Evolving and Controlling Perimeter, Rendezvous, and Foraging Behaviors in a Computation-Free Robot Swarm

Abstract

Designing and controlling the collective behavior of a swarm often requires complex range, bearing sensors, and peerto-peer communication strategies. Recent work studying swarm of robots that have no computational power has shown that complex behaviors such as aggregation and object clustering can be produced from extremely simple control policies and sensing capability. We extend previous work on computation-free swarm behaviors and show that it is possible to evolve simple control policies to form a perimeter around a target, rendezvous to a specific location, and perform foraging. We also demonstrate that simple manipulations of the environment can be used to control, these collective behaviors. The robustness and expressiveness of these behaviors, combined with the simple requirements for control and sensing, demonstrate the feasibility of implementing swarm behaviors at small scales or in extreme environments.

Circle Formation with Computation-Free Robots Shows Emergent Behavioural Structure

“…Recent work shows that complex collective behaviours can be achieved on computation-free robots, such as aggregation and foraging [9]. Among these behaviours, both Gauci et al and Brown et al [5], [2] identified what they called the cyclic pursuit.…”

A Minimalistic Approach to Segregation in Robot Swarms

“…A recent line of research in spatially organizing behaviors focuses on the minimal assumptions a swarm of robots must fulfill in order to perform the task. Johnson and Brown [7] and Brown et al [8] characterized the set of possible behaviors that can be obtained using primordial control strategies based on a simple 'if/then/else' structure, binary sensors, and differential-drive robots. Gauci et al provided the specific conditions for the emergence of aggregation [9] and object clustering [2], while St.-Onge et al [10] studied the emergence of circular formations.…”

Hierarchical Behaviour for Object Shape Recognition Using a Swarm of Robots

“…Some of the benefits from the foraging approach in swarm robotics are the robustness, collective intelligence and emerging behaviours to unexpected events. In this paper, a hierarchical cognitive architecture, composed of evolutionary aggregation, fuzzy logic, and Bayesian perception, is proposed for the control of the collective behaviour of a group of robots with basic computation capabilities [3], for exploration and identification of geometric shapes.…”