Robots and AI Are on the March — Literally

Robots and AI Are on the March — Literally

Artificial intelligence and robotics have long been subjects for those pursuing a computer science degree. Software would learn and make smart choices. Robots of all shapes and sizes would do specified tasks.

But what happens when you put these two concepts together? On the simpler side, you wind up with the Harvard kilobots. Harvard School of Engineering and Applied Sciences’ computer science professor Radhika Nagpal developed a set of 1,024 — 1K, that is — small robots that stand on thin legs and vibrate to move themselves about. The swarm of kilobots communicates using infrared laser beams and replicate aspects of an organism, as a fellow researcher was quoted in a Harvard release:

“Biological collectives involve enormous numbers of cooperating entities—whether you think of cells or insects or animals—that together accomplish a single task that is a magnitude beyond the scale of any individual,” says lead author Michael Rubenstein, a research associate at Harvard SEAS and the Wyss Institute.

Without central control, the kilobots can arrange themselves to roughly match shapes communicated to them. So, ask for a letter of the alphabet and they will collaborate and coordinate with one another, shifting around until they have replicated the pattern, as the video below shows.

The recreations may not be perfect, but they are astonishingly good. There are some organisms that cooperate, like the way a troop of ants can use their bodies to form bridges and rafts so other ants can cross an area. In the past, Nagpal’s group developed small robots that emulated termites to undertake construction tasks.

Get past the novelty and you start to see the potential usefulness, and the roll that computer scientists will play. For example, why not have robot colonies collaborate to build small devices? Instead of investing in specialized tooling, the bots could become a flexible crew, able to take on different automated tasks with only a change of instructions.

Artificial intelligence will eventually be applied on a wider basis, directing robots to perform all sorts of tasks that need flexibility and adaptability rather than a rigorously fixed set of instructions. It’s a radically different approach to automation than specially-built devices and software created to perform particular movements over rigidly-set constraints. And yet, it’s likely to represent one aspect of mankind’s future. Computer scientists who want to be involved in such areas will need deep learning and as high a degree of adaptability of thought and the kilobots have of motion.