This on-road field investigation employed, for the first time, a completely automated, trigger-based data collection system capable of evaluating driver performance in an extended duration real-world commercial motor vehicle environment. The complexities associated with the development of the system, both technological and logistical and the necessary modifications to the plan of research are presented herein This study, performed in conjunction with an on-going three year contract with the Federal Highway Administration, examined the use of self-rating of alertness and temporal separation (minimum time-to-collision, minimum headway, and mean headway) as indicators of driver fatigue. Without exception, the regression analyses for both the self-rating of alertness and temporal separation yielded models low in predictive ability; neither metric was found to be a valid indicator of driver fatigue. Various reasons for the failure of self-rating of fatigue as a valid measure are discussed. Dispersion in the data, likely due to extraneous (non-fatigue related) factors (e.g., other drivers) are credited with reducing the sensitivity of the temporal separation indicators. Overall fatigue levels for all temporal separation incidents (those with a time-to-collision equal to or less than four seconds) were found to be significantly higher than for those randomly triggered incidents. On this basis, it is surmised that temporal separation may be a sensitive indicator for time-to-collision values greater than the 4-second criterion employed in this study. Two unexpected relationships in the data are also discussed. A "wall" effect was found to exist for minimum time-to-collision values at 1.9 seconds. That is, none of the participants who participated in this research effort exhibited following behaviors with less than a 1.9-second time-to-collision criterion. In addition, based upon the data collected for this research, anecdotal evidence suggests that commercial motor vehicle operators do not appear to follow the standard progression of events associated with the onset of fatigue. / Ph. D.

[EN] The efficiency of free-collision motion planning results very sensible on robot and obstacle modelling technique selected. In this way, many works have been oriented to define models with proper throughput to speed up the collision detection proccess. This dissertation presents a new approach to the problem, whose complexity is reduced notably by means of using enveloping models of real objects, allowing security regions or distances. This objective is reached by means of the definition of a spherical model, composed of infinite spheres, generated from the application of linear or polynomial equations to a reduced number of control spheres, giving the so-called poly-spheres and spheroids respectively. These models, with evident simplicity, present a high modelling power, adapt easily to the requirements need in collision-detection and path planning applications for robotics systems. In order to represent a complete multi-robot cell, an extended hierarchical structure has been defined, in form of an AND-OR graph, with different degrees of accuracy, according to the different approximation model used. In order to generate automatically this structure, a procedure has been developed to compute the minimum volume enveloping spherical model in an off-line process with two levels based on Downhill Simplex method and Hough transform. This procedure can be greatly speed up by using clustering techniques to obtain appropiate initial conditions, allowing an on-line use. With a hierarchical structure computed in such a way, a fast procedure for collision detection in a multi-robot cell is introduced, based on several algorithms for distance computation including polyspheres and spheroids. This methodology presents a fast and anticipativa response, in the sense that every movement of a system has been validated before its execution, implying that not necessarily must be done in an off-line simulation. The use of spherical models, in addition to their fast distance computation, results suitable for the definition of artificial potential fields allowing a path planning in robotics systems with up to six degrees of freedom, including three for translation and three for rotation. The definition of these new potential fields and the study of new planning techniques based on classical optimisation methods allow their application straight forward in Cartesian space, with all their advantages. Last but not least, with the help of some systems for robot programming, simulation and control, the correctness of these contributions have been validated in a set of prototype applications, covering from robot-obstacle and multi-robot collision detection, to motion planning for a robot-arm or an auto-guided vehicle. / [ES] La eficiencia de la planificación de movimientos libres de colisión resulta muy sensible al modelado de los robots y obstáculos que se consideren, por lo que, frente al modelado tradicional con politopos, muchos trabajos en robótica han estado orientados a la definición de unos modelos que presenten buenas prestaciones de cara a acelerar el proceso de detección de colisiones. En esta Tesis se presenta una nueva perspectiva del problema, cuya complejidad queda reducida notablemente al utilizar envolventes de los objetos reales, lo que permite definir zonas o distancias de seguridad. Para ello se han definido unos modelos esféricos, compuestos de infinitas esferas generadas a partir de la aplicación de unas relaciones lineales o polinómicas a un número reducido de esferas de control, dando lugar a las llamadas poli-esferas y esferoides respectivamente. Estos modelos, de sencillez clara, presentan una potencia de modelado elevada, adaptándose fácilmente a los requisitos necesarios en las aplicaciones de detección de colisiones y planificación de movimientos en sistemas robotizados. Para la representación de una célula multi-robot completa, se ha definido una estructura jerárquica extendida, en forma de grafo AND-OR, con diferentes grados de precisión, mediante diferentes modelos de aproximación. De cara a generar automáticamente esta estructura, se ha desarrollado un procedimiento para generar el modelo esférico envolvente de mínimo volumen en un proceso off-line con dos niveles, basados en el método de minimización Downhill Simplex y en la transformada de Hough. Este procedimiento se acelera enormemente al utilizar técnicas de agrupamiento para obtener condiciones iniciales apropiadas, permitiendo su uso on-line. Con una estructura jerárquica generada de esta forma, se introduce un procedimiento rápido de detección de colisiones aplicable a una célula multi-robot, basado en algoritmos básicos de cálculo de distancias que pueden considerar poli-esferas y esferoides. Esta metodología presenta una respuesta rápida y anticipativa, entendiendo por tal que todo movimiento de cualquier sistema ha sido validado antes de su ejecución, por lo que no necesariamente debe realizarse en una simulación off-line. La utilización de modelos esféricos, así como el rápido cálculo de distancias entre ellos, resulta idónea para la definición de campos potenciales artificiales que permitan una planificación de movimientos en sistemas robotizados con hasta seis grados de libertad, incluyendo tres de traslación y tres de rotación. La definición de estos nuevos campos potenciales y el estudio de nuevas técnicas de planificación basados en métodos clásicos de optimización permiten su aplicación directamente en el espacio cartesiano, con las claras ventajas que esto conlleva. Finalmente, con la ayuda de varios sistemas de programación, simulación y control de robots, se ha demostrado la validez de estas aportaciones en una serie de aplicaciones prototipo que van desde la detección de colisiones de un robot con un obstáculo o entre sistemas multi-robot, a la planificación de movimientos de un brazo-robot o un vehículo autoguiado. / Mellado Arteche, M. (1996). Modelado, detección de colisiones y planificación de movimientos en sistemas robotizados mediante volúmenes esféricos [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/56621

Augmented reality (AR) has the potential to fundamentally change the way we interact with information. Direct perception of computer generated graphics atop physical reality can afford hands-free access to contextual information on the fly. However, as users must interact with both digital and physical information simultaneously, yesterday's approaches to interface design may not be sufficient to support the new way of interaction. Furthermore, the impacts of this novel technology on user experience and performance are not yet fully understood. Driving is one of many promising tasks that can benefit from AR, where conformal graphics strategically placed in the real-world can accurately guide drivers' attention to critical environmental elements. The ultimate purpose of this study is to reduce pedestrian accidents through design of driver interfaces that take advantage of AR head-up displays (HUD). For this purpose, this work aimed to (1) identify information requirements for pedestrian collision warning, (2) design AR driver interfaces, and (3) quantify effects of AR interfaces on driver performance and experience. Considering the dynamic nature of human-environment interaction in AR-supported driving, we took an ecological approach for interface design and evaluation, appreciating not only the user but also the environment. The requirement analysis examined environmental constraints imposed on the drivers' behavior, interface design translated those behavior-shaping constraints into perceptual forms of interface elements, and usability evaluations utilized naturalistic driving scenarios and tasks for better ecological validity. A novel AR driver interface for pedestrian collision warning, the virtual shadow, was proposed taking advantage of optical see-through HUDs. A series of usability evaluations in both a driving simulator and on an actual roadway showed that virtual shadow interface outperformed current pedestrian collision warning interfaces in guiding driver attention, increasing situation awareness, and improving task performance. Thus, this work has demonstrated the opportunity of incorporating an ecological approach into user interface design and evaluation for AR driving applications. This research provides both basic and practical contributions in human factors and AR by (1) providing empirical evidence furthering knowledge about driver experience and performance in AR, and, (2) extending traditional usability engineering methods for automotive AR interface design and evaluation. / Ph. D. / On average, a pedestrian was killed every 2 hours and injured every 8 minutes on U.S. roadways in 2013. Most common driver errors responsible for pedestrian collisions were drivers’ lack of situation awareness due to low visibility or unexpected appearance of pedestrians. As a solution to the problem, automakers introduced pedestrian collision warnings, taking advantage of recent advances in sensor technology and pedestrian detection algorithms. Once pedestrians are detected in the vehicle’s path, warnings are given to the driver typically through auditory alarms and/or simple visual symbols. However, with current warnings that often lack spatial information, drivers need to further localize and evaluate approaching pedestrians’ movement for appropriate decision and reaction. Augmented reality (AR) is one of the most promising solutions to address the limitations of current warning interfaces. By overlaying computer generated conformal graphics atop physical reality, AR head up displays (HUDs) can guide drivers’ attention to dangerous pedestrians, affording direct perception of spatial information about those pedestrians. The ultimate purpose of this work is to reduce pedestrian accidents by design of driver interfaces, taking advantage of AR HUDs. For this purpose, we aimed to (1) design a novel driver interface for cross traffic alerts, (2) prototype design ideas for a specific use-case of pedestrian collision warning, and (3) evaluate usability of the new design ideas in consideration of unique aspects of human-environment interaction with AR while driving. We proposed a novel driver interface for pedestrian collision warning, the virtual shadow, which can cast shadows of approaching pedestrians to the vehicle’s path via AR HUDs. Usability evaluations in a driving simulator and a roadway showed the potential benefits of the proposed idea over existing warnings in driver attention management, situation awareness, task performance with reduced workload. Thus, this work demonstrated the capabilities of AR HUDs as intuitive and effective interfaces for vehicle drivers.

Forward Collision Avoidance Systems (FCAS) are a type of active safety system which have great potential for rear-end collision avoidance. These systems use either radar, lidar, or cameras to track objects in front of the vehicle. In the event of an imminent collision, the system will warn the driver, and, in some cases, can autonomously brake to avoid a crash. However, driver acceptance of the systems is paramount to the effectiveness of a FCAS system. Ideally, FCAS should only deliver an alert or intervene at the last possible moment to avoid nuisance alarms, and potentially have drivers disable the system. A better understanding of normal driving behavior can help designers predict when drivers would normally take avoidance action in different situations, and customize the timing of FCAS interventions accordingly. The overall research object of this dissertation was to characterize normal driver behavior in car following events based on naturalistic driving data. The dissertation analyzed normal driver behavior in car-following during both braking and lane change maneuvers. This study was based on the analysis of data collected in the Virginia Tech Transportation Institute 100-Car Naturalistic Driving Study which involved over 100 drivers operating instrumented vehicles in over 43,000 trips and 1.1 million miles of driving. Time to Collision in both braking and lane change were quantified as a function of vehicle speed and driver characteristics. In general, drivers were found to brake and change lanes more cautiously with increasing vehicle speed. Driver age and gender were found to have significant influence on both time to collision and maximum deceleration during braking. Drivers age 31-50 had a mean braking deceleration approximately 0.03 g greater than that of novice drivers (age 18-20), and female drivers had a marginal increase in mean braking deceleration as compared to male drivers. Lane change maneuvers were less frequent than braking maneuvers. Driver-specific models of TTC at braking and lane change were found to be well characterized by the Generalized Extreme Value distribution. Lastly, driver's intent to change lanes can be predicted using a bivariate normal distribution, characterizing the vehicle's distance to lane boundary and the lateral velocity of the vehicle. This dissertation presents the first large scale study of its kind, based on naturalistic driving data to report driver behavior during various car-following events. The overall goal of this dissertation is to provide a better understanding of driver behavior in normal driving conditions, which can benefit automakers who seek to improve FCAS effectiveness, as well as regulatory agencies seeking to improve FCAS vehicle tests. / Ph. D.

In this dissertation, we seek to improve collision avoidance for autonomous underwater vehicles (AUVs). More specifically, we consider the case of a small AUV using a forward-looking sonar system with a limited number of beams. We describe a high-fidelity sonar model and simulation environment that was developed to aid in the design of the sonar system. The simulator achieves real-time visualization through ray tracing and approximation, and can be used to assess sonar design choices, such as beam pattern and beam location, and to evaluate obstacle detection algorithms. We analyze the benefit of using a few beams instead of a single beam for a low-cost obstacle avoidance sonar for small AUVs. Single-beam systems are small and low-cost, while multi-beam sonar systems are more expensive and complex, often incorporating hundreds of beams. We want to quantify the improvement in obstacle avoidance performance of adding a few beams to a single-beam system. Furthermore, we developed a collision avoidance strategy specifically designed for the novel sonar system. The collision avoidance strategy is based on posterior expected loss, and explicitly couples obstacle detection, collision avoidance, and planning. We demonstrate the strategy with field trials using the 690 AUV, built by the Center for Marine Autonomy and Robotics at Virginia Tech, with a prototype forward-looking sonar comprising of nine beams. / Doctor of Philosophy / This dissertation focuses on improving collision avoidance capabilities for small autonomous underwater vehicles (AUVs). Specifically, we are looking at the scenario of an AUV equipped with a forward-looking sonar system using only a few beams to detect obstacles in our environment. We develop a sophisticated sonar model and simulation environment to facilitate the design of the sonar system. Our simulator enables real-time visualization, offering insights into sonar design aspects. It also serves as a tool for evaluating obstacle detection algorithms. The research investigates the advantages of utilizing multiple beams compared to a single-beam system for a cost-effective obstacle avoidance solution for small AUVs. Single-beam sonar systems are small and affordable, while multi-beam sonar systems are more complex and expensive. The aim is to quantify the improvement in obstacle avoidance performance when adding additional sonar beams. Additionally, a collision avoidance strategy tailored to the novel sonar system is developed. This strategy, developed using a statistical model, integrates obstacle detection, collision avoidance, and planning. The effectiveness of the strategy is demonstrated through field trials using the 690 AUV, constructed by the Center for Marine Autonomy and Robotics at Virginia Tech, equipped with a prototype forward-looking sonar using nine beams.

This dissertation explores the development of sophisticated distributed layered control algorithms focused on the navigation, planning, and control of multi-agent quadrupedal robots collaborating in uncertain environments. Quadrupedal robots are high-dimensional, complex systems that are inherently unstable, posing significant challenges in designing predictive control laws. Template models offer a solution by providing a bridging layer of reduced-order models with fewer state variables and linearized dynamics. However, this approach compromises the agility and full potential of these sophisticated machines, as template models may fail to capture the intricate nonlinear dynamics of quadrupedal robots. Furthermore, in multi-robot systems (MRS) where numerous robots operate concurrently, it becomes crucial to develop strategies embedding collision safety mechanisms. One approach involves embedding Euclidean distance constraints in the predictive control formulation. While effective, this method significantly complicates the optimal control (OC) problem and increases computational overhead. To mitigate these challenges, this dissertation explores hierarchical and distributed control frameworks, focusing on developing real-time feasible controllers that guarantee collision avoidance while preserving the agility of these hardware platforms by utilizing fully nonlinear template models. In particular, this research investigates a multi-layered framework consisting of potential fields at the high-level layer, a distributed nonlinear model predictive control (DNMPC) based middle-level layer responsible for uncertainty mitigation, and full-order nonlinear controllers at the low-level layer. Additionally, the latter part of this dissertation examines the integration of safety-ensuring control barrier functions (CBFs) into the nonlinear model predictive control (NMPC) layer, thereby providing rigorous mathematical guarantees for collision avoidance. The crux of this research lies in addressing the following questions: How do we design layered control frameworks to guarantee optimal gait planning and collision avoidance while maintaining computational tractability? How do we mitigate uncertainty in the environment in real-time using safety-critical control algorithms? / Doctor of Philosophy / This dissertation investigates advanced control strategies for coordinating teams of quadrupedal robots in dynamic and uncertain environments. Quadrupedal robots present significant challenges in control and stability due to their complex, high-dimensional nature and inherent instability. Current approaches often employ simplified models for control, which, while computationally efficient, fail to fully capture the intricate dynamics of these sophisticated machines, thus limiting their agility and potential. Furthermore, in multi-robot systems, ensuring collision avoidance is a practically integral part of control. Conventional methods, including Euclidean distance constraints for collision avoidance, prove effective but substantially increase computational demands and complicate the optimal control problem. To address these challenges, this research explores a hierarchical, distributed control framework designed to guarantee collision-free navigation while maximizing the agility of quadrupedal platforms through the use of comprehensive nonlinear models. The proposed framework consists of three primary layers: a high-level layer utilizing potential fields for global path planning, a middle layer employing distributed nonlinear model predictive control for local navigation and uncertainty mitigation, and a low-level layer implementing full-order nonlinear controllers for precise motion execution. Additionally, this work examines the integration of control barrier functions into the predictive control layer, providing mathematical guarantees for collision avoidance. The core objectives of this research are twofold: first, to develop layered control frameworks that ensure optimal gait planning and collision avoidance while maintaining computational feasibility; and second, to create real-time algorithms capable of mitigating environmental uncertainties using safety-critical control methods. By addressing these fundamental questions, this dissertation aims to advance the field of multi-agent quadrupedal robotics, enhancing the capability of robotic teams to operate effectively in complex, unpredictable environments. The potential applications of this research extend to critical areas such as search and rescue operations, environmental monitoring, and exploration of hazardous terrains.

Autonomous mobile robots are beginning to try to help us provide different delivery services in people's lives, such as delivering medicines in hospitals, delivering goods in warehouses, and delivering food in restaurants. To realize this vision, robots need to navigate autonomously and efficiently through complex, crowded, and dynamic environments filled with static obstacles, such as tables and chairs, as well as people and/or other robots, and to achieve this using the computational resources available onboard a mobile robot. This dissertation improves the state-of-the-art in autonomous navigation by developing learning-based algorithms to model the environment around the robot, predict changes in the environment, and control the robot, all of which can run onboard a mobile robot in real time. Specifically, this dissertation first proposes a set of specialized preprocessed data representations to extract and encode useful high-level information about crowded dynamic environments from raw sensor data (i.e., a short history of lidar data, kinematic data about nearby pedestrians, and a sub-goal that leads the robots towards its final destination). Then, using these combined preprocessed data representations, this dissertation proposes a novel crowd-aware navigation control policy that can balance collision avoidance and speed in crowded dynamic scenes by designing a velocity obstacle-based reward function that is used to train the robot leveraging deep reinforcement learning techniques. This dissertation then proposes a series of hardware-friendly prediction algorithms, based on variational autoencoder networks, to predict a distribution of possible future states in dynamic scenes by exploiting the kinematics and dynamics of the robot and its surrounding objects. Furthermore, this dissertation proposes a novel predictive uncertainty-aware navigation framework to improve the safety performance of current existing control policies by incorporating the output of the proposed stochastic environment prediction algorithms into general navigation frameworks. Many different collected real-world datasets as well as a series of 3D simulation experiments and hardware experiments are used to demonstrate the effectiveness of these proposed novel learning-based prediction and control algorithms. The new algorithms outperform other state-of-the-art algorithms in terms of collision avoidance, robot speed, and prediction accuracy across a range of environments, crowd densities, and robot models. It is believed that all the work included in this dissertation will promote the development of autonomous navigation for modern mobile robots, provide highly innovative solutions to the open problem of autonomous navigation in crowded dynamic scenes, and make our daily lives more convenient and efficient. / Mechanical Engineering

Mass spectrometry is a commonly used technique in the modern sports anti-doping laboratory. Characteristic product ions observed in tandem mass spectrometry (MS/MS) can be used to identify prohibited substances. However, with continuous introduction of novel uncharacterized drugs, there is a need to increase the selectivity and coverage identification of mass spectrometry and non-mass spectrometry-based methods. The use of separations methods, (e.g., chromatography) is another means to identify substances using retention times, providing an additional dimension of analysis. Broadly, this work examines mass spectrometry of small molecules, with a focus on pharmaceuticals of sports anti-doping relevance. To gain a deeper understanding of characteristic product ions and their dissociation pathways, multi-stage mass spectrometry (MSn) and energy-resolved collision induced dissociation (E-resolved CID) were used. Using these methods, two classes of pharmaceuticals were studied: beta-2 agonists and beta blockers. Sequential versus competitive pathways were elucidated for four beta-2 agonists: isoetharine, salbutamol, formoterol, and salmeterol. Water loss is a common dissociation mechanism, with multiple water losses observed where structurally possible. A similar methodology was used for further investigation of the dissociation chemistry of five beta blockers (labetalol, bisoprolol, carteolol, acebutolol, and atenolol). Insights into the nature of the neutral losses and structures of product ions characteristic to the class are highlighted. Isomers that share product ions pose a special challenge, where differentiation is not possible using single collision energy CID-MS. Three sets of isomers with similar MS/MS patterns (leucine and tert¬-leucine, quinoline and isoquinoline, and para-, ortho-, and meta-aminobenzoic acid) were analyzed by E-resolved CID to investigate the analytical utility of this approach for isomer differentiation. Unique “fingerprints” were found among each set of isomers and additional analytical considerations were also investigated. Finally, separation of enantiomers is another special challenge, as MS techniques are “chirality blind”. Ion mobility spectrometry (IMS), a gas-phase separation technique, has been reported to show separation of enantiomers with the aid of drift gas modifiers (DGMs). Chiral butanol was used as a DGM to aid the IMS analysis of salbutamol enantiomers. These efforts were ultimately unsuccessful, which is in line with current literature.

Strobe lights have been used successfully in many transportation applications to increase conspicuity. It was hoped that a strobe signal could also be applied to more effectively warn distracted drivers of an unexpected rear end conflict. This "proof of concept study" used a 2 x 2 between-subjects design using thirty-three subjects (16 subjects in the strobe condition, 17 subjects in the no strobe condition) who were divided into two age groups: younger (25-35) and older (60-70). The driver unexpectedly encountered a stopped "surrogate" vehicle in the roadway (with or without a rear-facing strobe light) in a controlled on-road study at the Smart Road located at the Virginia Tech Transportation Institute (VTTI). Results suggested that younger subjects' perception times improved as a result of being exposed to the strobe signal. Faster perception of the situation allowed more time to initiate a brake response. Older subjects perception and response times remained unchanged by the strobe signal. More severe initial steering rate and subjective responses indicated that the strobe conveyed a sense of urgency irrespective of age. Visual distraction of subjects proved difficult. Hence, the impact of the strobe on attracting the attention of a visually distracted driver to the stimulus could not be as fully investigated as originally hoped. The formulation of a more difficult distraction task was suggested for future research to truly assess the ability of the strobe light at alerting visually distracted drivers. / Master of Science

This study investigates the structural integrity of hydrofoils under three scenarios: regular boating, turning, and collision with an underwater obstacle. To analyse the forces acting on the hydrofoil, calculations were performed and simulations were conducted in SolidWorks using a CAD model of the hydrofoil.The simulations reveal that the weld between the struts and the wing undergoes plastic deformation during both regular boating and turning. This deformation is particularly problematic during turning, as the forces on the weld increase significantly. Under the collision scenario, the bolts at the breakpoint fail before critical damage occurs to the components above.The results highlight the weakness of the weld and the need for a redesign of the hydrofoil to eliminate it. An alternative fastening method, such as bolted joints with watertight sealing, should be considered.