io9: robotics engineering

Robotic Rickshaws Wander the Streets of Beijing

Lauren Davis — Sun, 30 Nov 2008 09:00:00 PST

Wu Yulu may have only a primary school education, but his fascination with robots has inspired him to create 26 automata, a veritable robot army. Rather than leading his robots toward global conquest, Wu has designed them for more socially beneficially uses. His robotic rickshaw Wu No. 25 can pull an individual through the streets of China’s Beijing region for up to six hours on a single charge.

Wu was invited to participate in this year’s Microwave International New Media Arts Festival in Hong Kong. Wu has through his own research and experimentation created over two dozen robots based on his curiosity concerning human movement:

I remember once when I was a teen, I was sitting at the doorway of my home, bored. Then someone walked past. So I wondered about the two legs that we humans have, and I wondered if I could build a machine that walked like a human. I didn’t know about robots, but it was then that I got the idea.

In 1986, Wu built his first robot, Old Wu One, which was 20 cm tall and walked very slowly. But, after two decades of experimenting, Wu has created an impressive array of vehicles and automata, including a spider-legged chaise and a large, man-shaped rickshaw that can easily accommodate a passenger and driver. In the video below, Wu shows off several of his mechanical creations and explains his interest in robotics.

Rural Robots by Wu Yulu from microwavefest on Vimeo.

[via Design Boom]

Robots Learn By Doing Improv

Ed Grabianowski — Fri, 25 Apr 2008 08:00:00 PDT

Your household robot won't just clean and make repairs, it will come up with clever, novel solutions to problems by improvising. This hallmark of artificial intelligence is a little closer to reality thanks to a robot named Kurt3D. In a recent test, Kurt3D figured out how to activate a switch and open a door by improvising, using a limited set of instructions. The key to this A.I. breakthrough is a new way of teaching computers about objects by teaching them what something is for rather than simply what it is.

A great deal of A.I. research has focused on teaching computers to identify lists of objects and people. The Multi-sensory Autonomous Cognitive Systems (MACS) project uses a different paradigm - affordance learning. Instead of identifying a specific object as a hammer, an affordance-based system learns the parameters of what makes a hammer useful for hammering. It needs a shaft for leverage, a weight at the end and a flat surface for hammering. Then, if the robot needs to find something with which to hammer, it wouldn't be limited by a narrow visual recognition algorithm for a hammer. It could search for any object suitable for the purpose.

The only given parameters in the Kurt3D test stated that a door switch could be activated by placing a certain weight on a pressure sensitive plate. Kurt3D was able to examine the room, identify an appropriate object, pick it up, place it on the plate, and move through the open door. Photo by: Fraunhofer AIS.

What Can I, Robot, Do With That? [Science Daily]

Does Artificial Intelligence Require Artificial Emotion?

Ed Grabianowski — Wed, 16 Apr 2008 08:00:00 PDT

You might like it when your Tivo predicts what you want to watch, but you probably don't think that makes it intelligent. But what if your Tivo could cheer for a game, or cry with you when you're watching a poignant death scene in Battlestar Galactica? Researchers with the HUMAINE project are studying machine/human emotional interactions, and they're asking this very question. In essence, will people consider their machines intelligent when those machines can express what appear to be feelings?

HUMAINE has gathered psychologists, philosophers, sociologists and computer animation specialists along with database developers and programmers to tackle the issue of machine emotion. Whether HUMAINE's approach results in a better way of recognizing and displaying emotion might be beside the point. The reason they have philosophers on board is to help decide whether or not we should imbue machines with emotions at all.

The logical, emotionless decision-making of sci-fi A.I. is something we both admire (Data, good Terminators) and fear (HAL 9000, bad Terminators). Would it be ethical to give such machines emotions? I'm not sure I want to deal with an ATM that's been having a bad day, much less an armed police robot. In reality, we probably want a lesser degree of machine emotion, a realistic yet fake emotive ability that makes us feel better but doesn't affect the computer's decisions.

The bigger question might be: would an emotionless A.I. be any kind of intelligence at all? I'm not sure it would be possible for a machine to make the intuitive leaps and strokes of genius that we think of as measures of human intelligence in the absence of emotion. Photo by: Warner Bros.

Emotional Machines. [ICT Results]