We introduce a simulation environment to facilitate research into emergent collective behaviour, with a focus on replicating the dynamics of ant colonies. By leveraging real-world data, the environment simulates a target ant trail that a controllable agent must learn to replicate, using sensory data observed by the target ant. This work aims to contribute to the neuroevolution of models for collective behaviour, focusing on evolving neural architectures that encode domain-specific behaviours in the network topology. By evolving models that can be modified and studied in a controlled environment, we can uncover the necessary conditions required for collective behaviours to emerge. We hope this environment will be useful to those studying the role of interactions in emergent behaviour within collective systems.

We conducted a comprehensive tracking study of 64 unmarked ants within the same arena to examine the dynamics of individual behaviors within a collective, aiming to understand the underlying mechanisms that drive the colony's collective behaviors. Specifically, we analyzed the movement patterns of the ants to identify the “algorithm” governing their actions. One such approach we employed is the ϵ -machine method, pioneered by Crutchfield and colleagues, which predicts motion using a stochastic finite state machine. The results of our study revealed that individual ants exhibited either deterministic or stochastic behaviors, contingent upon their roles within the colony. Ants contributing to cluster formation displayed deterministic behaviors, whereas those exploring outside of the cluster were more likely to demonstrate stochastic behaviors.