Pushing the Envelope

Abstract

In the prior chapters, we modeled physics faithfully and limited the amount of information that was exchanged between robots. However, for some applications, performance can be improved by carefully modifying the underlying physics assumptions, introducing new physics, or slightly increasing the amount of information. This generalizes physicomimetics and expands its behavioral repertoire. This chapter demonstrates the increased repertoire by considering four different applications. First, we examine how a swarm can self-organize into globally perfect formations. Second, we consider the special situation where we want an initially disorganized swarm to self-organize into a chain formation that serves as a long communication relay. Third, we provide a solution to the problem of uniform coverage, where robots must search all areas of an environment equally. Finally, a different form of physics referred to as “kinetic theory” is used to control a swarm of robots to perform surveillance sweeps of an environment. All of these applications are accompanied by instructional NetLogo simulations. The latter three applications are covered in much greater detail later in this book.