Evolution of Solitary and Group Transport Behaviors for Autonomous Robots Capable of Self-Assembling

In this paper, we examine the use of artificial evolution to produce solitary transport and group transport behaviors for mobile autonomous robots. Our design and analysis are accomplished using a physics-based 3D simulator. The simulated robots are endowed with connection mechanisms that allow them to attach to (and detach from) each other to form collective physical structures. We show that robots, even though not being aware of each other, can display effective group transport behaviors. Thereby, they can benefit from behaving differently from robots engaged in solitary transport and vice versa. The best behaviors yielded by some of the evolutions let the group of robots organize into physically connected pushing chains. This proves that self-assembly can provide adaptive value to the group, and demonstrates the utility of self-assembling robots in a concrete task. Further analysis reveals the proximate mechanisms that cause the coordinated behavior in the group.