Squids and octopi use their camouflage abilities to hide from prey and their amorphous bodies to access hard-to-navigate spaces. So, why not a machine?
That insight led researchers at Harvard University to develop an innovative, new robot — part chameleon and part Gumby. Just like cephalopods in the sea, the robot’s soft, rubber-like body can move with surprising grace. Looking for inspiration underwater was a novel idea, according George Whitesides:
“Conventional robotics is a pretty highly developed area, and — if you look at various robots — you find that most are basically built on the body plan of a mammal.
“Our question is: Why do you have to do that? Why not think about organisms that are soft, that might have quite different structures and ways of moving and strategies for camouflage. And the obvious place to look is underwater.”
The soft robot — powered by air pumping through its four limbs — can crawl under or bend around obstacles in its way. It addition, it can change its appearance — choosing either to slyly blend into the background or proudly be set apart from it. Like its mobility system, the unit’s camouflage is a series of tiny channels, but with dye pumped into it.
In the current prototype, the power source and the camouflage dyes are connected by tubes to a base unit. Down the line, the researchers hope to ‘cut the cord’ — and incorporate these systems into the robot body itself — so it’s a truly autonomous entity.
It’s estimated that this type of light and flexible robot will be cheap to manufacture. One possible use may be in search-and-rescue operations. Professor Whitesides points out the benefits:
“For that kind of application, having it be able to advertise itself, for example, in a way that stood out against the dark would be a good thing.”
For a mission like search and rescue, these kind of robots could in principle be throwaway. So if you took a $25,000 robot and sent it in and the building falls down, then that is a real issue. If you send one in which is $100 and the roof falls in, you really don’t care.”
Though real-world applications are a few years away, all in all, it seems to be a very promising project.



















