Robots: Learning

Posted 20 Nov 2009 at 10:45 UTC by mwaibel

The latest episode of the Robots podcast interviews two experts in robot learning. Andrea Thomaz from Georgia Tech studies how humans teach and how humanoids learn with the goal to improve human-robot interaction. Among other things, Thomaz has designed her own humanoid robot Simon shown in the video above (have a look at this non-youtube video to see Simon's expressive ears in action). Our second guest Sethu Vijayakumar from the University of Edinburgh has done an astonishing amount of work spanning the fields of statistical machine learning, human motor control, Bayesian inference techniques and computational neuroscience. In the interview Vijayakumar speaks about the similarities and differences between human and machine learning, and explains how robots can learn about their environment through physical interactions. Read on and tune in!

Read more... (0 replies)

Consciousness May Not Require Big Brain

Posted 19 Nov 2009 at 19:28 UTC by steve

In an appropriate follow up to yesterday's article on IBM's large scale cortical simulation, comes news from brain researchers at Queen Mary, University of London and Cambridge University. Lars Chittka and other researchers, who have been studying both computer simulations and animal brains, now believe that brain size is not as significant as previously believed with respect to general intelligence and consciousness. "No one would seriously suggest that a bigger computer is automatically a better computer", say the researchers.

We know that body size is the single best way to predict an animal's brain size. However, contrary to popular belief, we can't say that brain size predicts their capacity for intelligent behaviour. In bigger brains we often don't find more complexity, just an endless repetition of the same neural circuits over and over.

The larger amount of IO needed in larger animals for things such as more muscles and higher resolution sensors may be the main driving factor for needing larger brains. The researchers say it's possible significant intelligence and consciousness could be present even in insect brains. This fits in with other recent research showing that, despite their tiny brain size, bees could match the best machine-based face recognition algorithms. For more, see the full text the researchers report, Are Bigger Brains Better (PDF format), or listen to the BBC Radio interview of Professor Lars Chittka on BBC Radio 4. CC-licensed image of honeybee by flickr user Agent Opie

Read more... (0 replies)

Real-Time Cat-Level Cortical Simulation Achieved

Posted 18 Nov 2009 at 17:28 UTC by steve

IBM announced today a new advance in the simulation of biological brains on computer hardware. As part of the DARPA SyNAPSE project, the Cognitive Computer Team at the IBM Almaden Research Center has achieved a large-scale simulation based on a new algorithm, called BlueMatter, that synthesizes neurological data.

BlueMatter, a new algorithm created in collaboration with Stanford University, exploits the Blue Gene supercomputing architecture in order to noninvasively measure and map the connections between all cortical and sub-cortical locations within the human brain using magnetic resonance diffusion weighted imaging.

The result achieved was real-time simulation of a cat-level cortex containing 1 billion spiking neurons and 10 trilling synapses. The simulation required 147,456 CPUs and 144 terabytes of memory. For more on this project, see IBM Research Center Cognitive Computing Manager Dharmendra S Modha's Cognitive Computing Blog.

Read more... (1 reply)

Random Robot Roundup

Posted 16 Nov 2009 at 20:27 UTC by steve

The Shadow Robot Company let us know about a great funding opportunity in the EU called ECHORD. Get your robot project proposal in by 1 Dec for a chance at 12 months funding for a staff plus hardware. The PlasticPals blog has some photos and information on the Russian AR-600 humanoid robot. If you're in need of some Java free software for most commonly available servo controllers, check the newest version of ServoMaster. Joanne Pransky, the Robot Psychiatrist, wrote to tell us about a new "low-cost" industrial robot arm from ST Robotics, the R12-500 Firefly. If you buy one, maybe you can combine it with MobileRobots, Inc. new MT400 "build-a-bot" autonomous robot base. Amber Johnson writes, "We just posted an article: 10 Unbelievable Robots Transforming Medicine". Rog-a-matic sent a link to a giant beetle robot built in Japan. And the Swirling Brain spotted stories on tiny sea-going robots, an update on the Robopocalypse film by the author of How to Survive a Robot Uprising, the worlds first Arabic-speaking interactive robot, and EVOLTA, a running, jumping, toy robot. Know any other robot news, gossip, or amazing facts we should report? Send 'em our way please. And don't forget to follow us on twitter.

Read more... (0 replies)

Roomba Pac-Man

Posted 11 Nov 2009 at 20:07 UTC by steve

The Research and Engineering Center for Unmanned Vehicles (RECUV) at the University of Colorado at Boulder has been developing software that helps robots form ad-hoc networks and distribute cooperative control of their operations. Some of the individuals at RECUV decided to create a cool demo on their own time to show off what their software can do. They've implemented a real-life version of Pac-Man using Roombas. They are quick to point out that despite the fact that the Blinky, Inky, Clyde, and Pinky Roombas seem determined to kill the Pac-Man Roomba, all the robots are actually quite safe. This is because, they say, all are "instilled with the Three Laws of Roombotics". And we'll note that the software is safe to download and use too because it protects your rights to examine, use, modify and distribute under the GNU GPL. There's one more video after the break with details on the waypoint navigation software used in the demo.

Read more... (0 replies)

Robust Optical Flow Model Derived From Biology

Posted 11 Nov 2009 at 01:09 UTC by steve

Optical Flow image processing algorithms have been around for a while but have yet to match the motion detection capabilities of insects. In particular current models tend to fail in real-world situations which include high-dynamic range as well as rapidly changing contrast, velocity, and acceleration. Russel Brinkworth and David O'Carroll of the University of Adelaide in Australia explain:

The extraction of accurate self-motion information from the visual world is a difficult problem that has been solved very efficiently by biological organisms utilizing non-linear processing. Previous bio-inspired models for motion detection based on a correlation mechanism have been dogged by issues that arise from their sensitivity to undesired properties of the image, such as contrast, which vary widely between images.

Brinkworth and O'Carroll have come up with a new optical flow model, described in their paper, Robust Models for Optic Flow Coding in Natural Scenes Inspired by Insect Biology (PDF format). Their new system is based on the actual neural processing pathways of the fly which could prove to be a very robust velocity estimator and accurate sensor for self-motion in robots. CC-licensed image of female Tabanus Horse Fly by flickr user Thomas Shahan.

Read more... (0 replies)

LaserMotive Wins Level 1 Space Elevator Prize

Posted 9 Nov 2009 at 19:17 UTC by steve

LaserMotive won the $900,000 Level 1 prize at this year's Space Elevator competition, doing a 1 km climb at an average speed of 4 m/s. The Kansas City Space Pirates came close but didn't couldn't beat LaserMotive's score. USST showed promise but ran into technical problems this year. Each team's space elevator is a robot climber powered by a high-powered laser beam directed at the climber from the ground. A NASA press release sums up this year's competition. For more details on the contest and this year's teams, see the Space Elevator Games website. There are additional photos of the winners in at Space Elevator Blog. For more vidoe see the Spaceward Foundation YouTube channel. LaserMotive also has more photos and info in their blog.

Read more... (0 replies)

Robots: Stroke Rehabilitation

Posted 6 Nov 2009 at 09:16 UTC by mwaibel

The latest episode of the Robots podcast looks at using robots to help stroke victims. Unlike other stroke rehabilitation projects like the MIT robotic brace which focus on arm motion, Ludovic Dovat is concerned with the considerably more difficult task of providing rehabilitation for the hand. Working as part of a multi-national team at the National University of Singapore, Dovat reports on three systems that allowed stroke victims to re-learn complex tasks such as handwriting, manipulation and coordination of their fingers. In a second interview physiotherapist David Brown describes how his Chicago-based company Kinea Design tries to aid stroke victims recover their walking skills. Read on and tune in!

Read more... (0 replies)

The Bad Robot Dance

Posted 29 Oct 2009 at 18:16 UTC by steve

This mysterious short video of a dancing robot turned up on YouTube recently. As it turns out, it's a teaser for an upcoming short film shot in LA. It's from Ruairi Robinson who also produced the short film The Silent City. About all that's know is a quote found in a production designer want ad:

5 day SAG Ultra-low Budget production. Written and directed by Academy Award nominated director, Ruairi Robinson, 'Bad Robot' is financed by the Irish Film Board and produced by Logolite Entertainment, Alturas Films and Image Now Films. Project is scheduled to begin shooting August 26.

Beyond those hints, we'll just have to wait and see what this short film is all about. via The Swirling Brain, via io9, via QuietEarth.

Read more... (0 replies)

Random Robot Roundup

Posted 28 Oct 2009 at 19:45 UTC by steve

Lindsey Carlin sent a link to an Innovation First news release describing their latest products, a line of tiny autonomous robots called HEXBUGs that look like they were inspired by BEAM robots and bristlebots. Roschler has written an update on the rise and fall (and resurrection) of the Pleo. Noriko Kageki of GetRobo has posted some new info on Stanford's autonomous car. David Wilson sent a YouTube link showing the roboseed, the MAV monocopter based on a Samara or maple seed. The Swirling Brain sent us a full brain dump of new robot links including a GadgetReview post on Amio, a PinkTentacle post on a robot that simulates the symptoms of H1N1 flu, a PopSci article on a robot skier, a story on Georgia Tech's robot cat that attempts to display emotional states in a way humans recognize, and finally a CNET story on SnackBot, a CMU robot designed to navigate a crowd dispensing snacks. Know any other robot news, gossip, or amazing facts we should report? Send 'em our way please. And don't forget to follow us on twitter.

Read more... (0 replies)

Robots: Slithering Creatures

Posted 27 Oct 2009 at 10:27 UTC by mwaibel

Snakes are exceptionally agile, which allows them to slide, squeeze, climb or slither into places that people and most other types of robots can't reach. This makes snake robots very attractive in spite of the hair-raising complexity of the required mechanics and control algorithms. The latest episode of the Robots podcast interviews two leading researchers working on snake robots: Howie Choset is an associate professor and the director of the BioRobotics Lab at Carnegie Mellon University where his research in path planning, motion planning and estimation have been used to control a range of snake-inspired robots. Among other things, his group has built robots capable of more than 10 different snake gaits, from sidewinding to pipe rolling. Check out the video above for a first glimpse of his work. This episode's second guest is Erik Kyrkjebø, Senior Researcher at the Applied Cybernetic department at Scandinavia's largest independent research organization, the SINTEF in Norway. Kyrkjebo talks about the PiKo pipe inspection robot and the fire fighting robot Anna Konda which is propelled by water. For more information read on on the Robots webpage or directly tune in to this episode.

Read more... (0 replies)

Boston Dynamics is Back with a Biped

Posted 26 Oct 2009 at 20:12 UTC by steve

Boston Dynamics, best known for the quadruped robots BigDog and Little Dog, have now developed a dynamically balancing bipedal robot that uses a toe-to-heel walking method very similar to a human. The robot, called PETMAN, can be seen in the YouTube video above going through tests on a treadmill. The eventual goal is an anthropomorphic robot that will be used in testing chemical protection clothing for the US Army. In addition to walking, the robot will be able to crawl like a human as well simulate human physiology including body temperature, humidity, and sweating to test the protective clothing under life-like conditions. Props to Robodave for the tip.

Read more... (0 replies)

Emotion Recognition System for Autistic Children

Posted 24 Oct 2009 at 22:45 UTC by steve

Children with the neurodevelopmental disorder known as Autism often have difficulty with social interaction, in part due to an impaired ability to intuit the emotional state of other people. Researchers at Nanyang Technological University have been working on a project that may help Autistic children by augmenting their emotional recognition artificially. Dr. Cho Siu Yeung with PhD students Nguwi Yok Yen and Teoh Teik Toe of the NTU Center for Computational Intelligence describe their system in a newly published paper, "Intelligent System to Help Autistic Children Recognize Emotions". The paper is in a pay per view commercial journal rather than an open scientific publication but the abstract gives a hint at what their working on:

The system is capable of locating the face region using derivative-based filtering and recognizing facial expressions using [a] boosting classifier. The portable device is being developed to help autistic children understand the emotions of surrounding people. The portable facial expression recognizer locates the edge of the human face through Gaussian derivatives, Laplacian derivatives and filters out non-face images using Adaboost. Secondly, the feature locator finds crucial fiducial points for subsequent feature extraction and selection processing. Finally, the meaningful features are classified into the corresponding classes.

The Autism assistance device is a case-study for a more complex Neuro-Cognitive Computational Framework that has been under development since 2005 and combines visual cues such as facial expression, gestures, and posture with "understanding of emotional tags in semantic memory learning structures" and auditory cues. This more general framework could be used for other applications in machine intelligence and robotics.

Read more... (0 replies)

Ruggedized Atom-based Vehicle Computer

Posted 23 Oct 2009 at 16:28 UTC by steve

Rave Computer has announced a new ruggedized MIL-SPEC-810F in-vehicle computer that looks good for use in autonomous vehicles and robots too. The RCV 6100 is based on a 1.6 GHz Atom N270 with up to 2GB of RAM. The fanless design can operate in temperatures ranging from -30 C to 60 C (-22 F to 140 F) and run on 6 - 36 VDC. IO includes a PCI-104 expansion connector, 2 mini PCI express slots, compact Flash and SIM slots, GB Ethernet, 802.11b/g/n, GSM or WCDMA, Bluetooth, GPS, 3 USB ports, 2 RS-232, 1 RS-485, 1 LVDS, 1 VGA, 1 DVI-D, audio in and out, 4 GPIO ports (4 in and 4 out). And, of course, you can run GNU/Linux on it. For even more info, see the RCV 6100 spec sheet (PDF format). How much will all this cost? Rave isn't saying but our guess is you should apply the old rule of thumb: if you have to ask, you can't afford it. Via LinuxDevices.com

Read more... (0 replies)