Supporting material for the article:

Towards group transport by swarms of robots

November 2008

Table of Contents Abstract Video recordings

1. Abstract:

We examine the ability of a swarm robotic system to transport cooperatively objects of different shapes and sizes. We simulate a group of autonomous mobile robots that can physically connect to each other and to the transported object. Controllers - artificial neural networks - are synthesised by an evolutionary algorithm. They are trained to let the robots self-assemble, that is, organise into collective physical structures, and transport the object towards a target location. We quantify the performance and the behaviour of the group. We show that the group can cope fairly well with objects of different geometries as well as with sudden changes in the target location. Moreover, we show that larger groups, which are made of up to 16 robots, make possible the transport of heavier objects. Finally, we discuss the limitations of the system in terms of task complexity, scalability, and fault tolerance, and point out potential directions for future research.

Keywords: artificial neural networks; bio-inspired computation; collective structures; cooperation; cooperative transport; evolutionary robotics; foraging; group transport; modular robots; object manipulation; reconfigurable robots; self-assembling robots; self-assembly; swarm intelligence; swarm robotics.

2. Video recordings:

• 16 robots transporting a heavy object while surrounding it: MPEG Movie Each robot is controlled by a simple recurrent neural network. The network enables the robot to decide autonomously where to look at, where to move, and where to grip.
• 16 robots transporting a heavy object while linking to it: MPEG Movie Each robot is controlled by a simple recurrent neural network. The network enables the robot to decide autonomously where to look at, where to move, and where to grip.