"The National Aeronautics and Space Administration (NASA) plans to send a human exploration team to Mars within the next 25 years. In support of this effort Hamilton Standard Space Systems International (HSSSI), current manufacturers of the Space Shuttle spacesuit, began exploring alternative solutions for supporting an astronaut during a Martian surface exploration. A design concept was developed by HSSSI to integrate a minimally equipped Martian spacesuit with a robotic support cart capable of providing life support assistance, communications, and independent navigational functions. To promote NASA's visionary efforts and increase university relations, HSSSI partnered with Worcester Polytechnic Institute (WPI) to develop a proof-of-concept robotic support cart system, FIDOE - Fully Independent Delivery of Expendables. As a proof-of-concept system, the primary goal of this project was to demonstrate the feasibility of current technologies utilized by FIDOE's communication and controls system for future Martian surface explorations. The primary objective of this project was to procure selected commercial-off-the-shelf components and configure these components into a functional robotic support cart. The design constraints for this project, in addition to the constraints imposed by the Martian environment and HSSSI's Martian spacesuit, were a one-year time frame and a $20,000 budget for component procurement. This project was also constrained by the protocols defined by the NASA demonstration test environment. The final design configuration comprised of 37 major commercial off-the-shelf components and three individual software packages that integrated together to provide FIDOE's communications and control capabilities. Power distribution was internally handled through a combination of a main power source and dedicated power supplies. FIDOE also provided a stowage area for handling assisted life support systems and geological equipment. The proof-of-concept FIDOE system proved that the current technologies represented by the selected components are feasible applications for a Mars effort. Specifically, the FIDOE system demonstrated that the chosen technologies can be integrated to perform assisted life support and independent functions. While some technologies represented by the proof-of-concept system may not adequately address the robustness issues pertaining to the Mars effort, e.g., voice recognition and power management, technology trends indicate that these forms of technology will soon become viable solutions to assisting an astronaut on a Martian surface exploration."

This dissertation details the design and development of a mobile autonomous platform from which to conduct a 3D ground-penetrating radar survey. The system uses a three-rover multiagent team to perform a site-selection activity during a lunar analog mission. The work took place beginning in 2008 and culminated in a final field test on Mauna Kea in Hawaii. This demonstration of autonomous acquisition of 3D ground-penetrating radar in a space robtic application is promising not only for in-situ resource utilization, but also for the concept of multiagent teaming.

This dissertation details the design and development of a mobile autonomous platform from which to conduct a 3D ground-penetrating radar survey. The system uses a three-rover multiagent team to perform a site-selection activity during a lunar analog mission. The work took place beginning in 2008 and culminated in a final field test on Mauna Kea in Hawaii. This demonstration of autonomous acquisition of 3D ground-penetrating radar in a space robtic application is promising not only for in-situ resource utilization, but also for the concept of multiagent teaming.

Como tópico de fundamental importância na robótica, o planejamento de rotas tem encontrado excelentes resultados nos últimos anos através da utilização de algoritmos baseados no conceito de árvore de exploração rápida, RRT. No entanto, a aplicação desses métodos em sistemas robóticos espaciais revela um cenário ainda a ser explorado. O comportamento não-holonômico e a presença de singularidades dinâmicas são alguns fatores que dificultam a consideração de obstáculos no planejamento de rotas desses sistemas. Além disso, os trabalhos relacionados ao planejamento de movimento para manipuladores espaciais mostram-se concentrados na estratégia ponto-a-ponto, com interesse especial nos aspectos particulares da dinâmica desses sistemas. De modo geral, para manipuladores espaciais, o planejamento de trajetória envolvendo o desvio de obstáculos depende de uma rota previamente computada. Contudo, essa tarefa carece de formulações ou técnicas solidificadas, sobretudo para manipuladores espaciais de base livre flutuante. Com esta motivação, o trabalho proposto nesta dissertação de mestrado cria um planejador de rotas com suporte a desvio de obstáculos para um manipulador espacial planar de base livre flutuante. O modelo dinâmico utilizado é baseado no conceito de manipulador dinamicamente equivalente e incorporado a um algoritmo baseado no conceito de RRT. / As major challenge in the field of robotics, path planning has experienced successful results in recent years by means of the RRT algorithm. However, the application of such algorithms in space manipulators reveals a scenario yet to be explored. The non-holonomic behavior, added to the presence of dynamic singularities are only a few factors that make collision-avoidance path planning of these systems such a hard task. Besides, works in the field of motion planning of space manipulators often concentrate in the strategy pointto- point, with particular interest in the complex dynamics of such systems. As a rule of thumb, collision-avoidance for space manipulators depends on a previous computed path. However, this task still lacks robust formulations, specially in the case of free-floating manipulators. With this motivation, the proposed work creates a collision-avoiding path planning for a free-floating planar manipulator. The dynamic model is based on the Dynamically Equivalent Manipulator and the concept of Rapidly-Exploring Random Trees serves as a frame for the developed algorithm.

In this thesis, we address the issue of fully localizing a known object through computer vision, using a monocular camera, what is a central problem in robotics. A particular attention is here paid on space robotics applications, with the aims of providing a unified visual localization system for autonomous navigation purposes for space rendezvous and proximity operations. Two main challenges of the problem are tackled: initially detecting the targeted object and then tracking it frame-by-frame, providing the complete pose between the camera and the object, knowing the 3D CAD model of the object. For detection, the pose estimation process is based on the segmentation of the moving object and on an efficient probabilistic edge-based matching and alignment procedure of a set of synthetic views of the object with a sequence of initial images. For the tracking phase, pose estimation is handled through a 3D model-based tracking algorithm, for which we propose three different types of visual features, pertinently representing the object with its edges, its silhouette and with a set of interest points. The reliability of the localization process is evaluated by propagating the uncertainty from the errors of the visual features. This uncertainty besides feeds a linear Kalman filter on the camera velocity parameters. Qualitative and quantitative experiments have been performed on various synthetic and real data, with challenging imaging conditions, showing the efficiency and the benefits of the different contributions, and their compliance with space rendezvous applications.