With this thesis, I would like to lay the foundations for designing human interfacing wearable exoskeleton robots that are truly designed for the human. Before starting the development of the first human arm exoskeleton prototype, tasked to telemanipulate a space robot, I extensively searched for prior art in databases to potentially find guidelines on how to design a wearable robotic exoskeleton. Information was very scarce, however, and I found only a handful of information at all. Previous records have either shown device concepts only, incomplete devices or devices built to interact with only a sub-set of joints of the human arm. No record has provided evidence of a successful robot control with a portable exoskeleton, let alone with force-feedback to the operator. Not even to speak of finding quantitative analyses about the goodness of physical human–robot interaction or about bilateral control performance with such exoskeletons. Most of the reference material rather raised new questions than providing answers to me. I noticed that previous exoskeleton devices had been designed similar to typical robotic manipulators, but aiming to encapsulate the human limb. This was done despite the fact that artificial robotic systems are fundamentally different in structure from biological human limbs. Moreover, all prior exoskeletons had been developed with anthropometric data of specific indivi-duals. This seemed like a wrong approach to me and inspired me to investigate how these fundamental differences between robots and humans can be harmonized. I was motivated to start this scientific research about finding the fundamentals of ergonomic exoskeleton design. Now, a couple of years later, I can present with this thesis a novel quantitative analysis approach for assessing combined physical human–exoskeleton interaction. The theory presented allows the design analysis and evaluation of exoskeletons, and the solutions provided offer a clear set of design guidelines helpful to the community in the future. The guidelines show, on scientific grounds, how to best conceive exoskeleton kinematics, motorization and mechanical structures for enabling smooth and comfortable physical human robot interaction with portable exoskeletons. Technological solutions are proposed as well, that allow conceiving of lightweight exoskeletons with little apparent inertia and good force-feedback performance for robotic telemanipulation. The feasibility of developing a portable and body-grounded exoskeleton for the entire human arm is shown for the first time. It is proven that the device can interact seamlessly with natural motion of the human arm, without variation of its mechanical structure, for different operators. A new paradigm for the design of kinematic exoskeleton structures is developed, as well as a novel actuator concept, based on Bowden Cable transmissions. The results presented in this thesis provide the lacking theoretical fundament and the technological solutions, which together enable the design of physically interacting human–robot systems that are truly conceived for the human.

Категория: Подборки книг | Нет комментариев »

Комментарии