Kube’s paper discusses the model of ants and how the workers cooperate to retrieve large prey. The robotic system demonstrated shows how a decentralized problem-solving by a group of robots and it provides the first formalized model of cooperative transport in ants. In autonomous robotics, swarm-based or collective robotics relies on the same metaphor for the design of distributed control algorithms for swarms of robots. A swarm of simple robots may also be more flexible and fault-tolerant because one or several robots may fail without affecting task completion.

The theory of swarm based systems is interesting because the activity can be regulated using only local perception and indirect communication through the environment as applied to algorithms for coordinating distributed building behavior and foraging tasks by multirobot systems. The part of the system that I do not entirely agree with is that the pathways to solutions are usually not predefined but emergent and solving a problem amounts to finding a trajectory for the system and its environment so that the states of both the system and its environment constitute the solution to the problem.

The other aspect of the research is that for an interesting idea with swarm based development, the idea of pushing boxes is not all that exciting. I was hoping that there would be a more interesting perspective and maybe a more practical example of the utility of this architecture and approach in the real world. The paper does go on to describe how social insects are nature’s proof-by-example of a decentralized multi-agent system whose control is achieved through locally sensed information.

The results of the experiments with physical robots presented in the article add support to the simulation studies which showed that cooperation in some tasks is possible without direct communication. For example, the cooperative transport of ants, if it is self-organized, is not always efficient because it takes ten minutes to get the first object to move. Barring efficiency, we can see that this is a unique approach to swarm based solution to a multi-agent system but is not truly practical from the perspective of pushing boxes.

Reference:

  1. Kube, C. Ronald. “Cooperative Transport By Ants and Robots.” <http://webdocs.cs.ualberta.ca/~kube/papers/KubeBonabeauRAS2000.pdf>