Assistive robots have the potential to enable persons with motor disabilities to live more independent lives. Object retrieval has been rated a high-priority task for assistive robots. A key challenge in creating effective assistive robots lies in designing control interfaces that enable the human user to control the robot. This thesis builds on prior work that uses a laser pointer to allow the person to intuitively communicate their goals to a robot by creating a `clickable world'. Specifically, this thesis reduces the infrastructure needed for the robot to recognize the user's goal by augmenting the laser pointer with a small camera, an inertial measurement unit (IMU), and a laser rangefinder to estimate the location of the object to be grasped. The robot then drives to the approximate target location given by input from the laser pointer while using an onboard camera to detect an object near the target location. Local autonomy on the robot is used to visually navigate to the detected object to enable object retrieval. Results show a successful proof of concept in demonstrating reasonable detection of user intent on a 1.23 x 1.83 meters squared test grid. Testing of the estimation of object location in the odometry frame fell within range of successful local autonomy object retrieval for an environment with a single object. Future work includes testing on a wide variety of dropped objects and in cluttered environments which is needed to validate the effectiveness of the system for potential end users.

abstract: The world population is aging. Age-related disorders such as stroke and spinal cord injury are increasing rapidly, and such patients often suffer from mobility impairment. Wearable robotic exoskeletons are developed that serve as rehabilitation devices for these patients. In this thesis, a knee exoskeleton design with higher torque output compared to the first version, is designed and fabricated. A series elastic actuator is one of the many actuation mechanisms employed in exoskeletons. In this mechanism a torsion spring is used between the actuator and human joint. It serves as torque sensor and energy buffer, making it compact and safe. A version of knee exoskeleton was developed using the SEA mechanism. It uses worm gear and spur gear combination to amplify the assistive torque generated from the DC motor. It weighs 1.57 kg and provides a maximum assistive torque of 11.26 N·m. It can be used as a rehabilitation device for patients affected with knee joint impairment. A new version of exoskeleton design is proposed as an improvement over the first version. It consists of components such as brushless DC motor and planetary gear that are selected to meet the design requirements and biomechanical considerations. All the other components such as bevel gear and torsion spring are selected to be compatible with the exoskeleton. The frame of the exoskeleton is modeled in SolidWorks to be modular and easy to assemble. It is fabricated using sheet metal aluminum. It is designed to provide a maximum assistive torque of 23 N·m, two times over the present exoskeleton. A simple brace is 3D printed, making it easy to wear and use. It weighs 2.4 kg. The exoskeleton is equipped with encoders that are used to measure spring deflection and motor angle. They act as sensors for precise control of the exoskeleton. An impedance-based control is implemented using NI MyRIO, a FPGA based controller. The motor is controlled using a motor driver and powered using an external battery source. The bench tests and walking tests are presented. The new version of exoskeleton is compared with first version and state of the art devices. / Dissertation/Thesis / Masters Thesis Mechanical Engineering 2018

This thesis presents three improvements to the current UCF MANUS systems. The first improvement modifies the existing fine motion controller into PI controller that has been optimized to prevent the object from leaving the view of the cameras used for visual servoing. This is achieved by adding a weight matrix to the proportional part of the controller that is constrained by an artificial ROI. When the feature points being used are approaching the boundaries of the ROI, the optimized controller weights are calculated using quadratic programming and added to the nominal proportional gain portion of the controller. The second improvement was a compensatory gross motion method designed to ensure that the desired object can be identified. If the object cannot be identified after the initial gross motion, the end-effector will then be moved to one of three different locations around the object until the object is identified or all possible positions are checked. This framework combines the Kanade-Lucase-Tomasi local tracking method with the ferns global detector/tracker to create a method that utilizes the strengths of both systems to overcome their inherent weaknesses. The last improvement is a particle-filter based tracking algorithm that robustifies the visual servoing function of fine motion. This method performs better than the current global detector/tracker that was being implemented by allowing the tracker to successfully track the object in complex environments with non-ideal conditions.

Exoskeletons are rapidly emerging from the realm of science-fiction myth to practical reality in everyday life. Various designs have provided viable means for individuals to regain capabilities that were lost or perform tasks not previously possible by their ability alone. In this research, I propose two novel exoskeletons for walking assistance and heavy load carriage. The first exoskeleton can be used to provide assistance for walking in various applications such as industrial productivity, rehabilitation, and military or space training. We introduce a design for a lower body wearable device that supports up to 80% of the user's body weight (667 N peak force) with a single actuator on each leg. Its underactuated design directs force through the user's center of mass with a single sprocket-chain driven prismatic actuator on each leg, allowing for natural gait and mobility. The device is optimized for simplicity, ease of assembly, low cost, and weight. The second design aims to counteract the one of the leading causes of injury in the workplace, repetitive and heavy lifting. The Heavy Lift and Carry Exoskeleton (HeavyLC Exo) is capable of safely lifting and carrying loads up to 36 kg (80 lbs) while minimizing the number of actuators to reduce weight and complexity. The HeavyLC Exo allows the user to direct the object, pause and hold the object steady mid-lift, and follow the natural kinematics of lifting. It is secured to the user with shoulder, chest, and dual thigh straps, along with an adjustable waist belt and overshoe attachment. Powered by two 14.8 V batteries and an off-board air compressor, the HeavyLC Exo has a total of 20 DOF, with 6 actuated DOF and 14 free DOF. The arms use only two actuators each, providing powered lifting and arm retraction/extension, and allowing a wide range of body postures; the legs are powered by single pneumatic actuators on each leg connected to the foot accompanied by a passive spring element to prevent excessive pelvic tilt and leg abduction during swing. The control system requires directional forces from the user at the tool handle of 19 N (4.3 lbf) on average. Current design limitations necessitate the user to provide up to 280 N (62.9 lbf) at the hip during worst load conditions, and future design optimization is proposed. A fully functional prototype of HeavyLC Exo is built, fully tested, and analyzed for improvement. / Master of Science / Exoskeletons, which were once only seen in science fiction, are now becoming a reality in everyday life. Various designs have made it possible for people to do things they couldn't do before or regain abilities they lost. In this research, two new exoskeletons are proposed - one for walking assistance and the other for carrying heavy loads. The first exoskeleton is designed to help people walk. It supports up to 80 % of the user's body weight with a single actuator on each leg, which directs force through the center of mass, allowing for natural gait and mobility. It's simple, easy to assemble, low-cost, and lightweight, making it useful in various applications such as medical rehabilitation, military or space training, and industrial productivity. The second exoskeleton is designed to help people lift and carry heavy objects, which is a common cause of workplace injuries. The Heavy Lift and Carry Exoskeleton (HeavyLC Exo) can safely lift and carry objects up to 36 kg (80 lbs) while minimizing the number of actuators to reduce weight and complexity. It's worn by the user using shoulder, chest, and dual thigh straps, along with an adjustable waist belt and overshoe attachment. The exoskeleton is powered by two batteries and an off-board air compressor, and has 20 degrees of freedom, with 6 powered and 14 non-powered, giving it significant flexibility to conform to lifting and walking motions allowing it to function with normal user range of motion. The arms use two actuators each to provide powered lifting and arm retraction/extension, while the legs are powered by single pneumatic actuators on each leg connected to the foot accompanied by a elastic spring element. The control system requires the user to lift and maneuver about 1.9 kg (4.2 lbs) to direct the object. A fully functional prototype has been built, tested, and analyzed for changes in the future.

There have been substantial improvements in the area of rehabilitation robotics in the recent past. However, these advances are inaccessible to a large number of people with disabilities who are in most need of such assistance. This group includes people who are in a severely paralyzed state, that they are completely "locked-in" in their own bodies. Such persons usually retain full cognitive abilities, but have no voluntary muscle control. For these persons, a Brain Computer Interface (BCI) is often the only way to communicate with the outside world and/or control an assistive device. One major drawback to BCI devices is their low information transfer rate, which can take as long as 30 seconds to select a single command. This can result in mental fatigue to the user, specially if it necessary to make multiple selections over the BCI to complete a single task. Therefore, P300 based BCI control is not efficient for controlling a assistive robotic device such as a robotic arm. To address this shortcoming, a novel vision based Brain Robot Interface (BRI) is presented in this thesis. This visual user interface allows for selecting an object from an unstructured environment and then performing an action on the selected object using a robotic arm mounted to a power wheelchair. As issuing commands through BCI is slow, this system was designed to allow a user to perform a complete task via a BCI using an autonomous robotic system while issuing as few commands as possible. Furthermore, the new visual interface allows the user to perform the task without losing concentration on the stimuli or the task. In our interface, a scene image is captured by a camera mounted on the wheelchair, from which, a dynamically sized non-uniform stimulus grid is created using edge information. Dynamically sized grids improve object selection efficiency. Oddball paradigm and P300 Event Related Potentials (ERP) are used to select stimuli, where the stimuli being each cell in the grid. Once selected, object segmentation and matching is used to identify the object. Then the user, using BRI, chooses an action to be performed on the object by the wheelchair mounted robotic arm (WMRA). Tests on 8 healthy human subjects validated the functionality of the system. An average accuracy of 85.56% was achieved for stimuli selection over all subjects. With the proposed system, it took the users an average of 5 commands to perform a task on an object. The system will eventually be useful for completely paralyzed or locked-in patients for performing activities of daily living (ADL) tasks.

On one hand, academic and industrial researchers have been developing and deploying robots that are used as educational tutors, mediators, and motivational tools. On the other hand, an increasing amount of interest has been placed on non-expert users being able to program robots intuitively, which has led to promising research efforts in the fields of machine learning and human-robot interaction. This dissertation focuses on bridging the gap between the two subfields of robotics to provide personalized experience for the users during educational, entertainment, and therapeutic sessions with social robots. In order to make the interaction continuously engaging, the workspace shared between the user and the robot should provide personalized contexts for interaction while the robot learns to participate in new tasks that arise. This dissertation aims to solve the task-learning problem using an instance-based framework that stores human demonstrations as task instances. These instances are retrieved when confronted with a similar task in which the system generates predictions of task behaviors based on prior solutions. The main issues associated with the instance-based approach, i.e., knowledge encoding and acquisition, are addressed in this dissertation research using interactive methods of machine learning. This approach, further referred to as interactive instance-based learning (IIBL), utilizes the keywords people use to convey task knowledge to others to formulate task instances. The key features suggested by the human teacher are extracted during the demonstrations of the task. Regression approaches have been developed in this dissertation to model similarities between cases for instance retrieval including multivariate linear regression and sensitivity analysis using neural networks. The learning performance of the IIBL methods were then evaluated while participants engaged in various block stacking and inserting scenarios and tasks on a touchscreen tablet with a humanoid robot Darwin. In regard to end-users programming robots, the main benefit of the IIBL framework is that the approach fully utilizes the explanatory behavior of the instance-based method which makes the learning process transparent to the human teacher. Such an environment not only encourages the user to produce better demonstrations, but also prompts the user to intervene at the moment a new instance is needed. It was shown through user studies that participants naturally adapt their teaching behavior to the robot learner's progress and adjust the timing and the number of demonstrations. It was also observed that the human-robot teaching and learning scenarios facilitate the emergence of various social behaviors from participants. Encouraging social interaction is often an objective of the task especially with children with cognitive disabilities, and a pilot study with children with autism spectrum disorder revealed promising results comparable to the typically developing group. Finally, this dissertation investigated the necessity of renewable context for prolonged interaction with robot companions. Providing personalized tasks that match each individual's preferences and developmental stages enhances the quality of the user experience with robot learners. Confronted with the limitations of the physical workspace, this research proposes utilizing commercially available touchscreen smart devices as a shared platform for engaging the user in educational, entertainment, and therapeutic tasks with the robot learners. To summarize, this dissertation attempts to defend the thesis statement that a robot learner that utilizes an IIBL approach improves the performance and efficiency of general task learning, and when combined with the state-of-the-art mobile technology that provides personalized context for interaction, enhances the user's experience for prolonged engagement of the task.

abstract: Wearable robotics is a growing sector in the robotics industry, they can increase the productivity of workers and soldiers and can restore some of the lost function to people with disabilities. Wearable robots should be comfortable, easy to use, and intuitive. Robust control methods are needed for wearable robots that assist periodic motion. This dissertation studies a phase based oscillator constructed with a second order dynamic system and a forcing function based on the phase angle of the system. This produces a bounded control signal that can alter the damping and stiffens properties of the dynamic system. It is shown analytically and experimentally that it is stable and robust. It can handle perturbations remarkably well. The forcing function uses the states of the system to produces stable oscillations. Also, this work shows the use of the phase based oscillator in wearable robots to assist periodic human motion focusing on assisting the hip motion. One of the main problems to assist periodic motion properly is to determine the frequency of the signal. The phase oscillator eliminates this problem because the signal always has the correct frequency. The input requires the position and velocity of the system. Additionally, the simplicity of the controller allows for simple implementation. / Dissertation/Thesis / Doctoral Dissertation Mechanical Engineering 2016

Esta investigación doctoral aborda la robótica asistencial (Assistive Robotics) como tema general, también llamada robótica social; específicamente el estudio de la interacción de la relación humano-robot. Con base en el estudio del estado del arte realizado, esta tesis se orientó hacia el siguiente problema de investigación: ¿es posible mejorar la efectividad terapéutica y de esta forma la calidad de vida de las personas con dificultades para comunicarse verbalmente debido a singularidades relacionadas con el habla y lenguaje; por medio del desarrollo de estrategias socio-terapéuticas que utilicen robots y mundos virtuales con frecuencias de exposición flexibles en comparación con la exposición programada utilizando únicamente robots en el mundo físico? El problema de investigación brinda una direccionalidad innovadora desde varias perspectivas científicas. Por un lado, la integración terapéutica de recursos en el mundo virtual así como en el mundo físico con robótica asistencial en colaboración con el ser humano para cumplir un objetivo de crecimiento personal. Además, la posibilidad de romper la barrera del tiempo controlado de exposición terapéutica por medio de la tecnología. Y, por otro lado, metodológicamente buscar un abordaje científico que demuestre causalidad y no sólo asociación por medios cualitativos; dado que se quiere saber si la efectividad terapéutica realmente puede incrementar como variable de respuesta. Por lo tanto, y como elemento de innovación adicional, esta investigación abordó el diseño de un prototipo tecnológico de programación robótica con emulación animada, que integra el uso de un avatar robótico virtual para facilitar la interacción social de personas que presentan dificultades de comunicación verbal relacionadas con el habla y lenguaje. El objetivo general de esta tesis se plantea como sigue: “Aportar al conocimiento científico sobre la interacción humano-robot con fines terapéuticos de comunicación verbal en el idioma castellano comparando el uso de robots en el mundo físico y virtual con flexibilidad de tiempos versus la interacción limitada a robots físicos por periodos de tiempo fijos para saber si es posible incrementar de forma relevante la efectividad terapéutica en términos de mejora de habilidades y tiempo invertido en terapia.” Esta investigación contribuye científicamente con la propuesta de un enfoque metodológico que busca obtener resultados basados en la evidencia experimental y no sólo en el análisis hermenéutico o el análisis léxico de datos cualitativos que constituye lo más frecuente en la investigación científica en este campo. En este sentido, el método exploratorio basado en datos cualitativos y abordajes epistemológicos subjetivistas pueden verse bien complementados con investigación positivista más orientada a la evidencia basada en resultados vinculados a la causalidad. Otra aportación de esta investigación está en el desarrollo tecnológico orientado hacia el uso de la experiencia de realidad virtual de personas con condiciones de limitación en habla o lenguaje en combinación con un robot físico. Esta es una forma innovadora de buscar la exposición continua y en tiempo real a los protocolos de terapia de habla sin supervisión física del terapeuta, teniendo en cuenta que las aplicaciones robóticas en mundos virtuales vinculados con el "mundo físico" no son frecuentes.

Brain-machine interfaces (BMIs) are the only currently viable means of communication for many individuals suffering from locked-in syndrome (LIS) – profound paralysis that results in severely limited or total loss of voluntary motor control. By inferring user intent from task-modulated neurological signals and then translating those intentions into actions, BMIs can enable LIS patients increased autonomy. Significant effort has been devoted to developing BMIs over the last three decades, but only recently have the combined advances in hardware, software, and methodology provided a setting to realize the translation of this research from the lab into practical, real-world applications. Non-invasive methods, such as those based on the electroencephalogram (EEG), offer the only feasible solution for practical use at the moment, but suffer from limited communication rates and susceptibility to environmental noise. Maximization of the efficacy of each decoded intention, therefore, is critical. This thesis addresses the challenge of implementing a BMI intended for practical use with a focus on an autonomous assistive robot application. First an adaptive EEG- based BMI strategy is developed that relies upon code-modulated visual evoked potentials (c-VEPs) to infer user intent. As voluntary gaze control is typically not available to LIS patients, c-VEP decoding methods under both gaze-dependent and gaze- independent scenarios are explored. Adaptive decoding strategies in both offline and online task conditions are evaluated, and a novel approach to assess ongoing online BMI performance is introduced. Next, an adaptive neural network-based system for assistive robot control is presented that employs exploratory learning to achieve the coordinated motor planning needed to navigate toward, reach for, and grasp distant objects. Exploratory learning, or “learning by doing,” is an unsupervised method in which the robot is able to build an internal model for motor planning and coordination based on real-time sensory inputs received during exploration. Finally, a software platform intended for practical BMI application use is developed and evaluated. Using online c-VEP methods, users control a simple 2D cursor control game, a basic augmentative and alternative communication tool, and an assistive robot, both manually and via high-level goal-oriented commands.

Les systèmes de robotique sociale sont de plus en plus présents dans nos vies. Ce ne sont plus des entités isolées, mais on s'attend à ce qu'ils soient capables d'interagir et de communiquer avec les humains. Ils doivent respecter les normes comportementales attendues par les humains avec qui les systèmes robotiques sont en interaction.L'une des principales pistes de recherche dans le domaine de la robotique sociale est représentée par la conception d'une interaction naturelle entre un robot social et un individu. Plus spécifiquement, cette interaction devrait prendre en considération le profil de l'individu, l'état émotionnel, l'état physiologique et l'humeur, entre autres.Dans cette thèse, nous explorons la relation qui existe entre l'échelle de typologie circadienne, la performance cognitive et l'état physiologique au cours de différents scénarios d'interaction homme-robot. L'administration de différents questionnaires psychologiques permet de déterminer le profil d'un individu. En outre, à l’aide de différents capteurs (par exemple, GSR, caméra thermique), de multiples méthodologies ont été développées pour déterminer l’état physiologique d’un individu. Plus spécifiquement, la variation de la température faciale, le clignotement des yeux et la réponse galvanique de la peau ont été étudiés.Plusieurs scénarios d'interaction homme-robot ont été conçus afin de tester le système développé. L'impact de l'empathie a également été étudié. En outre, le système développé a été testé avec succès dans deux environnements réels, avec deux populations vulnérables. La première application d'assistance est représentée par le projet de recherche EU H2020 ENRICHME, dans lequel un robot a été développé pour les personnes âgées atteintes d'un trouble cognitif léger. La deuxième population vulnérable est constituée d'individus souffrant de différents troubles du sommeil.Nous pensons que cette thèse représente une étape importante dans la compréhension de l'état physiologique de l'individu et est liée à la performance cognitive. / Social robotic systems are more and more present in our everyday lives. They are no longer isolated entities, but instead, they are expected to be capable of interacting and communicating with humans. They have to follow the behavioral norms that are expected by the individuals the robotic systems are interacting with.One of the main research directions in the field of social robotics is represented by the design of a natural interaction between a social robot and an individual. More specifically, this interaction should take into consideration the profile of the individual, the emotional state, the physiological internal state, and the mood, among others.In this thesis it is explored the relationship that exists between morningness-eveningness, cognitive performance, and the internal physiological state during different human-robot interaction scenarios. By administering different psychological questionnaires, the profile of an individual can be determined. Moreover, with the help of different sensors (e.g., GSR, thermal camera), multiple methodologies were developed to determine the internal physiological state of an individual. More specifically, the facial temperature variation, the blinking, and the galvanic skin response were investigated.Several human-robot interaction scenarios have been designed in order to test the developed system. The impact of empathy was also investigated. Furthermore, the developed system was successfully tested in two real-world environments, with two vulnerable populations. The first assistive application is represented by the ENRICHME EU H2020 research project, where a personal robot was developed for the elderly with mild cognitive impairment. The second vulnerable population consists of individuals suffering from different sleep disorders.We believe that this thesis represents an important step in understanding how the physiological internal state of an individual is related to cognitive performance, and to the user profile of that individual.