Forum for Science, Industry and Business

Sponsored by:     3M 
Search our Site:

 

Helping robots learn to see in 3-D

17.07.2017

Robots need to guess what they're seeing better, even when parts are hidden from view

Autonomous robots can inspect nuclear power plants, clean up oil spills in the ocean, accompany fighter planes into combat and explore the surface of Mars.


When fed 3-D models of household items in bird's-eye view (left), a new algorithm is able to guess what the objects are, and what their overall 3-D shapes should be. This image shows the guess in the center, and the actual 3-D model on the right.

Courtesy of Ben Burchfiel

Yet for all their talents, robots still can't make a cup of tea.

That's because tasks such as turning the stove on, fetching the kettle and finding the milk and sugar require perceptual abilities that, for most machines, are still a fantasy.

Among them is the ability to make sense of 3-D objects. While it's relatively straightforward for robots to "see" objects with cameras and other sensors, interpreting what they see, from a single glimpse, is more difficult.

Duke University graduate student Ben Burchfiel says the most sophisticated robots in the world can't yet do what most children do automatically, but he and his colleagues may be closer to a solution.

Burchfiel and his thesis advisor George Konidaris, now an assistant professor of computer science at Brown University, have developed new technology that enables machines to make sense of 3-D objects in a richer and more human-like way.

A robot that clears dishes off a table, for example, must be able to adapt to an enormous variety of bowls, platters and plates in different sizes and shapes, left in disarray on a cluttered surface.

Humans can glance at a new object and intuitively know what it is, whether it is right side up, upside down or sideways, in full view or partially obscured by other objects.

Even when an object is partially hidden, we mentally fill in the parts we can't see.

Their robot perception algorithm can simultaneously guess what a new object is, and how it's oriented, without examining it from multiple angles first. It can also "imagine" any parts that are out of view.

A robot with this technology wouldn't need to see every side of a teapot, for example, to know that it probably has a handle, a lid and a spout, and whether it is sitting upright or off-kilter on the stove.

The researchers say their approach, which they presented July 12 at the 2017 Robotics: Science and Systems Conference in Cambridge, Massachusetts, makes fewer mistakes and is three times faster than the best current methods.

This is an important step toward robots that function alongside humans in homes and other real-world settings, which are less orderly and predictable than the highly controlled environment of the lab or the factory floor, Burchfiel said.

With their framework, the robot is given a limited number of training examples, and uses them to generalize to new objects.

"It's impractical to assume a robot has a detailed 3-D model of every possible object it might encounter, in advance," Burchfiel said.

The researchers trained their algorithm on a dataset of roughly 4,000 complete 3-D scans of common household objects: an assortment of bathtubs, beds, chairs, desks, dressers, monitors, nightstands, sofas, tables and toilets.

Each 3-D scan was converted into tens of thousands of little cubes, or voxels, stacked on top of each other like LEGO blocks to make them easier to process.

The algorithm learned categories of objects by combing through examples of each one and figuring out how they vary and how they stay the same, using a version of a technique called probabilistic principal component analysis.

When a robot spots something new -- say, a bunk bed -- it doesn't have to sift through its entire mental catalogue for a match. It learns, from prior examples, what characteristics beds tend to have.

Based on that prior knowledge, it has the power to generalize like a person would -- to understand that two objects may be different, yet share properties that make them both a particular type of furniture.

To test the approach, the researchers fed the algorithm 908 new 3-D examples of the same 10 kinds of household items, viewed from the top.

From this single vantage point, the algorithm correctly guessed what most objects were, and what their overall 3-D shapes should be, including the concealed parts, about 75 percent of the time -- compared with just over 50 percent for the state-of-the-art alternative.

It was also capable of recognizing objects that were rotated in various ways, which the best competing approaches can't do.

While the system is reasonably fast -- the whole process takes about a second -- it is still a far cry from human vision, Burchfiel said.

For one, both their algorithm and previous methods were easily fooled by objects that, from certain perspectives, look similar in shape. They might see a table from above, and mistake it for a dresser.

"Overall, we make a mistake a little less than 25 percent of the time, and the best alternative makes a mistake almost half the time, so it is a big improvement," Burchfiel said. "But it still isn't ready to move into your house. You don't want it putting a pillow in the dishwasher."

Now the team is working on scaling up their approach to enable robots to distinguish between thousands of types of objects at a time.

"Researchers have been teaching robots to recognize 3-D objects for a while now," Burchfield said. What's new, he explained, is the ability to both recognize something and fill in the blind spots in its field of vision, to reconstruct the parts it can't see.

"That has the potential to be invaluable in a lot of robotic applications," Burchfiel said.

###

This research was supported in part by The Defense Advanced Research Projects Agency, DARPA (D15AP00104).

CITATION: "Bayesian Eigenobjects: A Unified Framework for 3D Robot Perception," Benjamin Burchfiel and George Konidaris. RSS 2017, July 12-16, 2017, Cambridge, Massachusetts.

Media Contact

Robin Ann Smith
ras10@duke.edu
919-681-8057

 @DukeU

http://www.duke.edu 

Robin Ann Smith | EurekAlert!

More articles from Information Technology:

nachricht NASA CubeSat to test miniaturized weather satellite technology
10.11.2017 | NASA/Goddard Space Flight Center

nachricht New approach uses light instead of robots to assemble electronic components
08.11.2017 | The Optical Society

All articles from Information Technology >>>

The most recent press releases about innovation >>>

Die letzten 5 Focus-News des innovations-reports im Überblick:

Im Focus: A “cosmic snake” reveals the structure of remote galaxies

The formation of stars in distant galaxies is still largely unexplored. For the first time, astron-omers at the University of Geneva have now been able to closely observe a star system six billion light-years away. In doing so, they are confirming earlier simulations made by the University of Zurich. One special effect is made possible by the multiple reflections of images that run through the cosmos like a snake.

Today, astronomers have a pretty accurate idea of how stars were formed in the recent cosmic past. But do these laws also apply to older galaxies? For around a...

Im Focus: Visual intelligence is not the same as IQ

Just because someone is smart and well-motivated doesn't mean he or she can learn the visual skills needed to excel at tasks like matching fingerprints, interpreting medical X-rays, keeping track of aircraft on radar displays or forensic face matching.

That is the implication of a new study which shows for the first time that there is a broad range of differences in people's visual ability and that these...

Im Focus: Novel Nano-CT device creates high-resolution 3D-X-rays of tiny velvet worm legs

Computer Tomography (CT) is a standard procedure in hospitals, but so far, the technology has not been suitable for imaging extremely small objects. In PNAS, a team from the Technical University of Munich (TUM) describes a Nano-CT device that creates three-dimensional x-ray images at resolutions up to 100 nanometers. The first test application: Together with colleagues from the University of Kassel and Helmholtz-Zentrum Geesthacht the researchers analyzed the locomotory system of a velvet worm.

During a CT analysis, the object under investigation is x-rayed and a detector measures the respective amount of radiation absorbed from various angles....

Im Focus: Researchers Develop Data Bus for Quantum Computer

The quantum world is fragile; error correction codes are needed to protect the information stored in a quantum object from the deteriorating effects of noise. Quantum physicists in Innsbruck have developed a protocol to pass quantum information between differently encoded building blocks of a future quantum computer, such as processors and memories. Scientists may use this protocol in the future to build a data bus for quantum computers. The researchers have published their work in the journal Nature Communications.

Future quantum computers will be able to solve problems where conventional computers fail today. We are still far away from any large-scale implementation,...

Im Focus: Wrinkles give heat a jolt in pillared graphene

Rice University researchers test 3-D carbon nanostructures' thermal transport abilities

Pillared graphene would transfer heat better if the theoretical material had a few asymmetric junctions that caused wrinkles, according to Rice University...

All Focus news of the innovation-report >>>

Anzeige

Anzeige

Event News

Ecology Across Borders: International conference brings together 1,500 ecologists

15.11.2017 | Event News

Road into laboratory: Users discuss biaxial fatigue-testing for car and truck wheel

15.11.2017 | Event News

#Berlin5GWeek: The right network for Industry 4.0

30.10.2017 | Event News

 
Latest News

Antarctic landscape insights keep ice loss forecasts on the radar

20.11.2017 | Earth Sciences

Filling the gap: High-latitude volcanic eruptions also have global impact

20.11.2017 | Earth Sciences

Water world

20.11.2017 | Life Sciences

VideoLinks
B2B-VideoLinks
More VideoLinks >>>