No description available.

No description available.

No description available.

Thesis (Ph. D.)--Ohio University, November, 1989. / Title from PDF t.p.

Thesis (M.S.)--Ohio University, June, 1990. / Title from PDF t.p.

Thesis (M.S.)--West Virginia University, 1999. / Title from document title page. Document formatted into pages; contains viii, 122 p. : ill. Includes abstract. Includes bibliographical references (p. 57-59).

Thesis (M.S. in Physics)-- Naval Postgraduate School, June 2009. / Thesis Advisor(s): Harkins, Richard ; Haegel, Nancy. "June 2009." Description based on title screen as viewed on July 13, 2009. Author(s) subject terms: Robotic Vision, Unmanned Ground Vehicle. Includes bibliographical references (p. 69-70). Also available in print.

Thesis (Ph.D.)--Hong Kong University of Science and Technology, 2007. / Includes bibliographical references (leaves 112-121). Also available in electronic version.

The main objective of my thesis is to design the constuction of strain gauge mea-surring head. This constrction allows testing float in the pool and facilitate the job of a human operator. By measuring we determined the physical properties of the float. The construction is mounted on the robot KUKA. I chose suitable sensor single-point for measuring buoyacy and tensile strength float on the surface. Strain sensor has an cover IP67. Material of the supportng structure is made of aluminium alloy from company MayTec. The sensors are located in the axis of the center of gravity of the float, which i experimentally calculated. From my experience, character of the float in work on the surface and under load be expected change center of gravity. Construction is adapted for this situation. Final product is light and durable construction with precise sensors and low price.

Les avancées récentes en robotique inspirent des visions de robots domestiques et de service rendant nos vies plus faciles et plus confortables. De tels robots pourront exécuter différentes tâches de manipulation d'objets nécessaires pour des travaux de ménage, de façon autonome ou en coopération avec des humains. Dans ce rôle de compagnon humain, le robot doit répondre à de nombreuses exigences additionnelles comparées aux domaines bien établis de la robotique industrielle. Le but de la planification pour les robots est de parvenir à élaborer un comportement visant à satisfaire un but et qui obtient des résultats désirés et dans de bonnes conditions d'efficacité. Mais dans l'interaction homme-robot (HRI), le comportement robot ne peut pas simplement être jugé en termes de résultats corrects, mais il doit être agréable aux acteurs humains. Cela signifie que le comportement du robot doit obéir à des critères de qualité supplémentaire. Il doit être sûr, confortable pour l'homme, et être intuitivement compris. Il existe des pratiques pour assurer la sécurité et offrir un confort en gardant des distances suffisantes entre le robot et des personnes à proximité. Toutefois fournir un comportement qui est intuitivement compris reste un défi. Ce défi augmente considérablement dans les situations d'interaction homme-robot dynamique, où les actions de la personne sont imprévisibles, le robot devant adapter en permanence ses plans aux changements. Cette thèse propose une approche nouvelle et des méthodes pour améliorer la lisibilité du comportement du robot dans des situations dynamiques. Cette approche ne considère pas seulement la qualité d'un seul plan, mais le comportement du robot qui est parfois le résultat de replanifications répétées au cours d'une interaction. Pour ce qui concerne les tâches de navigation, cette thèse présente des fonctions de coûts directionnels qui évitent les problèmes dans des situations de conflit. Pour la planification d'action en général, cette thèse propose une approche de replanification locale des actions de transport basé sur les coûts de navigation, pour élaborer un comportement opportuniste adaptatif. Les deux approches, complémentaires, facilitent la compréhension, par les acteurs et observateurs humains, des intentions du robot et permettent de réduire leur confusion. / The recent advances in robotics inspire visions of household and service robots making our lives easier and more comfortable. Such robots will be able to perform several object manipulation tasks required for household chores, autonomously or in cooperation with humans. In that role of human companion, the robot has to satisfy many additional requirements compared to well established fields of industrial robotics. The purpose of planning for robots is to achieve robot behavior that is goal-directed and establishes correct results. But in human-robot-interaction, robot behavior cannot merely be judged in terms of correct results, but must be agree-able to human stakeholders. This means that the robot behavior must suffice additional quality criteria. It must be safe, comfortable to human, and intuitively be understood. There are established practices to ensure safety and provide comfort by keeping sufficient distances between the robot and nearby persons. However providing behavior that is intuitively understood remains a challenge. This challenge greatly increases in cases of dynamic human-robot interactions, where the actions of the human in the future are unpredictable, and the robot needs to constantly adapt its plans to changes. This thesis provides novel approaches to improve the legibility of robot behavior in such dynamic situations. Key to that approach is not to merely consider the quality of a single plan, but the behavior of the robot as a result of replanning multiple times during an interaction. For navigation planning, this thesis introduces directional cost functions that avoid problems in conflict situations. For action planning, this thesis provides the approach of local replanning of transport actions based on navigational costs, to provide opportunistic behavior. Both measures help human observers understand the robot's beliefs and intentions during interactions and reduce confusion.