Polygon Formation in Distributed Multi-Agent Systems.

A primary challenge to form a geometric pattern for a fleet of randomly distributed agents is the determination of their locations on the boundary of a formation. The challenge is more acute if the agents need to be uniformly distributed on the formation. This study addresses this challenge for polygon formations via a two-phase procedure. In the first phase, the agents form an enclosing circle to the location of the formation. This prevents the agents from location-conflicts on the polygon and hence each agent can uniquely ascertain its projection point on the formation. In the second phase, the agents establish their projected points and move toward their locations on the polygon, while mitigating collisions. The circle formation is also used as a regrouping feature before the agents reconfigure themselves into a different polygon formation. The formation control laws developed are verified through simulation for circle formation, convex polygons, and some categories of concave polygons. The control laws include the cases for rotation, translation (relocation), and scaling of polygons as well. [ABSTRACT FROM AUTHOR]