MIT engineers, led by Anette Hosoi, have built a robotic snail to test out mathematical models of how snails move and stay stuck to surfaces.

The artificial gastropod has five movable segments lined up on its underside. One by one, each of these sections moves forwards along a track on the robot's body. After all five have advanced, the body itself slides forward, returning the segments to their original positions.

The team tested out their snail on a tilting platform, covered with slime made from Laponite, a type of clay that forms a clear, sticky gel when mixed with water. Even when the platform was flipped over so that the robot was upside down, it still made steady progress.

Because gastropods have only one foot, it is much easier to build mechanical analogues of snails than of two-footed people or four-footed animals. And although they are slow, snails can crawl around different environments.

The team is now working on a second generation of robosnails that are much faster and more manoeuvrable.

(Via robot gossip < nature)

Gastropods are so cool.

Posted by: Justus on December 14, 2005 5:09 PM

admit it, you wrote this whole post just so you could use the word "robosnail"!

Posted by: baloghblog on December 14, 2005 6:51 PM

Given their their stability and ability to climb walls and odd surfaces, let's say we build a few of these to explore Mars or one of the moons of the gas giants. That would certainly simplify the navigation and sensing algorithms. But how would they manufacture more slime in such a remote location?

Posted by: Pace Arko on December 14, 2005 9:53 PM

pet slugs can eat the mould out of yr shower at night too & they hide away during the day. No need to clean it yourself (with or without chemicals)!

Posted by: Roadside Flower on December 15, 2005 1:32 AM

At least for short-range applications, the slime trail could be useful. Imagine we have a task that requires the rapid exploration of a space difficult of access to human operatives - for example, searching the aftermath of an explosion for survivors.

We release the first lot of robosnails into the target area. They scatter into it using a semi-random process to get started, then basic AI rules to keep moving as far into the space as possible. Then we release the second wave, set to follow a marker(say a weak radioactive tracer) in the slime - that way, they follow the explorers' trails into the space for a set distance and then begin searching around it, or start searching when a sensor (say infrared) is triggered.

Alternatively, the trace could be altered either to mark the spot where the explorer's sensors went off, or where a human remote operator decided to put down a marker, and the searcher could react to that.

Posted by: Alex on December 15, 2005 2:36 AM