Tiny robots display lifelike teamwork skills.
A group of European researchers is looking to the insect world for ideas on how to innovate the field of robotics. A team of engineers and technicians has invented a batch of 35 tiny, 12 cm robots that work in concert to accomplish tasks that other robots would be incapable of performing. They’ve named the creations swarm-bots.
“Our machines demonstrated remarkable cooperative behavior,” says project coordinator Marco Dorigo of the Université Libre de Bruxelles. “They completed tasks, self-assembled, and coordinated their movement. . . . In our most complex experiment, we placed 20 robots in a big room to retrieve one object to ‘nest.’ This involved building a chain of eight robots placed 30 cm apart and visible to one another. The other robots followed the chain to find and retrieve the object, all in just 10 minutes.”
Dorigo tells THE FUTURIST that these are significant steps forward in this emerging subfield of robotics. “We were the first to be able to demonstrate on real robots autonomous self-assembly of a group of three or more (we ran experiments with up to 16 robots). Also, we were the first to demonstrate that cooperation [in] a group of autonomous robots can allow them to collectively search for an object, find it, self-assemble to it and retrieve it, to a target location. This is far more complex than any other work in swarm robotics we are aware of. Last, we have demonstrated ‘functional’ selfassembly. That is, self-assembly triggered by some environmental contingency. For example, the robots decide to self-assemble because they need to join into a bigger structure to be able to pass over a big hole in the terrain.”
The team’s swarm-bots could one day survey hostile terrains, such as the seafloor or the surface of other planets (NASA has already taken an interest). Dorigo projects that first responders may one day look toward swarm-bots for help in disaster situations such as earthquakes or bombings where people might be trapped under debris and unreachable by other means.
“The type of application will also depend on how far we will be able to go with miniaturization and how cheap these robots will be. An open question for the moment,” he says.
The team is hoping to experiment with materials for the bots besides metal and plastic; they also hope to improve the devices’ memory and autonomy. Most importantly, Dorigo hopes to endow each of the robots with a self-sufficient source of power.
“I think that energy is the big open question,” he says. “We will not have any useful swarm if they cannot reach some kind of energetic autonomy or at least greatly increase their energetic autonomy, which is at the moment one of the main limiting factors. For the rest (that is, concerning the other technologies of which these robots are made, software and hardware), I think that there will be no breakthrough, but a steady improvement on what these small robots can do when they join their forces. To get there, we need to integrate in new, clever ways the many technologies, both hardware and software, that are already available.”
The idea of tiny robots operating in concert conjures scenarios of little droid conspiracies against the human race. Fortunately, Dorigo doesn’t foresee a situation where the swarm-bots gang up on the rest of us. “At the current stage I do not see any danger,” he says. “Our robots are still too fragile, stupid, and far from autonomous. If, in the far future, we identify possible dangers, we will have time to take corrective actions.”
Originally published in THE FUTURIST, March-April 2007.