Technology
0 view | +0 today
Follow
Your new post is loading...
Your new post is loading...
Rescooped by stormey beecoolnow from Heron
Scoop.it!

DARPA Designs New Robots After Cheetah And Pack Mule (Videos) - Inventorspot

DARPA Designs New Robots After Cheetah And Pack Mule (Videos) - Inventorspot | Technology | Scoop.it
DARPA Designs New Robots After Cheetah And Pack Mule (Videos)InventorspotBut the Cheetah Bot, whose namesake runs at about 60 mph, will be out there in the field on tougher terrains than a racing track as soon as all of its algorithms are perfected.

Via cafonso
more...
No comment yet.
Rescooped by stormey beecoolnow from Heron
Scoop.it!

Here come the inflate-a-bots: iRobot's AIR blow-up bot prototypes - Ars Technica

Here come the inflate-a-bots: iRobot's AIR blow-up bot prototypes - Ars Technica | Technology | Scoop.it
Here come the inflate-a-bots: iRobot's AIR blow-up bot prototypesArs TechnicaA DARPA-funded research project at Massachusetts-based iRobot has developed a series of prototype robots with inflatable parts.

Via cafonso
more...
No comment yet.
Rescooped by stormey beecoolnow from Amazing Science
Scoop.it!

Future of Machine Learning: Automated tools aim to make it easier to teach a computer than to program it

Future of Machine Learning: Automated tools aim to make it easier to teach a computer than to program it | Technology | Scoop.it

Machine learning – the ability of computers to understand data, manage results, and infer insights from uncertain information – is the force behind many recent revolutions in computing. Email spam filters, smartphone personal assistants and self-driving vehicles are all based on research advances in machine learning. Unfortunately, even as the demand for these capabilities is accelerating, every new application requires a Herculean effort.  Even a team of specially-trained machine learning experts makes only painfully slow progress due to the lack of tools to build these systems.

 

The Probabilistic Programming for Advanced Machine Learning (PPAML) program was launched to address this challenge. Probabilistic programming is a new programming paradigm for managing uncertain information. By incorporating it into machine learning, PPAML seeks to greatly increase the number of people who can successfully build machine learning applications and make machine learning experts radically more effective. Moreover, the program seeks to create more economical, robust and powerful applications that need less data to produce more accurate results – features inconceivable with today’s technology.

 

“We want to do for machine learning what the advent of high-level program languages 50 years ago did for the software development community as a whole,” said Kathleen Fisher, DARPA program manager.

 

“Our goal is that future machine learning projects won’t require people to know everything about both the domain of interest and machine learning to build useful machine learning applications. Through new probabilistic programming languages specifically tailored to probabilistic inference, we hope to decisively reduce the current barriers to machine learning and foster a boom in innovation, productivity and effectiveness.”

 

To familiarize potential participants with the technical objectives of PPAML, DARPA will host a Proposers' Day on Wednesday, April 10, 2013. For details, visit:http://www.solers.com/BAAinfo-reg/ppaml. Registration closes on Friday, April 5, 2013 at 5 p.m. ET.  

 

The PPAML program is scheduled to run 46 months, with three phases of activity from 2013 to 2017. Fisher believes a successful solution will involve contributions from many areas, including statistics and probabilistic modeling, approximation algorithms, machine learning, programming languages, program analysis, compilers, high-performance software, and parallel and distributed computing.

 

The DARPA Special Notice document describing the specific capabilities sought is available at http://go.usa.gov/2PhW.

 

Associated images posted on www.darpa.mil may be reused according to the terms of the DARPA User Agreement, available here: http://go.usa.gov/nYr.


Via Dr. Stefan Gruenwald
more...
No comment yet.
Rescooped by stormey beecoolnow from anonymous activist
Scoop.it!

DARPA Drones & Robots - Welcome to The New World Order!!!

[video] DARPA Drones & Robots - Welcome to The New World Order!!! http://t.co/nGdF7spEbv #nwo

Via Iam Legion
more...
No comment yet.
Rescooped by stormey beecoolnow from Amazing Science
Scoop.it!

DARPA shows off 1.8-gigapixel surveillance drone that can spot a terrorist from 20,000 feet above

DARPA shows off 1.8-gigapixel surveillance drone that can spot a terrorist from 20,000 feet above | Technology | Scoop.it

DARPA and the US Army have taken the wraps off ARGUS-IS, a 1.8-gigapixel video surveillance platform that can resolve details as small as six inches from an altitude of 20,000 feet (6km). ARGUS is by far the highest-resolution surveillance platform in the world, and probably the highest-resolution camera in the world, period.

 

ARGUS, which would be attached to some kind of unmanned UAV (such as the Predator) and flown at an altitude of around 20,000 feet, can observe an area of 25 square kilometers (10sqmi) at any one time. If ARGUS was hovering over New York City, it could observe half of Manhattan. Two ARGUS-equipped drones, and the US could keep an eye on the entirety of Manhattan, 24/7.

 

It is the definition of “observe” in this case that will blow your mind, though. With an imaging unit that totals 1.8 billion pixels, ARGUS captures video (12 fps) that is detailed enough to pick out birds flying through the sky, or a lost toddler wandering around. These 1.8 gigapixels are provided via 368 smaller sensors, which DARPA/BAE says are just 5-megapixel smartphone camera sensors. These 368 sensors are focused on the ground via four image-stabilized telescopic lenses.

 

ARGUS’s insane resolution is only half of the story, though. It isn’t all that hard to strap a bunch of sensors together, after all. The hard bit, according to the Lawrence Livermore National Laboratory (LLNL), is the processing of all that image data. 1.8 billion pixels, at 12 fps, generates on the order of 600 gigabits per second. This equates to around 6 petabytes — or 6,000 terabytes — of video data per day. From what we can gather, some of the processing is done within ARGUS (or the drone that carries it), but most of the processing is done on the ground, in near-real-time, using a beefy supercomputer. We’re not entirely sure how such massive amounts of data are transmitted wirelessly, unless DARPA is waiting for its 100Gbps wireless tech to come to fruition.

 

The software, called Persistics after the concept of persistent ISR — intelligence, surveillance, and reconnaissance — is tasked with identifying objects on the ground, and then tracking them indefinitely. As you can see in the video, Persistics draws a colored box around humans, cars, and other objects of interest. These objects are then tracked by the software — and as you can imagine, tracking thousands of moving objects across a 10-square-mile zone is a fairly intensive task. The end user can view up to 65 tracking windows at one time.


Via Dr. Stefan Gruenwald
more...
Mercor's curator insight, January 31, 2013 8:53 AM

Rescooped by Christopher Baggett from Amazing Science onto You Can't Make This Stuff Up

Rescooped by stormey beecoolnow from Tracking the Future
Scoop.it!

Say hello to ATLAS, one of the most advanced humanoid robots ever built!

ATLAS was developed for DARPA by Boston Dynamics. Software-focused teams from Tracks B and C of the DARPA Robotics Challenge will use the robot to compete in the first physical competition of the Challenge in December 2013 at the Homestead-Miami Speedway. 
The DARPA Robotics Challenge seeks to advance the technology necessary to create robots capable of assisting humans in disaster response. 


Via Szabolcs Kósa
more...
No comment yet.