Amazon Picking Challenge aimed at improving warehouse robotics
By David Szondy for Gizmag: One of the biggest events at the recent 2015 IEEE International Conference on Robotics and Automation (ICRA) in Seattle was the first Amazon Picking Challenge, in which 31 teams from around the world competed for US$26,000 in prizes. The challenge set entrants with the real-world task of building a robot that can do the same job as an Amazon stock picker.According to Amazon Chief Technology Officer Peter Wurman, who initiated the challenge, the task of picking items off the shelf may seem simple, but it involves all domains of robotics. The robot has to capable of object and pose recognition. It must be able to plan its grasps, adjust manipulations, plan how to move, and be able to execute tasks while noticing and correcting any errors.
This might suggest that the robots would need to be of a new, specialized design, but for the Picking Challenge, Amazon made no such requirement. According to one participant we talked to, the more important factors were sensors and computer modelling, so ICRA 2015 saw all sorts of robots competing, such as the general purpose Baxter and PR2, industrial arms of various sizes, and even special-built frames that move up, down, left or right to position the arm. Even the manipulators used by the various teams ranged from hooks, to hand-like graspers, and vacuum pickups. Continue reading for competition results:
Paralyzed man can now use his mind to shake hands, drink beer using robotic arm
A man paralyzed by gunshot more than a decade ago can shake hands, drink beer and play "rock, paper, scissors" by controlling a robotic arm with his thoughts, researchers reported.
Two years ago, doctors in California implanted a pair of tiny chips into the brain of Erik Sorto that decoded his thoughts to move the free-standing robotic arm. The 34-year-old has been working with researchers and occupational therapists to practice and fine-tune his movements.
It's the latest attempt at creating mind-controlled prosthetics to help disabled people gain more independence. In the last decade, several people outfitted with brain implants have used their minds to control a computer cursor or steer prosthetic limbs. Full Article:
ABB's Acquisition of Gomtec Expands Collaborative Robotics Portfolio
ABB, a leading power and automation group, announced it acquired Gomtec GmbH to expand its offering in the field of collaborative robots. The parties agreed not to disclose financial terms of the transaction.
Gomtec, based near Munich, Germany, is a privately held company that develops mechatronic systems combining mechanical, electrical, telecommunications, control and computer engineering for customers in diverse industries. It has 25 employees.
Gomtec's technology platform will strengthen ABB's development of a new generation of "safe-by-design" collaborative robots that can be operated outside of cages or protective fencing, expanding opportunities to deploy them in new applications.
25 teams prepare for 2015 DARPA Robotics Challenge Finals
By David Szondy for Gizmag: On June 5 and 6, the 2015 DARPA Robotic Challenge (DRC) Finals will take place at Fairplex in Pomona, California. Open to the public, it will see 25 international teams compete for US$3.5 million in prizes as part of an effort to develop robots for disaster relief. Here's what to expect.
This year's challenge will see 25 teams competing. Half of the teams are from the United States, five are from Japan, three from Korea, two from Germany, one from Italy, one from Hong Kong, and one from the People’s Republic of China. They will be vying for a US$3.5 million total of prizes; including a $2 million first prize, a $1 million second prize, and a $500,000 third prize. The robots will be of a wide variety with some humanoid, some four-legged, and some tracked, but all will need to operate free of external power, mechanical support, and limited communications with their controllers.
The basic idea behind DRC 2015 is to make things much harder for the robots than previously.
Stephen Hawking fears robots could take over in 100 years
By Sharon Gaudin for ComputerWorld: Worried that one day we'll have robot overlords? You're in good company.
Reknowned physicist, cosmologist and author of A Brief History of Time, Stephen Hawking said this week that robots, powered by artificial intelligence (A.I.), could overtake humans in the next 100 years.
Speaking at the Zeitgeist conference in London, Hawking said: "Computers will overtake humans with AI at some within the next 100 years. When that happens, we need to make sure the computers have goals aligned with ours," according to a report in Geek.
This isn't the first time Hawking has spoken about the threat that comes along with machine learning, A.I. and robotics.
In December, Hawking said, "the development of full artificial intelligence could spell the end of the human race."
'First heavy-duty collaborative robot' can lift 35kg
Fanuc claims that it is the first robot manufacturer to produce a heavy-duty robot designed to work safely alongside humans. Its CR-35iA robot can perform tasks involving payloads of up to 35kg without needing the protective guards and fences that have previously been needed for robots with similar lifting capacities.
Although there are already several other collaborative robots on the market, most are designed for much lower payloads.
The new robot will stop automatically if it touches a human operator. A soft covering material also reduces the force of any impacts and prevents human operators from being pinched by the mechanism. And if the robot comes too close to an operator, they can simply push it away. The covering has a green colour to distinguish it from Fanuc’s usual yellow robots.
The six-axis robot is designed for duties such as transferring heavy workpieces or assembling parts. By avoiding the need for safety barriers, it is claimed to improve production efficiencies and allow higher levels of automation.
Building Enthusiasm for Construction Robotics
Vicki Speed for Inside Unmanned Systems: It would seem that robotic systems could provide an extra measure of safety, as well as a higher level of efficiency and machine-consistent quality. Yet, to date, the use of robotic systems on construction jobsites has been minimal. The building industry, however, is looking with fresh eyes at robots—including at least three new systems expected to be available this year—with a focus on near-term efficiencies that make investment in the systems make sense.
Demolition Days
Among the first fully-realized applications of robots in the construction environment are those used to support work that comes at the end of a structure’s life, namely demolition. In fact, remotely operated demolition robots have been around for more than a decade.
Robotic Building Blocks
The short answer is, ‘Yes.’ There are robotic systems in development around the world that can lay bricks, set tile or finish concrete floors.
Bionic Builders?
While not autonomous systems, robotic exoskeletons, those high-tech wearable suits seen in futuristic movies that help mere mortals defend Earth against other beings, could be a very real part of tomorrow’s jobsite and a possible precursor to autonomous robots in the field.
Microsoft HoloLens gets real with robotics, surgery, architecture
Jared Newman for PCWorld: At the 2015 Build conference, Microsoft tried to prove that HoloLens is more than just a neat gimmick.
The company showed off several new demos for its “mixed reality” headset, which can map digital imagery onto the user’s physical surroundings. While previous demos had focused on fun ideas like a virtual Mars walk and a living room-sized version of Minecraft, the Build presentation emphasized real-world applications for businesses and education.
For instance, Microsoft showed how architects could use HoloLens to interact with 3D models, laid out virtually in front of them on a table. They might also be able to examine aspects of a building site at full scale, with virtual beams and walls rendered before their eyes.
Not all the presentations were so serious. Microsoft also showed off an actual robot whose controls appeared in the virtual space above the robot’s head. Users could then create a movement pattern for the robot by tapping on the ground. Another demo showed how users could create their own personal screens that followed them around in real space.
Festo BionicANTs and eMotionButterflies
From Festo Bionic:
With the bionic butterflies, for the first time Festo combines the ultralight construction of artificial insects with collision-free flying behaviour in a collective. For coordination purposes, the eMotionButterflies make use of a guidance and monitoring system, which could be used in the networked factory of the future... ( additional info )
Like their natural role models, the BionicANTs work together under clear rules. They communicate with each other and coordinate their actions and movements among each other. The artificial ants thus demonstrate how autonomous individual components can solve a complex task together working as an overall networked system... ( additional info )
What You Wanted To Know About AI
Run down of the state of AI from FastML:
Let’s take a look at how advanced we are, really. Two representative and well known examples of the current state of the art are:
Automatic image annotation using a combination of convolutional and recurrent neural networks
DeepMind’s deep reinforcement learning for playing Atari games
( cont'd at FastML )
Runt Rover Robot Kits from Actobotics
From Servocity:
Simply mount your electronics using our innovative multi-board mounts that are compatible with a variety of micro-controllers; such as Raspberry Pi, Arduino and the SparkFun Redboard. The Runt Rovers™ are perfect for beginning light programming and educational applications... ( Servocity available options )
Robotnik's Mobile Manipulator RB-1
From Smashing Robotics:
Spanish company Robotnik introduced earlier this week their very own RB-1 mobile manipulator. The robot is designed for indoor use in household as well as professional environments, and is brought to life by using well known Dynamixel Pro series servo actuators which add up to 13 degrees of freedom (DOF), depending on variant. It is well suited for remote manipulation or human assistance applications and can be fully autonomous or manually controlled... ( full article ) ( datasheet )
Beating Super Hexagon With Computer Vision
Two great examples of using Computer Vision to beat Super Hexagon.
Super Hexagon is a really hard game. The goal of Super Hexagon is to control a small triangle which circles around a central hexagon (which occasionally collapses into a pentagon or square in the hexagon and hyper hexagon difficulty) attempting to avoid contact with incoming "walls".
First example from Valentin Trimaille's Super Hexagon bot:
Ray Casting Wall Detection
The point is that a bot for this game makes a really nice image processing project to start learning OpenCV: simple shapes but lots of human disturbing effects, fast-paced game meaning real-time is required, very simple controls: rotate CW or CCW... ( full article )
Second example from Shaun LeBron's Super Hexagon Unwrapper:
This project is written in Python. It employs Computer Vision algorithms provided by SimpleCV to establish a reference frame in the image. Then it warps (or "unwraps") the image based on that reference frame, using OpenGL fragment shaders... ( github code ) ( full explanation )
Proposed FAA Commercial UAS Rules
From IEEE Spectrum:
Unmanned aircraft must weigh less than 55 lbs. (25 kg).
Visual line-of-sight (VLOS) only; the unmanned aircraft must remain within VLOS of the operator or visual observer.
At all times the small unmanned aircraft must remain close enough to the operator for the operator to be capable of seeing the aircraft with vision unaided by any device other than corrective lenses.
Small unmanned aircraft may not operate over any persons not directly involved in the operation.
Daylight-only operations (official sunrise to official sunset, local time).
Must yield right-of-way to other aircraft, manned or unmanned.
May use visual observer (VO) but not required.
First-person view camera cannot satisfy “see-and-avoid” requirement but can be used as long as requirement is satisfied in other ways.
Maximum airspeed of 100 mph (87 knots).
Maximum altitude of 500 feet above ground level... (full article)
Records 826 to 839 of 839
Engineering - Featured Product
Using AI to Collect & Leverage Data
Data is the foundation of Industry 4.0. While skilled workers will always be essential, data is reshaping the manufacturing landscape, enabling automation of repetitive tasks, empowering smarter decision-making with AI assistance, and reducing defects and downtime. This shift allows small and medium-sized manufacturers (SMMs) to compete more effectively on quality, speed, and cost. While AI and machine learning systems typically require around two years to collect enough data to reach their full potential, manufacturers can start seeing benefits almost immediately with basic analysis tools and dashboards. Our MEP National NetworkTM expert explains how in this Manufacturing Tomorrow article.
