Autonomous Recognition of Collective Behaviour in Robot Swarms.

Developmental evolution of collective swarm behaviours promises new ways to evolve swarms with different movement characteristics. Preliminary work has developed value functions that can recognize emergent swarm behaviour and distinguish it from random behaviour in point-mass boid simulations. This paper examines the performance of several variants of such functions recognizing the emergent behaviour of simulated robots, which have different movement properties to point-mass boid simulations as they are constrained by the manoeuvrability of the physical robot. We designed two boid guidance algorithms for controlling Pioneer3DX robots. Five value functions were then examined and compared for their ability to distinguish swarming behaviour from unstructured behaviour. Results show that four of these can be used to distinguish structured collective behaviours of the robots and distinguish such behaviour from random movement patterns.