Monday, August 25, 2008

5 things your car will finally do in 2020

We'll spare you the far-future posturing and flying-car jokes, but the truth is -- for the 200 million automobile owners in America, the future looks bright. In fact, we've already made some pretty impressive headway. The 2008 Mercedes S-Class can change lanes on the highway automatically, and both the Toyota Prius and the Lexus LS-460 can self-park at the push of a button. But that's just a drop in the bucket compared to what's in the works. With eyes wide and mouths agape, we peeked under the curtain at the cars of the future. Here's what we can tell you about what you'll be driving in 2020.

1. Your car will predict the future: Self-parking cars are great and all, but there's a big difference between features of convenience and the kinds of safety technologies on the horizon. We're talking about cars that can see into the future and react on a dime -- whether that means detecting a person crossing the street or swerving to avoid oncoming traffic. Read More

Monday, August 11, 2008

The Invisibility Cloak

Scientists in the US say they are a step closer to developing materials that could render people invisible. Researchers at the University of California in Berkeley have developed a material that can bend light around 3D objects making them "disappear". The materials do not occur naturally but have been created on a nano scale, measured in billionths of a metre. The team says the principles could one day be scaled up to make invisibility cloaks large enough to hide people.
The findings, by scientists led by Xiang Zhang, were published in the journals Nature and Science. The new system works like water flowing around a rock, the researchers said. Because light is not absorbed or reflected by the object, a person only sees the light from behind it - rendering the object invisible. The new material produces has "negative refractive" properties. It has a multi-layered "fishnet" structure which is transparent over a wide range of light wavelengths. The research, funded by the US government, could one day be used in military stealth operations - with tanks made to disappear from the enemies' sight. Read More

Thursday, August 7, 2008

Robots learn to move themselves

Researchers in Leipzig have demonstrated software designed for robots that allows them to "learn" to move through trial and error. The software mimics the interconnected sensing and processing of a brain in a so-called "neural network". Armed with such a network, the simulated creatures start to explore. In video demonstrations, a simulated dog learns to jump over a fence, and a humanoid learns how to get upright, as well as do back flips.Ralf Der at the Max Planck Institute for Mathematics in the Sciences has also applied the software to simulated animals and humans. The only input to the network is the types of motion that the robot can achieve; in the case of a humanoid, there are 15 joints and the angles through which they can move. No information about the robot's environment is given. The network then sends out signals to move in a particular way, and predicts where it should end up, based on that movement. If it encounters an obstacle such as itself, a wall or the floor, the prediction is wrong, and the robot tries different moves, learning about itself and its environment as it does so. "In the beginning, we just drop a robot into a space. But they don't know anything, so they don't do anything," Professor Der said. The neural network eventually picks up on electronic noise, which causes small motions. Read More