TITLE: A Distributed Architecture for Mobile Navigation

PRESENTER: Julio Rosenblatt

AFFILIATION: Australian Centre for Field Robotics, The University of Sydney

DATE: Friday October 5th 2001

TIME: 12:00 - 13:00

PLACE: CSE K17 1st Floor Seminar Room

ABSTRACT: An architecture is presented in which distributed decision-making processes, or behaviours, cooperatively determine a mobile robot's path by sending votes to a centralized arbiter. One type of arbiter defined within this architecture performs a form of command fusion similar to fuzzy logic, thus allowing multiple goals and constraints to be considered simultaneously. Another type of arbiter performs action selection via a new means of action selection known as utility fusion, in which behaviors indicate the utility of various possible world states and the arbiter determines the next action based on the maximization of expected utility in a decision-theoretical framework. The centralized arbitration of votes from distributed behaviors provides coherent, rational, goal-directed behavior while preserving real-time responsiveness to its immediate physical environment. Experimental results are provided for an autonomous land-based vehicle, as well as underwater one, and the efficacy of the two approaches to decision-making are compared.

BIOGRAPHY OF SPEAKER:

Dr. Julio Rosenblatt is a Senior Research Associate at the Australian Centre for Field Robotics at Sydney University, and is an Australian Research Fellow. He received his Ph.D. and M.S. degrees in Robotics from Carnegie Mellon University in 1996 and 1991, respectively, and the S.B. in Computer Science and Engineering from the Massachusetts Institute of Technology in 1985. He was formerly Director of the Autonomous Mobile Robotics Laboratory at the University of Maryland, College Park, and a research scientist at the Hughes Research Labs in Malibu, CA. He is currently investigating co-operative robotics, as well as the combination of quantitative and qualitative methods for mobile robot navigation.

Seminar Convenor:

Bernhard Hengst
Tel: +61-2-9385-3988,
E-mail: bernhardh@cse.unsw.edu.au
School of Computer Science & Engineering, UNSW.