A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation
From Computer Vision Freiburg: Recent work has shown that optical flow estimation can be formulated as a supervised learning task and can be successfully solved with convolutional networks. Training of the so-called FlowNet was enabled by a large synthetically generated dataset. The present paper extends the concept of optical flow estimation via convolutional networks to disparity and scene flow estimation. To this end, we propose three synthetic stereo video datasets with sufficient realism, variation, and size to successfully train large networks. Our datasets are the first large-scale datasets to enable training and evaluating scene flow methods. Besides the datasets, we present a convolutional network for real-time disparity estimation that provides state-of-the-art results. By combining a flow and disparity estimation network and training it jointly, we demonstrate the first scene flow estimation with a convolutional network.
his video shows impressions from various parts of our dataset, as well as state-of-the-art realtime disparity estimation results produced by one of our new CNNs... (full paper)
Kagan Pittman for Engineering.com: As Chinese companies move to expand their market share in industrial robotics, so too are companies in the West, with some of the biggest power-plays coming from R&D departments.
A recent report by Technavio predicts that global R&D spending in the robotics industry will grow at a CAGR of more than 17 percent between 2016-2020.
Technavio analysts cite the following factors as key drivers in the growth of R&D spending:
- A race for robotics patents
- Demand for lower system engineering and installation costs
- Growing demand for industrial robots from non-automotive industries
The report’s analysis considers market trends across user segments including: defense, healthcare, automotive, domestic, food and beverage, electricals, electronics, oil and gas, textiles and packaging. Cont'd...
Caitlin Ju for The Stanford Daily: Stanford researchers in the Computational Vision and Geometry Lab have designed an autonomously navigating robot prototype that they say understands implicit social conventions and human behavior. Named “Jackrabbot” after the swift but cautious jackrabbit, the visually intelligent and socially amicable robot is able to maneuver crowds and pedestrian spaces.
A white ball on wheels, the Jackrabbot is built on a Segway system and contains a computing unit and multiple sensors that acquire a 3-D structure of the surrounding environment. 360-degree cameras and GPS also enhance the robot’s navigation and detection capabilities.
To interact smoothly in public settings, the robot has to know how to avoid someone in a natural way, how to yield the right-of-way and how to leave personal space, among other skills. Cont'd...
From Gizmag: Researchers at Siemens have created a new prototype electric motor specifically designed for aircraft that weighs in at just 50 kg (110 lb) and is claimed to produce about 260 kW (348 hp) at just 2,500 RPM. With a quoted power five times greater than any comparable powerplant, the new motor promises enough grunt to get aircraft with take-off weights of up to 1,800 kg (2 ton) off the ground...
... As a result, the new aircraft electric drive system achieves a claimed weight-to-performance ratio of 5 kW per kilogram. This ratio is an exceptional figure – especially if compared to similarly powerful industrial electric motors used in heavy machinery that produce less than 1 kW per kilogram, or even to more efficient electric motors for vehicles that generate around 2 kW per kilogram... (full article)
Manish Sablokk for IoTTech: Cutting-edge robots and other advanced smart machines are set to be added into the rapidly expanding Internet of Things, which is projected to reach 25 billion devices by 2020. Robotics has already been used in manufacturing to great effect for over a decade, performing delicate and precise tasks with a higher success rate than humans. With advancements such as 'deep learning' robots, delivery drones and ubiquitous knowledge-sharing between machines, widespread robotics adoption is becoming far more feasible.
In healthcare, there are already robotic services in operation with automated pharmacy dispensing and robotic trolleys - robots that can navigate between floors and even call the lift using a Wi-Fi sensor. The hospitality sector has also been a keen adopter of robotics to deliver services and in education, robots are being deployed successfully as a tutor, tool or peer in learning activities, providing language, science and technology education. Cont'd...
ABIGAIL BEALL FOR MAILONLINE: Many people spend their childhood peering up into the vast expanse of the sky, dreaming of growing up to become an astronaut. But these dreams could be dashed as the idea of people venturing into space will one day become a distant memory, according to a report published today. Robots will eventually have enough capabilities to replace humans and other animals on space missions, experts have said. Many missions involving humans in space are dangerous and expensive. But for years robots have been sent to places humans could not venture, like the rovers venturing to the edges of our solar system. According to European Space Agency (Esa) Astronaut Roberto Vittori, who launched a paper on space robotics and autonomous systems, robots can help carry out these dangerous missions. Cont'd...
JG Randall for The National Interest: Things tend to happen in threes. An unlikely triumvirate on the surface, it would appear that Asimov’s laws on robotics and the UN Convention on Conventional Weapons (CCW) will outflank the Third Offset—the nation’s search for its next silver bullet in war fighting is robotics—knowing that many nations will agree on moral grounds. These nations will reject Asimov based on semantics, and though the debate might be perceived as strictly academic, or even rhetorical, it is worth discussing for the sake of a good cautionary tale. Because, whether we like it or not, killer bots are coming to a theater of operation near you.
Before we get deep in the weeds, let’s get some clarity. First, let’s outline Asimov’s robotic laws. The Three Laws of Robotics are a set of rules devised by the science fiction author Isaac Asimov. They were introduced in his 1942 short story “Runaround,” although they had been foreshadowed in earlier stories. Cont'd...
Phys.org: Soft robots do a lot of things well but they're not exactly known for their speed. The artificial muscles that move soft robots, called actuators, tend to rely on hydraulics or pneumatics, which are slow to respond and difficult to store.
Dielectric elastomers, soft materials that have good insulating properties, could offer an alternative to pneumatic actuators but they currently require complex and inefficient circuitry to deliver high voltage as well as rigid components to maintain their form—both of which defeat the purpose of a soft robot.
Now, researchers at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) have developed a dielectric elastomer with a broad range of motion that requires relatively low voltage and no rigid components. They published their work recently in Advanced Materials. Cont'd...
Ford Motor Company announced today its early testing of a new type of assembly line robot that were co-developed with German robotics company KUKA Roboter GmbH with the intention of assisting human line workers. Two of these three-foot-tall machines are in use at the Cologne, Germany factory, where they assist human workers to install shock absorbers on Ford Fiestas. These workers would have originally had to juggle the shocks and tools to install them, but now the robot helps them position and install the parts. More...
Thomas Claburn for InformationWeek: Among programmers, there's a principle called DRY, which stands for "Don't repeat yourself." It's an attempt to avoid writing code that duplicates the function of other code.
DRY embodies the same resistance to needless repetition as the more common idiom, "Don't reinvent the wheel."
Among those making robots, a group that includes software and hardware engineers attempts to adhere to these principles, as can be seen in designs that borrow from nature, from the evolved forms of life on Earth.
Biomimicry and bioinspired design provide a way to avoid reinventing the wheel. The biological systems of living things have been honed through eons of Darwinian user testing.
Borrowing aspects of animal physiology isn't the only option or necessarily the best option for robot designers. For some purposes, something new may be necessary. For others, biomechanically systems can't be easily duplicated. Cont'd...
CADE METZ for WIRED: HANNS TAPPEINER TYPES a few lines of code into his laptop and hits “return.” A tiny robot sits beside the laptop, looking like one of those anthropomorphic automobiles that show up in Pixar’s Cars movies. Almost instantly, it wakes up, rolls down the table, and counts to four. This is Cozmo—an artificially intelligent toy robot unveiled late last month by San Francisco startup Anki—and Tappeiner, one of the company’s founders, is programming the little automaton to do new things.
The programs are simple—he also teaches Cozmo to stack blocks—but they’re supposed to be simple. Tappeiner is using Anki’s newly unveiled software development kit—an SDK, in coder parlance—that he says even the greenest of coders can use to tweak the behavior of the toy robot. And that’s a big deal, at least according to Anki. The company claims the SDK is the first of its kind: a kit that lets anyone program such an intelligent robot, a robot that recognizes faces and navigates new environments and even mimics emotions. With the kit, Tappeiner says, “we’re trying to advance the field of robotics.” He compares the move to Apple letting people build apps for the iPhone. Cont'd...
Samuel Bouchard for Engineering.com: Collaborative robots (also known as cobots) are changing how robots and humans interact in our factories and manufacturing facilities.
No longer separated by cages, humans and cobots can work beside each other on complex tasks from picking and placement to assembly and logistics.
Human-cobot systems bring together the best of human capabilities (complex reasoning, ease of learning new tasks, pattern and object recognition in cluttered environments) and robot functionality (the ability to perform complex, tedious tasks 24/7 and with high precision).
The close proximity between humans and cobots and its advantages are exciting for manufacturers, SMEs, and the robotics industry, but it also brings a unique set of safety challenges.
Enter ISO/TS 15066 – the world's first specifications of safety requirements for collaborative robot applications. Cont'd...
From New York Magazine: Snowden’s body might be confined to Moscow, but the former NSA computer specialist has hacked a work-around: a robot. If he wants to make his physical presence felt in the United States, he can connect to a wheeled contraption called a BeamPro, a flat-screen monitor that stands atop a pair of legs, five-foot-two in all, with a camera that acts as a swiveling Cyclops eye. Inevitably, people call it the “Snowbot.” The avatar resides at the Manhattan offices of the ACLU, where it takes meetings and occasionally travels to speaking engagements. (You can Google pictures of the Snowbot posing with Sergey Brin at TED.) Undeniably, it’s a gimmick: a tool in the campaign to advance Snowden’s cause — and his case for clemency — by building his cultural and intellectual celebrity. But the technology is of real symbolic and practical use to Snowden, who hopes to prove that the internet can overcome the power of governments, the strictures of exile, and isolation... (full article)
Linda A. Thompson for Bloomberg: European lawmakers warn that the growing use of robots and artificial intelligence may cause job losses across the continent, threatening to result in plummeting tax revenues if current tax frameworks aren't revised to account for the rise of the robotic workforce.
Practitioners told Bloomberg BNA that taxing robots as “electronic persons,” as the EU contemplates in a recent report, would hinder innovation and that other ways of taxing the value that robotics create should be explored.
The recent European Parliament Committee on Legal Affairs draft reportrecommends the European Commission adopt a resolution to require companies to report on “the extent and proportion of the contribution of robotics and AI to the economic results of a company for the purpose of taxation and social security contributions.” Its first paragraph references Frankenstein, and comes amid mounting concerns that the rise in automation and artificial intelligence in the workplace will fundamentally alter economies, destroy jobs and jeopardize social welfare programs such as social security. Cont'd...
From All About Circuits: Google ATAP is bringing touchless interfaces to the market using a miniaturized radar chip no bigger than a dime. This is Project Soli.
Soli’s radar sensor is a marvel in many respects. For one thing, it solves a long-lived issue when it comes to gesture-recognition technology. Previous forays into the topic yielded almost-answers such as stereo cameras (which have difficulty understanding the overlap of fingers, e.g.) and capacitive touch sensing(which struggles to interpret motion in a 3D context).
Google ATAP’s answer is radar.
Radar is capable of interpreting objects’ position and motion even through other objects, making it perfect for developing a sensor that can be embedded in different kinds of devices like smartphones... (full article)
Records 121 to 135 of 700