Distributed Algorithms and Systems of Self-Organizing Robots

From DRLWiki

Revision as of 15:58, 20 October 2006 by Drl
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Our goal is to study self-organizing systems. A self-organizing system consists of multiple autonomous components that make local decisions leading to a global coordinated behavior for the system. Such systems can organize autonomously in different ways, thus adapting to task and environment. Local sensing and communication to nearby neighbors allow the individual units to find their place in the larger system goal.

Self-organizing robots are well-suited for tasks in hazardous and remote environments, especially when the environmental model and the task specifications are uncertain. A collection of simple, modular robots endowed with self-organizing capabilities can conform to the shape of the terrain for locomotion by implementing compliant ``water-flow-like locomotion gaits. They can distribute themselves to form a sensor network for surveillance and monitoring. The robots in such a system can also collaborate to perform sophisticated manipulation tasks. Moreover, a modular platform could carry a collection of self-reconfiguring modules to a site. The modules could then grow into a tower, enter the site through a small opening (such as a window) and reconfigure to survey the site.

To create autonomous robot systems capable of such applications, a fundamental goal of this research is to further the science-base for modular self-organizing systems. This is a considerable challenge, which we propose to meet by the synergistic integration of the following new directions of work with our existing work:

  1. Designing heterogeneous robot systems with self-organization capabilities. In the past we designed the Molecule Robot and the Crystal Robot. Our current focus is to design self-organizing adaptive mobile trusses that consist of active and passive parts.
  2. Focus on distributed and generic rather than architecture-specific algorithms.
  3. Focus on heterogeneous rather than homogeneous systems.
  4. Focus on automating the development and analysis of the algorithms used to control these robots.
Personal tools