In the year 2002, Michael Crichton wrote Prey, a science fiction novel in which a corporation manufactures, in a secret facility in the middle of the desert of Nevada, a swarm of nanorobots with autonomy and the capacity to self-organize. It’s a work of fiction that today has acquired premonitory overtones in view of the latest research and advances in the field of robotics. Some teams, such as a group of engineers from Cornell University, have begun to design and program minirobots to behave like a swarm of autonomous insects.
Those who have read Prey will find some of these latest achievements in robotic technology, and the scenario they seem to be leading to, both familiar and disconcerting. Vicente Matellán Olivera, from the Robotics Group of the Department of Mechanical, Computer and Aerospace Engineering at the University of León, has made an analysis for OpenMind that reflects the parallels between Michael Crichton’s fiction and the present-day reality.
“That swarm is a set of micro-robot machines. Where is it getting power?”
“We build the units with a piezo wafer to generate current from photons.”
“So the units are solar-powered.”
As in the book, several robotic groups from around the world are currently working on the development of swarms: groups of small robots that behave and interact like a swarm of insects. A team of engineers from the Harvard Microrobotics Laboratory is working towards this goal and has created the “RoboBee.” Shown to the public in 2013, the first version of this tiny insect-inspired flying robot weighed only 80 milligrams and had a 3 cm wingspan. It was equipped with vision and motion sensors and depended on a central power source. However, at the end of 2017 the next generation was unveiled—the hybrid RoboBee—weighing 175 mg, capable of autonomously flying and navigating underwater, and incorporating electrolytic plates that convert water into oxyhydrogen, which acts as a fuel. The goal of its creators is that in the near future these RoboBees will be able to carry out reconnaissance, search and rescue missions, as well as environmental monitoring.
“The main problem of mechatronic robots is the energy source,” says Matellán. “But we can think of swarms that are responsible for performing simple and repetitive tasks: applying treatments to plants one by one, taking samples over large surface areas or monitoring volcanoes, for example.”
Organize and interact with each other
“The technical definition of emergent behavior was behavior that occurred in a group but was not programmed into any member of the group. Emergent behavior could occur in any population, including a computer population. Or a robot population. Or a nanoswarm.
Individual agents—in this case, actual microrobots—could be programmed to cooperate under certain circumstances. They could be given goals. They could be instructed to pursue their goals. But the result of these interactions could not be programmed. It just emerged, with often surprising outcomes. For the first time a program could produce results that absolutely could not be predicted by the programmer.”
The goal of the Laboratory for Intelligent Systems and Controls (LISC) at Cornell University is to develop programming systems so that robots exhibit behaviour similar to that of living beings, i.e. that they are able to adapt in real time through experience and to organize and interact with each other. LISC focuses on the development and evolution of so-called neuromorphic chips, which instead of processing binary code, process the pulses transmitted by an electrical signal, similar to the signal by which neuronal communication is established in the brain.
Magnus Egerstedt explains how the robotarium got its name and provides a demo of the robots. Credit: Georgia Tech
A robotic laboratory housing available to everyone
“This is just a cluster of microbots. You can make it do what you want. If the programming’s not right, you adjust it. What don’t I understand?”
“We can’t control it.”
“So you have a runaway swarm.”
Finally, there are initiatives such as Robotarium, the robot terrarium from the Georgia Institute of Technology (USA). It is a robotic laboratory housing a swarm of dozens of small robots that can be controlled remotely and that has been made available to everyone. Anyone can submit their project on the university website and, if approved, put it to the test by writing their own programming code to study how the robotic swarm behaves.
In the opinion of Vicente Matellán, this is “the most-educational project. There is nothing better than having someone interacting with a real system to counter the ideas reflected in works of science fiction.”
“What do you think?”
“You’ve got a breakaway robotic nanoswarm. That some idiot made self-powered and self-sustaining.”
“You think we can get it back?”
“From what I’ve seen, there’s not a chance…”
The question inevitably arises: will current research in robotics lead to an outcome similar to that of the novel? (Spoiler alert) In the plot of the book, the uncontrollable swarm of nanorobots turns against its creators, forcing an intervention by the army in order to contain it.
It’s an outcome that, according to Vicente Matellán, “is not real at the present time. The crux of the matter lies in the concept of self-organization. Current neural networks are interactive methods of parameter optimization. To think that consciousness could arise from them is the same as thinking that dumping a load of stones from a truck could make a cathedral appear.”