Developing an improved understanding and awareness of how gender impacts perceptions of robots and interactions with them is crucial for the ongoing advancement of the human-robot interaction (HRI) field, as a lack of awareness of gender issues increases the risk of robot rejection and poor performance. This thesis provides a theoretical grounding for gender-studies in HRI, and contributes to the understanding of how gender affects attitudes toward and interaction with robots via the findings from an on-line survey and a laboratory user study. We envision that this work will provide HRI designers with a foundation and exemplary account of how gender can influence attitudes toward and interaction with robots, serving as a resource and a sensitizing discussion for gender studies in HRI.

"Human-Robot Interaction (HRI) is a rapidly expanding field of study that focuses on allowing non-roboticist users to naturally and effectively interact with robots. The importance of conducting extensive user studies has become a fundamental component of HRI research; however, due to the nature of robotics research, such studies often become expensive, time consuming, and limited to constrained demographics. This work presents the Robot Management System, a novel framework for bringing robotic experiments to the web. A detailed description of the open source system, an outline of new security measures, and a use case study of the RMS as a means of conducting user studies is presented. Using a series of navigation and manipulation tasks with a PR2 robot, three user study conditions are compared: users that are co-present with the robot, users that are recruited to the university lab but control the robot from a different room, and remote web-based users. The findings show little statistical differences between usability patterns across these groups, further supporting the use of web-based crowdsourcing techniques for certain types of HRI evaluations."

An important challenge for human-robot interaction is for both the human and the robot to agree on a model of motion. However, most applications require machine learning and heavy communication for the robot to be an active member in executing a joint task and adapt to human behavior. In this work, we develop a decentralized approach to control and track a catenary cable for object transportation in a human robot interaction scope. Our system is composed of a linked chain that is attached to a quadrotor on one end, and the human on the other. The chain is defined as a catenary curve with five degrees of freedom, and Motion Capture technology is used to track the components of our system. Given the human’s position, we use shape estimation of the curve to determine the drone’s position and control the trajectory of the chain and thus the load attached to it. We then proceed to implement a swing load controller that minimizes the oscillations of the load created by the chain’s movement.

abstract: What if there is a way to integrate prosthetics seamlessly with the human body and robots could help improve the lives of children with disabilities? With physical human-robot interaction being seen in multiple aspects of life, including industry, medical, and social, how these robots are interacting with human becomes even more important. Therefore, how smoothly the robot can interact with a person will determine how safe and efficient this relationship will be. This thesis investigates adaptive control method that allows a robot to adapt to the human's actions based on the interaction force. Allowing the relationship to become more effortless and less strained when the robot has a different goal than the human, as seen in Game Theory, using multiple techniques that adapts the system. Few applications this could be used for include robots in physical therapy, manufacturing robots that can adapt to a changing environment, and robots teaching people something new like dancing or learning how to walk after surgery. The experience gained is the understanding of how a cost function of a system works, including the tracking error, speed of the system, the robot’s effort, and the human’s effort. Also, this two-agent system, results into a two-agent adaptive impedance model with an input for each agent of the system. This leads to a nontraditional linear quadratic regulator (LQR), that must be separated and then added together. Thus, creating a traditional LQR. This new experience can be used in the future to help build better safety protocols on manufacturing robots. In the future the knowledge learned from this research could be used to develop technologies for a robot to allow to adapt to help counteract human error. / Dissertation/Thesis / Masters Thesis Engineering 2019

One of the problems affecting robot operators' spatial awareness involves their ability to infer a robot's location based on the views from on-board cameras and other electro-optic systems. To understand the vehicle's location, operators typically need to translate images from a vehicle's camera into some other coordinates, such as a location on a map. This translation requires operators to relate the view by mentally rotating it along a number of axes, a task that is both attention-demanding and workload-intensive, and one that is likely affected by individual differences in operator spatial abilities. Because building and maintaining spatial awareness is attention-demanding and workload-intensive, any variable that changes operator workload and attention should be investigated for its effects on operator spatial awareness. One of these variables is the use of automation (i.e., assigning functions to the robot). According to Malleable Attentional Resource Theory (MART), variation in workload across levels of automation affects an operator's attentional capacity to process critical cues like those that enable an operator to understand the robot's past, current, and future location. The study reported here focused on performance aspects of human-robot interaction involving ground robots (i.e., unmanned ground vehicles, or UGVs) during reconnaissance tasks. In particular, this study examined how differences in operator spatial ability and in operator workload and attention interacted to affect spatial awareness during human-robot interaction (HRI). Operator spatial abilities were systematically manipulated through the use of mental transformation training. Additionally, operator workload and attention were manipulated via the use of three different levels of automation (i.e., manual control, decision support, and full automation). Operator spatial awareness was measured by the size of errors made by the operators, when they were tasked to infer the robot's location from on-board camera views at three different points in a sequence of robot movements through a simulated military operation in urban terrain (MOUT) environment. The results showed that mental transformation training increased two areas of spatial ability, namely mental rotation and spatial visualization. Further, spatial ability in these two areas predicted performance in vehicle localization during the reconnaissance task. Finally, assistive automation showed a benefit with respect to operator workload, situation awareness, and subsequently performance. Together, the results of the study have implications with respect to the design of robots, function allocation between robots and operators, and training for spatial ability. Future research should investigate the interactive effects on operator spatial awareness of spatial ability, spatial ability training, and other variables affecting operator workload and attention.

In the process of Human-Robot Interaction, improving the intelligibility of robots is crucial. Intelligibility refers to the degree to which humans can understand robot behavior and decision-making. When humans interact with low-intelligibility robots, it can lead to a series of problems, such as misunderstanding and trust issues. For old people, various impairments may occur in their physiological functions. Therefore, in the process of Human-Robot Interaction, when the target group is older adults, it is more critical to improve intelligibility. In this thesis project, we propose a solution to improve the intelligibility of robots when interacting with older adults. Specifically, we design an intelligent dialogue system that combines the Rasa dialogue management assistant, the Pepper robot, and the "Nerve" module to provide cognitive exercises. Among them, the user model is used to expand the Rasa dialogue management assistant to adjust the difficulty of cognitive exercises according to the user's hearing, seeing, and cognitive impairment. Rasa could interact with older adults through the Pepper robot. At the same time, we also consider environmental factors in Human-Robot Interaction, such as lighting and noise. For this purpose, we design a "Nerve" module that could adjust the communication modality and parameters during interaction according to environmental factors and user models. We also use some designed personas to evaluate the intelligent system from the five perspectives of speech recognition accuracy, user experience, system stability, intelligibility, and robustness. Finally, we analyze the limitations of the system properties and give a brief introduction to future work.

Transparency in HRI describes the method of making the current state of a robotor intelligent agent understandable to a human user. Applying transparencymechanisms to robots improves the quality of interaction as well as the userexperience. Explanations are an effective way to make a robot’s decision making transparent. We introduce a framework that uses natural language labels attached to a region inthe continuous state space of the robot to automatically generate local explanationsof a robot’s policy. We conducted a pilot study and investigated how the generated explanations helpedusers to understand and reproduce a robot policy in a debugging scenario.

Worldwide, about 35.7 million people were estimated to be affected by dementia in 2010. One way to treat dementia is by exercising, but human trainers are few and expensive. Robots can be mass produced and work at anytime of the day. This report describes research done for developing a robot exercise coach intended for treating dementia. Three main problems for people with dementia were identified: memory, attention and motivation. By using computer vision the robot can help count repetitions, grade exercise correctness and make sure that the user is still paying attention. The Kinect was used for skeleton tracking to count repetitions and provide video. For motivation, motivational models and flow theory were used to design the users interaction with the robot and make it more enjoyable and engaging. Feedback was believed to be an important part of this interaction. To provide extra feedback skeleton tracking was turned into the robot mimicking the user. To test which combination of feedback and interaction was most enjoyable, a user study was done. The user study consisted of 11 subjects, each interacting with three different systems, each system with varying levels of feedback. After interacting, the subject filled out a survey and had an interview. The results from the user study showed evidence that repetition counting and exercise correctness feedback but no mimicking is the most enjoyable. With a statistically significant difference in regards to repetition counting at the 0.05 level. Younger people found the mimicking enjoyable but still preferred the system without it, and older people found it confusing. In future systems like this, repetition counting and exercise correctness feedback should be seen as important parts of the interaction.

With the increasing longevity that accompanies advances in medical technology comes a host of other age-related disabilities. Among these are neuro-degenerative diseases such as Alzheimer's disease, Parkinson's disease, and stroke, which significantly reduce the motor and cognitive ability of affected individuals. As these diseases become more prevalent, there is a need for further research and innovation in the field of motor rehabilitation therapy to accommodate these individuals in a cost-effective manner. In recent years, the implementation of social agents has been proposed to alleviate the burden on in-home human caregivers. Socially assistive robotics (SAR) is a new subfield of research derived from human-robot interaction that aims to provide hands-off interventions for patients with an emphasis on social rather than physical interaction. As these SAR systems are very new within the medical field, there is no standardized approach to developing such systems for different populations and therapeutic outcomes. The primary aim of this project is to provide a standardized method for developing such systems by introducing a modular human-robot interaction software framework upon which future implementations can be built. The framework is modular in nature, allowing for a variety of hardware and software additions and modifications, and is designed to provide a task-oriented training structure with augmented feedback given to the user in a closed-loop format. The framework utilizes the ROS (Robot Operating System) middleware suite which supports multiple hardware interfaces and runs primarily on Linux operating systems. These design requirements are validated through testing and analysis of two unique implementations of the framework: a keyboard input reaction task and a reaching-to-grasp task. These implementations serve as example use cases for the framework and provide a template for future designs. This framework will provide a means to streamline the development of future SAR systems for research and rehabilitation therapy.

A Robótica Educacional consiste na utilização de robôs para aplicação prática dos conteúdos teóricos discutidos em sala de aula. Porém, os robôs mais usados apresentam uma carência de interação com os usuários, a qual pode ser melhorada com a inserção de robôs humanoides. Esta dissertação tem como objetivo a combinação de técnicas de visão computacional, robótica social e reconhecimento e síntese de fala para a construção de um sistema interativo que auxilie em sessões pedagógicas por meio de um robô humanoide. Diferentes conteúdos podem ser abordados pelos robôs de forma autônoma. Sua aplicação visa o uso do sistema como ferramenta de auxílio no ensino de matemática para crianças. Para uma primeira abordagem, o sistema foi treinado para interagir com crianças e reconhecer figuras geométricas 3D. O esquema proposto é baseado em módulos, no qual cada módulo é responsável por uma função específica e contém um grupo de funcionalidades. No total são 4 módulos: Módulo Central, Módulo de Diálogo, Módulo de Visão e Módulo Motor. O robô escolhido é o humanoide NAO. Para visão computacional, foram comparados a rede LEGION e o sistema VOCUS2 para detecção de objetos e SVM e MLP para classificação de imagens. O reconhecedor de fala Google Speech Recognition e o sintetizador de voz do NAOqi API são empregados para interações sonoras. Também foi conduzido um estudo de interação, por meio da técnica de Mágico-de-Oz, para analisar o comportamento das crianças e adequar os métodos para melhores resultados da aplicação. Testes do sistema completo mostraram que pequenas calibrações são suficientes para uma sessão de interação com poucos erros. Os resultados mostraram que crianças que tiveram contato com uma maior interatividade com o robô se sentiram mais engajadas e confortáveis nas interações, tanto nos experimentos quanto no estudo em casa para as próximas sessões, comparadas às crianças que tiveram contato com menor nível de interatividade. Intercalar comportamentos desafiadores e comportamentos incentivadores do robô trouxeram melhores resultados na interação com as crianças do que um comportamento constante. / Educational Robotics is a growing area that uses robots to apply theoretical concepts discussed in class. However, robots usually present a lack of interaction with users that can be improved with humanoid robots. This dissertation presents a project that combines computer vision techniques, social robotics and speech synthesis and recognition to build an interactive system which leads educational sessions through a humanoid robot. This system can be trained with different content to be addressed autonomously to users by a robot. Its application covers the use of the system as a tool in the mathematics teaching for children. For a first approach, the system has been trained to interact with children and recognize 3D geometric figures. The proposed scheme is based on modules, wherein each module is responsible for a specific function and includes a group of features for this purpose. In total there are 4 modules: Central Module, Dialog Module, Vision Module and Motor Module. The chosen robot was the humanoid NAO. For the Vision Module, LEGION network and VOCUS2 system were compared for object detection and SVM and MLP for image classification. The Google Speech Recognition speech recognizer and Voice Synthesizer Naoqi API are used for sound interactions. An interaction study was conducted by Wizard-of-Oz technique to analyze the behavior of children and adapt the methods for better application results. Full system testing showed that small calibrations are sufficient for an interactive session with few errors. Children who had experienced greater interaction degrees from the robot felt more engaged and comfortable during interactions, both in the experiments and studying at home for the next sessions, compared to children who had contact with a lower level of interactivity. Interim challenging behaviors and support behaviors brought better results in interaction than a constant behavior.