The day when robots will do all of our dangerous work for us is quickly approaching, and Google-owned Boston Dynamics‘ walking, self-balancing robot Atlas is a big step (no pun intended) toward that goal. The 6’2″, 330-pound humanoid robot walks on two legs to leave its upper appendages free for lifting, carrying, and manipulating objects. It’s designed to navigate difficult terrain and even to crawl on all fours or climb when its surroundings call for some more flexibility in getting from one point to another.

The Florida Institute for Human & Machine Cognition has been working on Atlas’ walking and balancing algorithms, including a feature that lets the robot recover from tripping or being pushed. Their newest video demonstrating Atlas’ impressive abilities shows the enormous robot balancing on one foot atop a pile of cinder blocks. This may seem like a small feat for a robot – you could stand your old tin robot on a pile of bricks and it would stay there, right? But Atlas is a dynamic moving robot that can self-correct its position to keep it in place as it moves its arms, much in the way a human body does.

In the video, Atlas adopts the “crane” posed demonstrated in the final fight scene of The Karate Kid: one leg bent and raised high and both arms raised over the head. Atlas adjusts its center of gravity as its arms move around, which would be a feat for a lot of humans. Disappointingly, though, it doesn’t actually do the crane kick in the end. The robot is currently in development, with large groups of researchers dedicating their time to various aspects of its movement. They envision Atlas being able to perform actions that are too dangerous for humans, such as ammunition handling, rescue missions, and bomb disposal.

The post Linebacker-Sized Humanoid Robot Can Balance on One Foot first appeared on Gajitz.

If the idea of computer-generated farming reminds you of an early cyberpunk novel, then you’re as skeptical as we are. But one designer has attempted to do just this, using what he calls precision farming.

UK-based designer Benedikt Groß has come up with algorithmic models to plant a variety of crops within close quarters. His goal is to create better harmony between plant and animal life. Planting one solitary crop has historically leaved the field vulnerable to pests, who could potentially wipe out an entire yield. The algorithm model prevents this, as well as uses sustainable practices.

While not a completely new idea (combination farming has been done for centuries to improve soil quality), the employment of technological computer models is pretty original. Groß has refined his vision enough to put a smiley face in the middle of a piece of land he plotted last spring.

The post Digitally ‘Printed’ Fields Can Benefit Farms + Environment first appeared on Gajitz.

Despite the fact that we all know what happens when machines get too smart, some researchers are still trying to teach them new tricks. Zdenek Kalal, a researcher at the University of Surrey, wants to teach machines to “see.” By using a unique object tracking algorithm that he calls Predator, Kalal has enabled computers to learn various faces, track objects regardless of position, scale or orientation changes, and even recognize tracked objects through disappearing and reappearing.

Predator learns as it goes, meaning that every mistake and misstep it makes improves its future performance. Predator could, in theory, be used for some terribly scary things: things like tracking faces on public sidewalks or specific cars on public streets, building a database of recognized faces or any number of other privacy-invading applications. But for now, it is just nicer to believe that it would be used for positive things like helping disabled people use computers.

The post It’s Got Its Eye on You: Smart Camera Learns as it Tracks first appeared on Gajitz.