The overall objective of this thesis is to build an integrated, inexpensive, human-sized humanoid robot from scratch that looks and behaves like a human. More specifically, my goal is to build an android robot called Marie Curie robot that can act like a human actor in the Portland Cyber Theater in the play Quantum Debate with a known script of every robot behavior. In order to achieve this goal, the humanoid robot need to has degrees of freedom (DOF) similar to human DOFs. Each part of the Curie robot was built to achieve the goal of building a complete humanoid robot. The important additional constraints of this project were: 1) to build the robot from available components, 2) to minimize costs, and 3) to be simple enough that the design can be replicated by non-experts, so they can create robot theaters worldwide. Furthermore, the robot appears lifelike because it executes two main behaviors like a human being. The first behavior is tracking where the humanoid robot uses a tracking algorithm to follow a human being. In other words, the tracking algorithm allows the robot to control its neck using the information taken from the vision system to look at the nearest human face. In addition, the robot uses the same vision system to track labeled objects. The second behavior is grasping where the inverse kinematics (IK) is calculated so the robot can move its hand to a specific coordinate in the surrounding space. IK gives the robot the ability to move its end-effector (hand) closer to how humans move their hands.

Thesis (Ph.D)--Psychology, Georgia Institute of Technology, 2010. / Committee Chair: Richard Catrambone; Committee Member: Gregory Corso; Committee Member: Jack Feldman; Committee Member: John T. Stasko; Committee Member: Wendy A. Rogers. Part of the SMARTech Electronic Thesis and Dissertation Collection.

The objective of this research effort is to further rehabilitation techniques for children by developing and validating the core technologies needed to integrate therapy instruction with child-robot play interaction in order to improve upper-arm rehabilitation. Using computer vision techniques such as Motion History Imaging (MHI), Multimodal Mean, edge detection, and Random Sample Consensus (RANSAC), movements can be quantified through robot observation. Also incorporating three-dimensional data obtained via an infrared projector coupled with a Principle Component Analysis (PCA), depth information can be utilized to create a robust algorithm. Finally, utilizing prior knowledge regarding exercise data, physical therapeutic metrics, and novel approaches, a mapping to therapist instructions can be created allowing robotic feedback and intelligent interaction.

Orientadores: Antonio Celso Fonseca de Arruda, Paulo Roberto Gardel Kurka / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-27T18:15:09Z (GMT). No. of bitstreams: 1 Quiroz_CesarHenriqueCordova_M.pdf: 12262976 bytes, checksum: 366530cff1ef38066096c0ad2e8735fe (MD5) Previous issue date: 2015 / Resumo: O trabalho desenvolveu um conjunto de ferramentas simples (interface gráfica e comunicação remota) que permitiram o controle e monitoração dos diferentes sensores e atuadores embarcados em uma cadeira de rodas elétrica. A solução encontrada para realizar os comandos necessários, de modo remoto, foi utilizar um dispositivo móvel (smartphone) baseado em sistema operacional Android. Os principais aspectos considerados no desenvolvimento foram utilizar o hardware e o software que propiciaram uma interface de interação entre o usuário e os atuadores presentes na cadeira de rodas. Adicionalmente foi previsto e viabilizada a possibilidade de um acompanhante efetuar comandos, por meio de outro celular móvel, segundo critérios de prioridade inseridos na programação. A opção por utilizar dispositivos móveis convencionais está vinculada com a capacidade de recorrer, por meio de aplicativos específicos, aos diversos recursos (bibliotecas) que são embarcadas em um telefone móvel ou tablet. O aplicativo criado utilizou o sistema operacional Android pela facilidade de acesso ao código fonte. O aplicativo possibilita o uso dos botões presentes nos smartphones para cumprir tarefas de controle direcional e de velocidade. Adicionalmente sensores ultrassônicos , instalados na cadeira, impedem a colisão com obstáculos presentes no trajeto. Os testes realizados demonstraram a eficácia da movimentação da cadeira, do controle de velocidade e da atuação dos sensores de colisão. A proposta inicial de uma cadeira de rodas elétrica a ser utilizada por crianças, foi estendida para adultos, o que foi comprovado em teste realizado com uma pessoa de 80 quilogramas. A velocidade máxima final independente da carga foi de 0,8m/s (2,88km/h). Na programação do aplicativo foi considerado e implementado um controle para impedir o tranco no início da movimentação, objetivando o fator ergonômico do usuário / Abstract: The work developed a set of simple tools (graphical user interface and remote communication) that allowed the control and monitoring of different sensors and actuators embedded in a electric wheelchair. The solution to make necessary commands remotely, was to use a mobile device (smartphone) based operating system Android. The main aspects considered in the development were using the hardware and software that provided an interaction interface between the user and actuators in the wheelchair. Additionally it was intended and the possibility of a viable accompanying perform controls by means of another mobile phone according to priority criteria included in the programming. The option to use conventional mobile devices is linked with the capacity to use through specific applications, the various resources (libraries) that are embedded in a mobile phone or tablet. The application created used the Android operating system for ease of access to the source code. The application enables the use of buttons present in the smartphones to fulfill directional control and velocity. Additionally ultrasonic sensors installed in the chair, prevent collision with obstacles present in the path. The tests demonstrated the effectiveness of the movement of the chair, speed control and the performance of the collision sensors. The initial proposal of an electric wheelchair to be used by children, has been extended to adults, which has been proven in test performed with a person of 80kg. The independent final maximum speed of the load was $0.8m/s$ or (2,88km/h). In the application programming was considered and implemented a control to prevent stride in the early movement, aiming the ergonomic factor user / Mestrado / Materiais e Processos de Fabricação / Mestre em Engenharia Mecânica

Most of work on motion of mobile robots is to generate plans for avoiding obstacles or perform some meaningful and useful actions. In modern robot theatres and entertainment robots the motions of the robot are scripted and thus the performance or behavior of the robot is always the same. In this work we want to propose a new approach to robot motion generation. We want our robot to behave more like real people. People do not move in mechanical way like robots. When a human is supposed to execute some motion, these motions are similar to one another but always slightly or not so slightly different. We want to reproduce this property based on the introduced by us new concept of probabilistic regular expression, a method to describe sets of interrelated similar actions instead of single actions. Our goal is not only to create motions for humanoid robots that will look more naturally and less mechanically, but also to program robots that will combine basic movements from certain library in many different and partially random ways. While the basic motions were created ahead of time, their combinations are specified in our new language. Although now our method is only for motions and does not take inputs from sensors into account, in future the language can be extended to input/output sequences, thus the robot will be able to adapt the motion in different ways, to some sets of sequences of input stimuli. The inputs will come from sensors, possibly attached to limbs of controlling humans from whom the patterns of motion will be acquired.

This paper analyzes the blurred boundaries between human and artificial intelligence in Philip K. Dick’s novel Do Androids Dream of Electric Sheep. A postmodern theoretical and critical approach that employs Jean Baudrillard’s concepts of simulacra and hyperreality and Jaques Derrida’s deconstructive theory provides the conceptual framework for the analysis. The primary focus is on the main character, Rick Deckard, as he grapples with identity, ethics, and the very nature of humanity in a world where androids are indistinguishable from humans. The essay identifies two focus points in the novel. The first analyzes and deconstructs the real versus artificial (human/android) dichotomy in the book and shows how isreconstructs the essence of identity and reality. The second focus point is the novel’s portrayal of empathy, supposedly a defining human trait and yet one mimicked by androids so well that it is practically useless as a criterion for distinguishing androids from humans. The conclusion drawn by this analysis is that Rick Deckard and the characters he meets may indeed illustrate Baudrillard’s hyperreality, depicting a world where humans are willfully stunting their own emotions and autonomy, lost in the false reality that society has constructed. Furthermore, the ethical dilemmas raised by playing God with life, even if it is artificial, align with Derrida’s deconstructive views of how non-binary all life is. This is especially seen in the contrasting depictions of Rachael (android) and Resch (human). Derrida’s views on humanity’s dissimulation of the cruelty of its exploitations, justified by humanity’s own parameters for what is deemed right and wrong, is also seen in Deckard’s moral struggles and the cruelty androids must endure as described by Rachael. This is lastly compared and contrasted with the contemporary development of AI and its potential dangers, shedding light on ethical considerations. While the AI available to us today is far from the kind of artificial intelligence Dick presents in the form of androids, it is, perhaps, not too soon for us to begin considering the moral and ethical implications now so that should the day come, we will be ready and avoid the crisis humanity has fallen into in Do Androids Dream of Electric Sheep.

In my thesis, Philip K. Dick's Do Androids Dream of Electric Sheep? is examined using French psychoanalyst Jacques Lacan's mirror stage theory. In the novel, humans have built androids that are almost indistinguishable from humans except that they lack a sense of empathy, or so the humans believe. The Voigt-Kampff Machine is a polygraph-like device used to determine if a subject shows signs of empathy in order to confirm if one is an android or a human. Yet, should empathy be the defining quality of determining humanity? In his article "The mirror stage as formative of the function of the ‘I’ as revealed in psychoanalytic experience," Lacan refers to a particular critical milestone in an infant's psychological development. When the baby looks in a mirror, they come to the realization that the image they are seeing is not just any ordinary image; it is actually themselves in the mirror. This "a-ha" moment of self-realization is what Lacan's Mirror Stage Theory is based on. According to Lacan's theory, the image that the child sees in a mirror becomes an "Other" through which they will always scrutinize and pass judgment on, for it is not how they have pictured themselves to be in their mind’s eye. I hypothesize that the androids are humans' artificial and technological Other. It is my thought that Dick uses the conflict of determining the biological from the artificial, the effort to differentiate humans from androids and biological animals from artificial ones, to illustrate Lacan's psychoanalysis of the mirror stage and its importance in our continual search for determining what humanity is and who we really are.

An algorithm for generating communicative, human-like motion for social humanoid robots was developed. Anticipation, exaggeration, and secondary motion were demonstrated as examples of communication. Spatiotemporal correspondence was presented as a metric for human-like motion, and the metric was used to both synthesize and evaluate motion. An algorithm for generating an infinite number of variants from a single exemplar was established to avoid repetitive motion. The algorithm was made task-aware by including the functionality of satisfying constraints. User studies were performed with the algorithm using human participants. Results showed that communicative, human-like motion can be harnessed to direct partner attention and communicate state information. Furthermore, communicative, human-like motion for social robots produced by the algorithm allows humans partners to feel more engaged in the interaction, recognize motion earlier, label intent sooner, and remember interaction details more accurately.

This body of work explores an emerging aspect of human-robot interaction, transparency. Socially guided machine learning has proven that highly immersive robotic behaviors have yielded better results than lesser interactive behaviors for performance and shorter training time. While other work explores this transparency in learning by demonstration using non-verbal cues to point out the importance or preference users may have towards behaviors, my work follows this argument and attempts to extend it by offering cues to the internal task representation. What I show is that task-transparency, or the ability to connect and discuss the task in a fluent way implores the user to shape and correct the learned goal in ways that may be impossible by other present day learning by demonstration methods. Additionally, some participants are shown to prefer task-transparent robots which appear to have the ability of "introspection" in which it can modify the learned goal by other methods than just demonstration.

A framework for understanding the social responses to virtual humans suggests that human-like characteristics (e.g., facial expressions, voice, expression of emotion) act as cues that lead a person to place the agent into the category "human" and thus, elicit social responses. Given this framework, this research was designed to answer two outstanding questions that had been raised in the research community (Moon&Nass, 2000): 1) If a virtual human has more human-like characteristics, will it elicit stronger social responses from people? 2) How do the human-like characteristics interact in terms of the strength of social responses? Two social psychological (social facilitation and politeness norm) experiments were conducted to answer these questions. The first experiment investigated whether virtual humans can evoke a social facilitation response and how strong that response is when participants are given different cognitive tasks (e.g., anagrams, mazes, modular arithmetic) that vary in difficulty. They did the tasks alone, in the company of another person, or in the company of a virtual human that varied in terms of features. The second experiment investigated whether people apply politeness norms to virtual humans. Participants were tutored and quizzed either by a virtual human tutor that varied in terms of features or a human tutor. Participants then evaluated the tutor's performance either directly by the tutor or indirectly via a paper and pencil questionnaire. Results indicate that virtual humans can produce social facilitation not only with facial appearance but also with voice recordings. In addition, performance in the presence of voice synced facial appearance seems to elicit stronger social facilitation (i.e., no statistical difference compared to performance in the human presence condition) than in the presence of voice only or face only. Similar findings were observed with the politeness norm experiment. Participants who evaluated their tutor directly reported the tutor's performance more favorably than participants who evaluated their tutor indirectly. In addition, this valence toward the voice synced facial appearance had no statistical difference compared to the valence toward the human tutor condition. The results suggest that designers of virtual humans should be mindful about the social nature of virtual humans.