Schedule of Papers and Events

To accomplish the next generation of challenging missions to the Moon, Mars, and beyond, researchers at NASA centers and at academic institutions have made significant progress over the last five years towards developing autonomous systems that can make critical decisions independently of human operators. Autonomy technology will extend the boundary on what can be accomplished in future missions by overcoming limitations due to communications delays, light-speed constraints, mission complexity, and cost. Autonomous systems will enable future space missions by maintaining vehicle health and safety, accomplishing complex science and mission goals, and adapting to changing circumstances or opportunities.

This talk will provide an overview of the current state of autonomy technology applied to deep space exploration, with particular emphasis placed on robotic surface explorers. First, I will motivate and describe the notion of autonomy as an enabler for deep space exploration. Second, I will discuss the contribution of Artificial Intelligence in current autonomy architectures, illustrating how automated systems for planning, plan execution and health management are being integrated into traditional control systems. Third, I will illustrate the challenges for developing autonomous systems through a case study of involving a Mars scenario, in which a rover is required to traverse to a target of interest, extend a flexible arm and acquire a close-up image. The ability to perform this scenario autonomously required advances along a broad technological front, including technology related to target tracking and instrument placement, planning and execution, and automated ground tools for coordinating with science teams. Finally, I will discuss the future challenges in developing autonomy technology for realizing NASA's goal for humans to return to the Moon and eventually establish a permanent presence on Mars.

Dr. Robert Morris is a Computer Science researcher in the planning and scheduling group in the Intelligent Systems division at NASA Ames Research Center. He is part of NASA's Intelligent Systems program, managing a number of projects that build automated reasoning systems for mission operations and autonomy. Currently, he is technical lead in a project to develop an observation scheduling system for constellations of Earth orbiting sensors. He received a B.A from the University of Minnesota, a Ph.D. in philosophy from Indiana University and an M.S. in Computer Science from Wright State University. The focus of his research is the area of temporal reasoning for planning and scheduling.

Representation is a fluent. A mismatch between the real world and an agent's representation of it can be signalled by unexpected failures (or successes) of the agent's reasoning. Such mismatches can be repaired by refining or abstracting an agent's representation. These refinements or abstractions may not be limited to changes of belief, but may also change the signature of the agent's ontology. We describe the implementation and successful evaluation of these ideas in the ORS system. ORS diagnoses failures in plan execution and then repairs the faulty ontologies. Such dynamic ontology repair will be an essential tool in realising the vision of the Semantic Web.

Alan Bundy is a professor in the School of Informatics, and was formerly Head of the Division of Informatics, at Edinburgh University. His research is concerned with the analysis of, development of and interaction between, reasoning processes for a wide range of logics, and the application of that work to areas such as proof by mathematical induction, building ecological models, hardware and software verification, and cognitive modelling. He has been sole or joint holder of 39 EPSRC, SERC, Alvey, ESPRIT or ESRC grants, and is the sole or joint author of over 140 published papers and books.

The World Wide Web provides a wealth of data that can be harnessed to help improve information retrieval and increase understanding of the relationships between different entities. In many cases, we are often interested in determining how similar two entities may be to each other, where the entities may be pieces of text, descriptions of some object, or even the preferences of a group of people. In this work, we examine several instances of this problem, and show how they can be addressed by harnessing data mining techniques applied to large web-based data sets. Specifically, we examine the problems of: (1) determining the similarity of short texts--even those that may not share any terms in common, (2) learning similarity functions for semi-structured data to address tasks such as record linkage between objects, and (3) measuring the similarity between on-line communities of users as part of a recommendation system. While we present rather different techniques for each problem, we show how measuring similarity between entities in all these domains has a direct application to the overarching goal of improving information access for users of web-based systems.

Mehran Sahami is a Lecturer in the Computer Science Department at Stanford University, and also a Senior Research Scientist at Google. His research interests include machine learning, data mining, intelligent agents, and information retrieval on the Web.