Segregation of multiple heterogeneous units in a robotic swarm

Abstract

Several natural systems adopt self-sorting mechanisms based on segregative behaviors. Among these, cell segregation is of particular interest since it plays an important role in the formation of tissues, organs, and living organisms. The Differential Adhesion Hypothesis states that cells naturally segregate because of differences in affinity, which lead similar cells to strongly adhere to each other. By exploring this principle, we propose a controller that can segregate a heterogeneous swarm of robots according to the characteristics of each agent, such that similar robots form homogeneous teams and dissimilar robots are segregated. We apply LaSalle's Invariance Principle to show convergence and perform simulated experiments in order to demonstrate the robustness and effectiveness of the proposed controller. Results show that our approach allows a swarm of multiple heterogeneous robots to segregate in a coherent and smooth fashion, without any inter-agent collisions.