu navigation. Experiment 1 with 32 participants employed a 2x4 mixed-model design, and explored the effect of four auditory display conditions (auditory icon, earcon, spearcon, and no-sound) and two menu-generation interfaces (fixed and adaptive) on driving performance, eye glance behavior, secondary task performance and subjective perception. Each auditory display (within-subjects) was tested with both a fixed and adaptive menu-generation interface (between-subjects). Results from Experiment 1 demonstrated that spearcon provided the least visual distraction, least workload, best system usability and was favored by participants; and that fixed menu generation outperformed adaptive menu generation in driving safety and secondary task performance. Experiment 2 with 24 participants utilized the best interface to emerge from Experiment 1 to further explore the auditory display with the most potential: spearcon. 70% spearcon and 40% spearcon were compared to text-to-speech (TTS) and no audio conditions. Results from Experiment 2 showed that 70% spearcon induced less visual distraction than 40% spearcon, and that 70% spearcon resulted in the most accurate but slowest secondary task selections. Experimental results are discussed in the context of the multiple resource theory and the working memory model, design guidelines are proposed, and future work is discussed. / Master of Science / Driver distraction is a major cause of car accidents, and using in-vehicle infotainment systems (IVIS) while driving can distract drivers and increase the risk of crashes. Air gesture interfaces and auditory displays were created to help reduce driver distraction, and using auditory displays has been shown to decrease the number of times a driver looks away from the road and to improve the perceived workload of the driver. However, the design of auditory displays has not been thoroughly studied. This study examined the design of auditory displays for air gesture IVIS through two experiments in which participants drove a simulator and used air gesture menus while navigating. The first experiment, which included 32 participants, looked at the effect of four different types of auditory displays (auditory icon, earcon, spearcon, and no sound) and two different types of menu-generation interfaces (fixed and adaptive) on driving performance, eye glance behavior, secondary task performance, and subjective perception. The second experiment, which included 24 participants, compared the use of 70% and 40% spearcon displays to text-to-speech and no audio conditions. The results of these experiments showed that using spearcon displays resulted in the least visual distraction and workload, the best system usability, and the most accurate but slowest secondary task selections. These findings are discussed in relation to existing theories of how the brain processes multiple tasks, and design guidelines for auditory displays are proposed for future research.

The SINDy (Sparse Identification of Non-linear Dynamics) algorithm is a method of turning a set of data representing non-linear dynamics into a much smaller set of equations comprised of non-linear functions summed together. This provides a human readable system model the represents the dynamic system analyzed. The SINDy algorithm is important for a variety of applications, including high precision industrial and robotic applications. A Hardware Accelerator was designed to decrease the time spent doing calculations. This thesis proposes an efficient hardware accelerator approach for a broad range of applications that use SINDy and similar system identification algorithms. The accelerator is leverages both systolic arrays for integrated neural network models with other numerical solvers. The novel and efficient reuse of similar processing elements allows this approach to only use a minimal footprint, so that it could be added to microcontroller devices or implemented on lower cost FPGA devices. Our proposed approach also allows the designer to offload calculations onto edge devices from controller nodes and requires less communication from those edge devices to the controller due to the reduced equation space.

Soil organic matter (SOM) is the largest terrestrial pool of organic carbon (C), and potential carbon-climate feedbacks involving SOM decomposition could exacerbate anthropogenic climate change. Despite the importance of SOM in the global C cycle, our understanding of the controls on SOM stabilization and decomposition is still developing, and as such, SOM dynamics are a source of major uncertainty in current Earth system models (ESMs), which reduces the effectiveness of these models in predicting the efficacy of climate change mitigation strategies. To improve our understanding of controls on SOM decomposition at scales relevant to such modeling efforts, A and upper B horizon soil samples from 22 National Ecological Observatory Network (NEON) sites spanning the conterminous U.S. were incubated for 52 weeks under conditions representing site-specific mean summer temperature and horizon-specific field capacity (-33 kPa) water potential. Cumulative CO2 respired was periodically measured and normalized by soil organic C content to obtain cumulative specific respiration (CSR). A two-pool decomposition model was fitted to the CSR data to calculate decomposition rates of fast- (kfast) and slow-cycling pools (kslow). Post-LASSO best subsets multiple linear regression was used to construct horizon-specific models of significant predictors for CSR, kfast, and kslow. Significant predictors for all three response variables consisted mostly of proximal factors related to clay-sized fraction mineralogy and SOM composition. Non-crystalline minerals and lower SOM lability negatively affected CSR for both A and B horizons. Significant predictors for decomposition rates varied by horizon and pool. B horizon decomposition rates were positively influenced by nitrogen (N) availability, while an index of pyrogenic C had a negative effect on kfast in both horizons. These results reinforce the recognized need to explicitly represent SOM stabilization via interactions with non-crystalline minerals in ESMs, and they also suggest that increased N inputs could enhance SOM decomposition in the subsoil, highlighting another mechanism beyond shifts in temperature and precipitation regimes that could alter SOM decomposition rates. / Master of Science / Soils contain a large amount of carbon (C) in the form of soil organic matter (SOM), and there is the potential for the increased decomposition of SOM due to warmer temperatures to cause climate change to become worse through the release of additional CO₂ into the atmosphere. However, we still do not know exactly what is most important for predicting how vulnerable SOM is to decomposition at continental scales, and this results in a substantial amount of uncertainty in Earth system models used to predict climate change. To address this question, the proportion of organic C decomposed in soil samples from the topsoil and subsoil from 22 sites across the conterminous U.S. was monitored over the course of a year under optimal moisture conditions and at site-specific summer temperature. Additionally, a mathematical model was fitted to the proportion of organic C decomposed over time to estimate decomposition rates of a quickly decomposing pool of SOM and a slowly decomposing pool of SOM. The proportion of organic C decomposed and decomposition rates were related to soil and site properties using multiple linear regression to find which soil and site properties were most important for predicting these response variables. The type of clay-sized mineral and SOM chemical composition were found to be important predictors of the proportion of organic C decomposed for both topsoil and subsoil samples. The important predictors for decomposition rates varied by pool and by topsoil vs. subsoil. For subsoil decomposition rates, it was found that a greater availability of nitrogen (N) increased decomposition rates, and in the quickly decomposing pool, it was found that fire-derived organic matter slowed decomposition rates. The results of this study showed the general importance of local factors for controlling SOM decomposition. Specifically, it showed that the type of clay-sized mineral present at a site needs to be considered as well as the fact that N might increase SOM decomposition in the subsoil.

This study describes the development of the Propulsion Supervisory Controller for an independent axle All-Wheel Drive Electric Vehicle, using a model-based approach. The vehicle has a main rear motor and a smaller front motor. Features like power moding, transmission range selection and torque architecture are discussed. For the torque architecture, different torque distribution strategies are explored in detail. Initially, a comparison of torque distribution strategies considering positive torques only, is used to assess the impact on the vehicle's energy consumption. Firstly, an optimal strategy with and without power-rate penalties is explored, which distributes the torque request to minimize the losses in both drive-units. Secondly, a fixed-ratio strategy is considered where both axles contribute with a predetermined torque ratio to meet the total torque demand. Thirdly, a torque-assist approach is examined, wherein only the rear motor contributes to the torque demand till it is operating at instantaneous maximum torque, after which the front motor starts contributing. Similar evaluations are then performed including regenerative braking or negative torque domain. Additionally, the performance of the penalized optimal strategy (PO) for positive torques is evaluated when combined with the torque assist regenerative braking strategy, where the front motor is primarily used for regenerative braking. The performance of PO combined with the ideal regenerative braking strategy is also assessed. This study aims to provide an overview of the controller development approach and an insight of the feasibility of deploying sophisticated computational algorithms for enhanced efficiency on it. / Master of Science / This study focuses on the development of a propulsion controller for a modified all-electric 2023 Cadillac LYRIQ. The Sport Utility Vehicle (SUV) is equipped with a main rear motor and a smaller front motor. Functional features such as the power-up and power-down sequence and vehicle range selection are discussed along with performance features like torque control. The objective is to enable safe vehicle functionality and enhance the vehicle's powertrain efficiency through the development of software for its Propulsion Supervisory Controller (PSC). The study initially evaluates various strategies for distributing torque during forward acceleration. Three primary strategies are analyzed: an optimal approach aimed at minimizing overall energy losses, a fixed-ratio strategy where torque ratios are predetermined to meet the total demand, and a torque-assist method where the front motor provides torque only after the rear motor reaches its instantaneous maximum torque, triggered by the accelerator pedal input exceeding a threshold. Similarly, these strategies are examined within the context of regenerative braking to assess their impact on range. Finally, the penalized optimal torque distribution strategy is combined with a torque assist regenerative braking strategy as well as a strategy that adheres to the ideal braking distribution. This study provides an overview of the vehicle controller development and demonstrates the feasibility and benefits of employing advanced computational algorithms in the propulsion controller to achieve enhanced efficiency and an improved range in electric vehicles.

With an increase in age comes an increase in the risk of frailty and mobility decline, which can lead to dangerous falls and can even be a cause of mortality. Despite these serious consequences, healthcare systems remain reactive, highlighting the need for technologies to predict functional mobility decline. In this thesis, we present an end-to-end autonomous functional mobility assessment system that seeks to bridge the gap between robotics research and clinical rehabilitation practices. Unlike many fully integrated black-box models, our approach emphasizes the need for a system that is both reliable as well as transparent to facilitate its endorsement and adoption by healthcare professionals and patients. Our proposed system is characterized by the sensor fusion of multimodal data using an optimization framework known as factor graphs. This method, widely used in robotics, enables us to obtain visually interpretable 3-D estimations of the human body in recorded footage. These representations are then used to implement autonomous versions of standardized assessments employed by physical therapists for measuring lower-limb mobility, using a combination of custom neural networks and explainable models. To improve the accuracy of the estimations, we investigate the application of the Koopman operator framework to learn linear representations of human dynamics: We leverage these outputs as prior information to enhance the temporal consistency across entire movement sequences. Furthermore, inspired by the inherent stability of natural human movement, we propose ways to impose stability constraints in the dynamics during the training of linear Koopman models. In this light, we propose a sufficient condition for the stability of discrete-time linear systems that can be represented as a set of convex constraints. Additionally, we demonstrate how it can be seamlessly integrated into larger-scale gradient descent optimization methods. Lastly, we report the performance of our human pose detection and autonomous mobility assessment systems by evaluating them on outcome mobility datasets collected from controlled laboratory settings and unconstrained real-life home environments. While we acknowledge that further research is still needed, the study results indicate that the system can demonstrate promising performance in assessing mobility in home environments. These findings underscore the significant potential of this and similar technologies to revolutionize physical therapy practices.

A comparison of two methods to deal with allocating controls for unattainable moments in an aircraft was performed using a testbed airframe that resembled an F/A-18 with a large control effector suite. The method of preserving the desired moment direction to deal with unattainable moments is currently used in a specific control allocator. A new method of prioritizing the pitch axis is compared to the moment-direction preservation. Realtime piloted simulations are completed to evaluate the characteristics and performance of these methods. A direct comparison between the method of preserving the moment direction by scaling the control solution vector and prioritizing the pitching moment axis is performed for a specific case. Representative maneuvers are flown with a highly unstable airframe to evaluate the ability to achieve the specific task. Flight performance and pilot interpretation are used to evaluate the two methods. Pilot comments and performance results favored the method of pitch-axis prioritization. This method provided favorable flight characteristics compared to the alternative method of preserving the moment direction for the specific tasks detailed in this paper. NOTE: An updated copy of this ETD was added on 09/28/2010. / Master of Science

<p dir="ltr">This study explores the use of hand warmers to enhance hermetic storage. Hand warmers work similarly to conventional oxygen scavengers by consuming oxygen through oxidation. This research aimed to determine how effective hand warmers are in accelerating oxygen reduction and insect mortality in hermetic storage, hence preserving grain quality. The experiments conducted in hermetic containers showed that hand warmers rapidly reduce oxygen levels, achieving insect-lethal conditions much faster than conventional oxygen absorbers like Oxy-Sorb. Our results show that hand warmers brought oxygen levels down to below 5% within 24-48 h and kept them low for an extended period up to 240 h. This quick oxygen depletion led to high insect mortality, reduced oviposition, and inhibited progeny development. Additionally, hand warmers did not negatively impact seed moisture content or germination rates, demonstrating their effectiveness in maintaining grain quality during storage.</p>

As required by the Corporate Sustainability Reporting Directive (CSRD) of the European Union, this thesis thoroughly examines the transition in corporate sustainability reporting from limited to reasonable assurance. The directive is being implemented at a crucial time as companies are progressively incorporating environmental, social and governance (ESG) factors into their core operating and reporting practices. The CSRD is intended to improve sustainability disclosures’ comparability, credibility and reliability. It is a legislative response to a larger movement for more corporate accountability and transparency. The study uses a qualitative technique using semi-structured interviews with professionals from leading audit firms. The purpose of these interviews is to provide many aspects of the transition with a particular emphasis on how businesses and audit firms are modifying their operations to comply with the stricter requirements of the CSRD. The study identifies the main factors behind this shift such as increased stakeholder demands for transparency and a changing regulatory environment that makes it necessary to reevaluate current assurance practices.  The study’s important conclusion is that putting reasonable assurance requirements into practice is difficult. Companies must create advanced data management systems that can handle the CSRD’s requirements for an increasing volume of information. In addition to ensuring data accuracy, these systems need to make it easier to analyse and report on the data in a way that complies with the new assurance standards.  Additionally, the shift needs a major improvement in auditor competencies. It is necessary for auditors to have a better understanding of both traditional financial auditing and the specific challenges presented by sustainability reporting. This includes having a thorough understanding of ESG factors and how they affect risk profiles and corporate performance. Hence, the CSRD acts as a catalyst for a significant shift in auditors’ professional development and training, emphasising the necessity to combine sustainability with financial auditing skills.  Strong internal controls are also important, as the research shows. Establishing and maintaining strict internal mechanisms is necessary for businesses to ensure the reliability of their sustainability reports. This involves a thorough review and a redesign of internal processes in order to meet the higher standards of reasonable assurance. These controls are essential for reducing the risks associated with sustainability reporting such errors or misrepresentations which can have a big influence on stakeholder trust and regulatory compliance.  This change has significant implications for society at large. Companies may give stakeholders more reliable and detailed disclosures about their sustainable practices by shifting toward reasonable assurance. This can have a big impact on investment choices and build stakeholder confidence. This shift promotes a more transparent and sustainable corporate environment by supporting the incorporation of ESG factors into core business strategies and by strengthening the credibility of sustainability reports.

As the use of small satellites for advanced space missions continues to grow, the importance of low mass and cost three-axis attitude stabilization systems increases as well, with these systems requiring high accuracy attitude knowledge. Star trackers provide the most accurate attitude knowledge of any type of attitude sensor, but the high cost, size, and weight of commercial star trackers can be prohibitive to small satellite missions. Many simple star trackers have been developed using commercial off-the-shelf camera sensors and processing hardware, but the challenge remains in testing and characterizing these devices. A common solution is night sky tests, in which the star tracker is held up to the night sky to image the star field and perform attitude determination. Commercial star trackers, on the other hand, are regularly tested with manufacturer provided star field images that attach directly to the sensor. These methods, however, severely limit the sky conditions that can be used in testing. Night sky tests depend on weather and can only image regions of the sky the user has access to, while lab-based testing uses the few provided still images. This thesis presents a hardware-in-the-loop star tracker test bed developed for comprehensive ground-based testing of both in-house and commercial star trackers. The system consists of a small screen to display a star field, a simple in-house camera star tracker, and a microprocessor. This test bed allows any star field image to be simulated. The system is set up for use on a stationary tabletop, but its small size lends itself for use with a spacecraft dynamics platform, which can facilitate testing of control algorithms using real star tracker output.

Semi-active suspensions have drawn particular attention due to their superior performance over the other types of suspensions. One of their advantages is that their damping coefficient can be controlled without the need for any external source of power. In this study, a handful of control approaches are implemented on a car models using MATLAB/Simulink. The investigated control methodologies are skyhook, groundhook, hybrid skyhook-groundhook, Acceleration Driven Damper, Power Driven Damper, H∞ Robust Control, Fuzzy Logic Controller, and Inverse ANFIS. H∞ Robust Control is an advanced method that guarantees transient performance and rejects external disturbances. It is shown that H∞ with the proposed modification, has the best performance although its relatively high cost of computation could be potentially considered as a drawback. Also, the proposed Inverse ANFIS controller uses the power of fuzzy systems along with neural networks to help improve vehicle ride metrics significantly. In this study, a novel approach is introduced to analyze and fine-tune semi-active suspension control algorithms. In some cases, such as military trucks moving on off-road terrains, it is critical to keep the vehicle ride quality in an acceptable range. Semi-active suspensions are used to have more control over the ride metrics compared to passive suspensions and also, be more cost-effective compared to active suspensions. The proposed methodology will investigate the skyhook-groundhook hybrid controller. This is accomplished by conducting sensitivity analysis of the controller performance to varying vehicle/road parameters. This approach utilizes sensitivity analysis and one-at-a-time methodology to find and reach the optimum point of vehicle suspensions. Furthermore, real-time tuning of the mentioned controller will be studied. The online tuning will help keep the ride quality of the vehicle close to its optimum point while the vehicle parameters are changing. A quarter-car model is used for all simulations and analyses. / Ph. D. / Passenger safety and comfort have always been two major concerns in designing and engineering vehicles. Suspensions play a vital role in this regard. They are there to ensure a very smooth and comfortable ride experience. Many technologies have been developed to increase performance of suspension and customize their functionality. However, only a few developments led to a new family of suspensions and opened a broad field in automotive engineering for researchers to do their twist and tweaks. One fascinating technology that was developed a few decades ago, was semi-active suspensions. Their advantage over conventional ones is that its stiffness can be adjusted on the fly. This property can be combined with a control methodology in order to improve the ride experience further more compared to conventional suspensions. In this dissertation, some novel control methodologies are developed and compared with existing ones. The results are discussed exclusively for each controller.