hjalli.com - Hjálmar Gíslason » Robotics

“Female” robots

Hjalmar Gislason — Sun, 14 Sep 2003 16:51:00 +0000

Australian robotic software firm Kadence Photonics has made the world’s first “female” robots. Peter Hill, the founder of Kadence says that after seeing a TV program about the female brain he decided the day after that his line of robots should be based on some of its merits. The female robots differ from other robots in their ability to co-operate and multitask and work toward their goals in a flexible way, whereas in traditional robots on manufacturing lines work on single tasks in sequential order. Or as Dr Hill puts it so elegantly: “If a man does the housework, he’ll load the washing machine then stand there and watch it. A woman will go off and do something else.”

This difference between the male and female brain is proven and was for example portrayed in the book Why Men Don’t Iron, that was also made into a TV series (I wonder if that was the one that Hill saw). The “female” analogy here is still of course just a great way to state what is special about Kadence’s robots and a pretty certain way to grab the media’s attention (not that there is anything wrong with that).

I see this as a step towards where I think robotics and a lot of other technology is destined to go – systems of simple objects (girls, don’t get me wrong!), that interact to work together in a flexible way towards a single goal. Once again an ant colony makes a great analogy, where the simple individual ants work together to make up the complex thing that is the anthill with its entire infrastructure. Methods like these can be flexible, effective, redundant and probably inexpensive as each “module” can be relatively simple and mass produced.

Methods like these could and should work for a lot of manufacturing tasks, but make note however that the more complex problems each “module” can solve and the better understanding it has of the local and overall goals of the system, the more complex tasks it will be able to take on as a whole. Compare construction workers to ants and you quickly see what I mean.

Links:
Enter the female robots (The Sydney Morning Herald)
Female Robots Are Coming (Roland Piquepaille)
T-X – the coolest female robot

Robosnail and mimcking of animal locomotion

Hjalmar Gislason — Sun, 07 Sep 2003 15:35:25 +0000

MIT’s Fluid Dynamics Lab seems to be on a biomimicry roll. I recently wrote about the robostrider. Now it is the robosnail that mimics the locomotion of snails. Their 3-link swimmer is actually very interesting as well, but has yet to receive its 15 minutes of media-fame like its siblings. This kind of research obviously helps biologists understand animals’ locomotion and is probably very interesting from the microfluid perspective (I wouldn’t know), but making use of animal locomotion methods in robots and vehicles definetly opens a range of new possibilities.

More examples of mimicking of animal locomotion can be found at MIT’s Leg lab of which I’ve been a fan for a long time. A well designed legged robot or vehicle could make its way to places that wheeled vehicles have no change of getting to, just ask anybody who is into horses. In a similar way snail locomotion could allow small robots to climb vertical or even overhanging surfaces to reach difficult places. Such use probably has limits in size and weight of the robot. And now that people have been able to reproduce how geckos’ manage to stick to almost any surface, it can’t be too long until we see robots that can climb walls with ease (think: window cleaning and construction aides).

In order to make full use of animal locomotion methods one has to be able to grasp its full dynamics and that has proven to be no easy task. When for example an animal runs, the gait is not simply set to a certain rythm, but much rather to be able to instantly respond to any hole, bump or hurdle it may encounter (otherwise you’d twist your ancle and/or stumble every time you stepped on a stone in the path). The point could be made that the locomotion alone is only half the achievement without the intelligence (artificial or human) to select the routes and deal with unexpected hurdles.

More links:
MIT�s RoboSnails model novel movements (MIT News)
MIT�s RoboSnails (Roland Piquepaille’s Technology Trends)
MIT scientists looking to unlock the secrets of a snail’s pace
(SFGate.com)

Water striders mimicked

Hjalmar Gislason — Wed, 03 Sep 2003 23:23:55 +0000

Researchers at MIT have solved the mystery of how the insects known as water striders move across water’s surface. While it is no mystery how they manage to stay afloat (many of us have surely done experiments with water’s surface tension), previous theories on how they manage to move on the surface have proven wrong. The striders actually “row” on the water surface and the researchers have made a very simple robot that mimickes this behavior.

Graduate students David Hu (mathematics) and Brian Chan (mechanical engineering) at MIT solved the riddle by carefully studying water striders’ motion and built the robot out of a 7-up can, stainless steel and an elastic band. It is not quite as elegant as its biological role model that can move at a speed of up to 1.50 m/sec, but nevertheless proves the theory in practice.

I can’t quickly think of any other machine that makes use of water’s surface tension (comments welcomed), but this could lead the way. Maybe successive water strider robots will be used to clean leaves and other dirt from swimming pools or simply make fun toys.

Links:
Article in MIT News (don’t miss the beautiful images and the videos)
Infographic from Nature magazine
The Nature article
Discussion on Slashdot
Purcell’s swimmer (another project in the same lab at MIT)

Snake-like robot uses genetic algorithms to make up for injury

Hjalmar Gislason — Mon, 01 Sep 2003 23:30:00 +0000

New Scientist recently reported on a snake-like robot that uses genetic algorithms to learn how to move. If a part of the robot is damaged, the GA process will find a new method to move about, if the damage is not too severe for it to be able to move at all.

This seems like just the task genetic algorithms will be good for solving.

Slashdot had an interesting discussion on the matter.

“I wonder if they could extend this to create a worm type robot that if it got cut in two it could still carry on. Say a brain in each segment”.

Very cool. If the whole mechanism was built in to every segment in the snake, parts as small as a couple of segments, could still go on and try to complete the tasks at hand. The snake might even use this deliberately, and split itself up from time to time, e.g. to cover more ground in a search operation or squeeze itself into tight corners.

“While it is certainly natural for all living creatures to learn to cope with an injury, I think this is far more important for robots, since they don’t have a survival instinct. Humans will go to great lengths to avoid getting injured (well, most sane ones). However, robots will just do what they are told. While they may be told to avoid any hazards, they just don’t have that instinct that says I *really* don’t want to get my limb chopped off today.”

Good point and very philosophical. What exactly is a “survival instinct”, and how do you build one into a robot? Could a robot “think of” damaging itself in order to escape from a tight situation, like some animals do when in a trap?

Links:

Another snake robot project at PARC: PolyBot
Brilliant similar stuff by a guy named Mark Tilden.
New York steam pipe fixing robot WISOR.