background preloader

Neuro Evolving Robotic Operatives

Neuro Evolving Robotic Operatives
Neuro-Evolving Robotic Operatives, or NERO for short, is a unique computer game that lets you play with adapting intelligent agents hands-on. Evolve your own robot army by tuning their artificial brains for challenging tasks, then pit them against your friends' teams in online competitions! New features in NERO 2.0 include an interactive game mode called territory capture, as well as a new user interface and more extensive training tools. NERO is a result of an academic research project in artificial intelligence, based on the rtNEAT algorithm. It is also a platform for future research on intelligent agent technology. Currently, we are developing an open source successor to NERO , OpenNERO , a game platform for AI research and education. Related:  Robots and Robotics

Salad-Making Robot Gives Us a Creepy View Of the Future If ever you wanted a dead-eyed robot wielding a knife to be your next personal chef, you're in luck. The Korean Institute of Science and Technology recently unveiled CIROS, a robot that can make you salad. CIROS is designed to help out around the house; it can grasp items from the refrigerator, serve tea, scrub dishes and slice vegetables... very, very slowly. In the below video from Robot World 2012, CIROS slices a cucumber for a salad at a glacial speed, then dumps a load of dressing atop it. Gizmag writes that CIROS' head contains stereoscopic cameras and a 3D IR sensor that helps it recognize objects, like microwaves, sinks, refrigerators and dishwashers. Watch CIROS in action in the below video. Also on HuffPost: RSLSteeper launches third version of its bebionic myoelectric hand Nigel Ackland could be mistaken for a cyborg. He has a high-tech robotic hand that looks like it started life as a Formula 1 car and its movements are alarmingly lifelike. It’s called the “bebonic3” and is the latest version of bebonic series of artificial hands produced by RSLSteeper of Leeds, U.K. The myoelectric hand has been under development for a couple of years now, but the bebonic3 is moving prosthetic limbs from Captain Hook to Luke Skywalker territory. View all Artificial hands have come a long way in recent years, but it turns out the human hand is amazingly complex. While there have been a lot of advances over the past fifty years (as evidenced by the i-LIMB, ProDigits and SmartHand), many artificial hands are little more than powered hooks or pincers that often require a great deal of effort to work. The bebionic3 is designed to not only look human, but also to provide the wearer with a large degree of natural movement. The bebionic3 hand's silicone glove

AMP-Foot 2.0 prosthesis mimics human ankle's spring An amputee tests the AMP-Foot 2.0 on a treadmill Image Gallery (2 images) The majority of protheses available today that replace the lower leg, ankle, and foot are passive devices that store energy in an elastic element (similar to a coiled spring) at the beginning of a step and release during push-off to give you some added boost. While this type of prosthetic is energy efficient, it doesn't replicate the full power we get from our muscles. In order to provide that kind of energy an actuator is required, and these are often heavy and bulky. The latest version, AMP-Foot 2.0, uses an actuator to store energy in springs, which is released when needed. CAD rendering of the AMP-Foot 2.0 The result is a prosthesis that, despite utilizing actuators, weighs just five and a half pounds (2.5 kg), which is roughly the weight of a healthy foot. This isn't the only prosthesis with an actuated ankle, but the low power is key. Source: Vrije Universiteit Brussel About the Author

Mind-controlled robot avatars inch towards reality A researcher minds the robot's balance as it is commanded to pick up a canned drink by an operator (off camera) Image Gallery (5 images) Researchers at the CNRS-AIST Joint Robotics Laboratory (a collaboration between France's Centre National de la Recherche Scientifique and Japan's National Institute of Advanced Industrial Science and Technology) are developing software that allows a person to drive a robot with their thoughts alone. The technology could one day give a paralyzed patient greater autonomy through a robotic agent or avatar. View all The system requires that a patient concentrate their attention on a symbol displayed on a computer screen (such as a flashing arrow). The system does not provide direct fine-grain motor control: the robot is simply performing a preset action such as walking forward, turning right or left, and so on. With training, the user can direct the robot's movements and pick up beverages or other objects in their surroundings. About the Author

DARPA's Robotics Challenge gives birth to new humanoid robots The Defense Advanced Research Projects Agency has published concept artwork depicting the robots that will compete in its ambitious DARPA Robotics Challenge (DRC). The DRC will require robots to drive a car, travel through rubble, open doors, climb ladders, manipulate tools, and more. However, due to the current limitations in artificial intelligence, the robots will be teleoperated by a team of people behind the scenes. View all The winning team will be awarded US$2 million, but that's not all. The software groups don't exactly have it easy, but because they will be given a robot and simulator based on Boston Dynamics' Atlas, they only get funded up to $2.125 million. Here's a quick look at the competitors, with captions provided by DARPA. Carnegie Mellon University – National Robotics Engineering Center CMU-NREC proposed to develop the CHIMP (CMU Highly Intelligent Mobile Platform) robot for executing complex tasks in dangerous, degraded, human-engineered environments. Drexel University

Disney Research robot can juggle, play catch The robot created at Disney Research can play catch with a human partner Image Gallery (4 images) With the aim of providing some physical interaction between entertainment robots and guests at its theme parks, while still maintaining a safe distance between the two, Disney Research has created an animatronic robot that can play catch and juggle balls with a human partner. A Kalman filter algorithm is used to analyze video captured on an external camera system consisting of a Kinect-like ASUS Xtion PRO LIVE to track a colored ball in three-dimensional space and predict its destination and timing. The ball's predicted location is then relayed to the robot, which moves its hand accordingly. The balls need to be thrown in the general vicinity of the robot’s hand, which has been modified with a cup-like shape to boost its catching ability. The developers were also able to speed up the throw/catch cycle to give the robot the ability to juggle three balls with a human partner. About the Author

Bipedal hobby robot walks a tightrope Dr. Guero's Primer-V4 robot works its way across the tightrope Image Gallery (2 images) A Japanese roboticist that goes by the handle Dr. The following video shows Dr. Dr. Source: Dr. About the Author Jason is a freelance writer based in central Canada with a background in computer graphics. Post a CommentRelated Articles Just enter your friends and your email address into the form below For multiple addresses, separate each with a comma Privacy is safe with us because we have a strict privacy policy. Robots are learning to walk and run at Delft University Leo and Phides – two planar biped robots built at the Delft University of Technology – are walking and running, respectively. Leo improved its walking gait through reinforcement learning, which shapes behavior by rewarding success and punishing failure. Phides, the running robot, has achieved an impressive flight phase (the period in a running stride in which both feet are off the ground). Reinforcement learning (RL) normally favors computer simulation, not only because the hundreds (if not thousands) of trials can be run faster, but because real hardware would likely break down from the stress. Leo was built to test how RL would work in the real world. To prevent Leo from falling apart, Schuitema gave it a head start by providing it with a walking gait that it could learn from and improve upon. "If the hardware could be made more knock-proof, a robot like Leo could indeed learn to walk all by itself," Schuitema said. Leo learns to walk Phides running