New RFID technology helps PR2 robots find household objects

A Georgia Tech research team has developed a new search algorithm that improves a robot’s ability to find and navigate to tagged objects.

Mobile robots could be much more useful in homes, if they could locate people, places and objects.

Today's robots usually see the world with cameras and lasers, which have difficulty reliably recognizing things and can miss objects that are hidden in clutter.

A complementary way robots can "sense" what is around them is through the use of small ultra-high frequency radio-frequency identification (UHF RFID) tags.

Inexpensive self-adhesive tags can be stuck on objects, allowing an RFID-equipped robot to search a room for the correct tag's signal, even when the object is hidden out of sight.

Once the tag is detected, the robot knows the object it's trying to find isn't far away.

"But RFID doesn't tell the robot where it is," said Charlie Kemp, an associate professor in Georgia Tech's Wallace H. Coulter Department of Biomedical Engineering.

"To actually find the object and get close to it, the robot has to be more clever."

A PR2 robot successfully navigates to a medication bottle. 

Credit: Georgia Tech/Travis Deyle

That's why Kemp, former Georgia Tech student Travis Deyle and University of Washington Professor Matthew Reynolds developed a new search algorithm that improves a robot's ability to find and navigate to tagged objects.

The team has implemented their system in a PR2 robot, allowing it to travel through a home and correctly locate different types of tagged household objects, including a medication bottle, TV remote, phone and hair brush.

The research was presented September 14-18 in Chicago at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

The researchers have equipped a PR2 robot with articulated, directionally sensitive antennas and a new algorithm that allows the robot to successfully find and navigate to the intended object.

Due to the physics of radio-frequency propagation, these antennas tend to receive stronger signals from a tag when they are closer to it and pointed more directly at it.

By moving around the antennas on its shoulders and driving around the room, the PR2 can figure out the direction it should move to get a stronger signal from a tag and thus become closer to a tagged object.

In essence, the robot plays the classic childhood game of "Hotter/Colder" with the tag telling the PR2 when it's getting closer to the target object.

Charlie Kemp is an associate professor in the Wallace H. Coulter Department of Biomedical Engineering. Credit: Georgia Tech

In contrast to other approaches, the robot doesn't explicitly estimate the 3D location of the target object, which significantly reduces the complexity of the algorithm.

"Instead the robot can use its mobility and our special behaviors to get close to a tag and oriented toward it," said Deyle, who conducted the study in Kemp's lab while earning his doctoral degree in Electrical and Computer Engineering from Georgia Tech.

Deyle, who currently works at Google, says the research has implications for future home robots and is particularly compelling for applications such as helping people with medicine, as RFID is able to provide precise identification information about an object or a person.

"This could allow a robot to search for, grasp and deliver the right medication to the right person at the right time," he added.

"RFID provides precise identification, so the risk of delivering the wrong medication is dramatically reduced. Creating a system that allows robots to accurately locate the correct tag is an important first step."

More information: 
Travis Deyle, Matt Reynolds and Charles C. Kemp, "Finding and Navigating to Household Objects with UHF RFID Tags by Optimizing RF Signal Strength." IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014. www.hsi.gatech.edu/hrl/pdf/iro… le_reynolds_kemp.pdf

JAMSTEC: Japan deep sea drilling boat Chikyu casts off to find Earthquakes

Japan's deep-sea drilling vessel Chikyu is anchored at Shimizu port, Shizuoka prefecture on September 11, 2013. 

A Japan-led team of seismologists set off Friday on a mission to drill deep beneath the seabed in a search for the origin of earthquakes.

A Japan-led team of seismologists set off Friday on a mission to drill deep beneath the seabed in a search for the origin of earthquakes.

The scientists weighed anchor on Japan's deep-sea drilling vessel Chikyu, heading for a spot in the ocean off the Kii peninsula, southwestern Japan, and a fracture in the Earth's crust known as the Nankai Trough.

Experts have warned the trough, which marks the place where the Philippine Sea plate slides under the Eurasian plate, is the likely source of a monster earthquake sometime in the near future.

Tamano Omata

Japan's government last year unveiled a worst-case scenario, warning a big quake in the area could kill over 320,000 people, dwarfing the March 11, 2011, quake-tsunami disaster.

In its four-month mission, the latest stage of a multi-year project that began in 2007, the team plans to drill 3,600 metres (2.2 miles) down and take samples from the crust.

They will also be readying for another trip next year in which they hope to get 5,200 metres down, to the spot where the action actually happens.

"It would be unprecedented to drill directly into a seismogenic zone, the area believed to release great energy and cause crusts to slide along fault lines and trigger tsunami," said Tamano Omata, a researcher for the Japan Agency for Marine-Earth Science and Technology (JAMSTEC).

Scientists want to plant sensors—such as seismometers, deformation-measuring devices and thermometers—in the zone, that will form part of a system called Dense Oceanfloor Network System for Earthquakes and Tsunamis (DONET), which is linked directly to onshore monitors.

"We expect to become able to monitor how the crusts move immediately before a quake hits," Omata said.

Shinichi Kuramoto

Shinichi Kuramoto, deputy director of JAMSTEC's Center for Deep Earth Exploration, said recent research has shown mild earthquakes, in which the two crusts slip gently past each other, have occurred frequently over stretches of the Nankai Trough in the past five years.

He said it was possible these were precursors to a mega-quake.

"Directly drilling into and observing the place that may release a big quake would be a big step towards understanding the seismological mechanism," he said.

The 56,752-ton Chikyu—"Earth" in Japanese—has been anchored in central Shimizu port, and was open to foreign press this week ahead of the mission.

The vessel, built in 2005 at a cost of $500 million, is equipped with a 121-metre (400-foot) drill tower that can descend 7,000 metres below the seabed, nearly three times as deep as its predecessors.

It depends on satellite location systems with pinpoint accuracy that allow its captain to know exactly where the ship is in relation to the Earth's crust.

Seismically-active Japan experiences 20 percent of the world's major earthquakes every year.

Building standards are high and its people are well-practised at taking cover when quakes strike, meaning damage and death tolls are often much lower than in other parts of the world.

But its proximity to major tectonic faults means the risk is ever-present.

DONET is a submarine cabled real-time seafloor observation infrastructure which was designed to realize precise earthquakes and tsunamis monitoring on seafloor in the long period of time. 

It consists of an approximately 300km length of backbone cable system, 5 science nodes, and 20 observatories. 

Its installation on 20 stations at Kumanonada started in 2006 and has been completed in July 2011.

Researchers find phosphate in more soluble form on Mars

Synthetic crystals of the calcium phosphate mineral whitlockite similar to those used to produce the extraterrestrial mineral merrillite. 

If life ever arose on Mars, merrillite may have been a major source of biologically required phosphate. Largest crystal are ~1mm. 

Credit: C. T. Adcock / University of Nevada Las Vegas 

A trio of researchers at the University of Nevada has found that phosphate found in minerals on Mars, is far more soluble than it is in natural Earth minerals.

In their paper published in the journal Nature Geoscience, the researchers describe how they synthesized mineral types found on Mars and then tested how well they dissolved in water releasing phosphate as compared to samples from natural Earth minerals.

Most scientists agree that phosphate is a key ingredient for life. Put another way, they believe that life couldn't have evolved without it.

For that reason, scientists have been studying ways in which minerals that contain phosphate could have broken down to allow the phosphate to escape.

Such studies have thus far found that minerals that hold phosphate on Earth are not very soluble—they don't break down easily when soaked in sea water.

That has led to what Earth scientists call "the phosphate problem." How did life get started on Earth if there wasn't enough phosphate around when life was first beginning?

Some have suggested the answer is that it didn't, instead, it started on another planet, such as Mars, and made its way here via meteorites.

Prior research has already shown that Mars has much more phosphate than does Earth. In this new effort, the team in Nevada looked at minerals that exist on Mars to see if they might be more soluble in water as well.

Lacking samples from Mars to test, the researchers synthesized chlorapatite and merrillite in their lab—two common phosphate bearing minerals found on the Red Planet.

They then soaked samples in several tubs, each with a different pH level for varying amounts of time. As they did so, they measured how much phosphate made its way into the water and how long it took.

In analyzing their results, the researchers found that more phosphate made its way into the water with both types of minerals and they did so at a faster rate than minerals that contain phosphate found naturally on Earth.

In some cases, they report that the Mars rocks released phosphate up to 45 percent faster than Earth rocks.

The findings by the team don't prove that life began on Mars and migrated to Earth—after all, scientists have yet to prove life ever existed Mars. But it does add some credence to the argument that perhaps life did start somewhere other than our home planet, which if true, might mean it's still out there waiting for us to discover it.

More information: Readily available phosphate from minerals in early aqueous environments on Mars, Nature Geoscience (2013) DOI: 10.1038/ngeo1923

Hubble Astronomers find new Neptune moon

This composite Hubble Space Telescope picture shows the location of a newly discovered moon, designated S/2004 N 1, orbiting Neptune. 

The black and white image was taken in 2009 with Hubble's Wide Field Camera 3 in visible light. 

Hubble took the colour inset of Neptune on August 2009. 

Credit: NASA, ESA, M. Showalter/SETI Institute

NASA and ESA's Hubble Space Telescope has discovered a new moon orbiting the distant blue-green planet Neptune, the 14th known to be circling the giant planet.

The moon, designated S/2004 N 1, is estimated to be no more than 12 miles across, making it the smallest known moon in the Neptunian system.

It is so small and dim that it is roughly 100 million times fainter than the faintest star that can be seen with the naked eye.

It even escaped detection by NASA's Voyager 2 spacecraft, which flew past Neptune in 1989 and surveyed the planet's system of moons and rings.

Mark Showalter of the SETI Institute in Mountain View, Calif., found the moon July 1, while studying the faint arcs, or segments of rings, around Neptune.

"The moons and arcs orbit very quickly, so we had to devise a way to follow their motion in order to bring out the details of the system," he said.

"It's the same reason a sports photographer tracks a running athlete—the athlete stays in focus, but the background blurs."

The method involved tracking the movement of a white dot that appears over and over again in more than 150 archival Neptune photographs taken by Hubble from 2004 to 2009.

This diagram shows the orbits of several moons located close to the planet Neptune. 

All of them were discovered in 1989 by NASA's Voyager 2 spacecraft, with the exception of S/2004 N 1, which was discovered in archival Hubble Space Telescope images taken from 2004 to 2009.

The moons all follow prograde orbits and are nestled among Neptune's rings (not shown). 

The outer moon Triton was discovered in 1846 — the same year the planet itself was discovered. 

Triton's orbit is retrograde, suggesting it is a captured Kuiper Belt object and therefore a distant cousin of Pluto. 

The inner moons may have formed after Triton's capture several billion years ago. 

Credit: NASA, ESA, and A. Feild (STScI)

On a whim, Showalter looked far beyond the ring segments and noticed the white dot about 65,400 miles from Neptune, located between the orbits of the Neptunian moons Larissa and Proteus.

The dot is S/2004 N 1. Showalter plotted a circular orbit for the moon, which completes one revolution around Neptune every 23 hours.

Durham Uni's Galactic Archaeologists Find Origin Of Milky Way's Ancient Stars

This simulation shows a Milky Way-like galaxy around five billion years ago when most satellite galaxy collisions were happening. Credit: Andrew Cooper/John Helly, Durham University

Many of the Milky Way's ancient stars are remnants of other smaller galaxies torn apart by violent galactic collisions around five billion years ago, according to researchers at Durham University.

Scientists at Durham's Institute for Computational Cosmology and their collaborators at the Max Planck Institute for Astrophysics, in Germany, and Groningen University, in Holland, ran huge computer simulations to recreate the beginnings of our galaxy.

The simulations revealed that the ancient stars, found in a stellar halo of debris surrounding the Milky Way, had been ripped from smaller galaxies by the gravity generated by colliding galaxies.

Cosmologists predict that the early Universe was full of small galaxies which led short and violent lives. These galaxies collided with each other leaving behind debris which eventually settled into more familiar looking galaxies like the Milky Way.

The researchers say their finding supports the theory that many of the Milky Way's ancient stars had once belonged to other galaxies instead of being the earliest stars born inside the galaxy when it began to form about 10 billion years ago.

The research, funded in the UK by the STFC, appears in the Monthly Notices of the Royal Astronomical Society.

Lead author Andrew Cooper, from Durham University's Institute for Computational Cosmology, said: "Effectively we became galactic archaeologists, hunting out the likely sites where ancient stars could be scattered around the galaxy.

"Our simulations show how different relics in the galaxy today, like these ancient stars, are related to events in the distant past.

"Like ancient rock strata that reveal the history of Earth, the stellar halo preserves a record of a dramatic primeval period in the life of the Milky Way which ended long before the Sun was born."

Four-legged tracks in mud cause paleontologists re-think

Muddy tracks left 397 million years ago may reset the age when the first four-legged animals crawled onto land.

In the Nature journal study, reported Wednesday and led by Grzegorz Niedz´wiedzki of Poland's Warsaw University, a European team reports "numerous" tracks from the Zachelmie mountain quarry that reveal the era when clumsy-limbed fishes forsook the seas to live on land.

The tracks show that four-legged creatures, one of them more than 8 feet long, probably originated in lagoons, not swamps, as suggested by past fossil finds.

A beach "provides a ready food source of stranded marine animals," the study's authors say. That would have allowed marine

ancestors of four-legged creatures to evolve "terrestrial competence" while dining on "a new and essentially untouched resource."

The trackway discovery "is undoubtedly one of the most extraordinary and unexpected finds in paleontology for decades, perhaps a century," says paleontologist John Long of the Natural History Museum of Los Angeles County.

Until the discovery, paleontologists thought the evolutionary transition from fishes to land animals was tightly nailed down to fossils dated about 379 million years old, he says.

The tracks not only push back the age of the first four-legged animals by a "whopping 18 million years," he says, but also "show that quite advanced kinds of tetrapods (four-limbed creatures), with incipient wrist joints, had already appeared by this much earlier stage of animal evolution."

The tracks "lob a grenade" into fossil studies of the earliest land creatures, Philippe Janvier and Gael Clement of the Muséum National d'Histoire Naturelle in Paris say in a commentary accompanying the study.

Paleontologists had considered fish-like creatures such as "Tiktaalik," a discovery that made headlines in 2006, as the likely earliest land animals. Instead, the trackways suggest Tiktaalik was a relic species still adapting to land long after other species had made the transition.

Not so fast, Tiktaalik's discoverers say. "Trace fossils such as these presumed vertebrate tracks and trackways, however, are a notoriously difficult class of evidence to interpret with full confidence," says Ted Daeschler of the Academy of Natural Sciences in Philadelphia by e-mail. "With all respect to the scientists involved in this study, there may be other explanations for these suggestive tracks."

The study authors acknowledge they have only fossil tracks, not bones, from their four-legged creature, which perhaps resembled an oversized salamander. Based on its gait, the critter floated in the tide while making tracks. The authors recommend from their finding that other paleontologists look for fossils of four-legged creatures as far back as 420 million years ago.

In Liquid and Air, Scientists Find Order Among the Chaos - NYTimes.com

In Liquid and Air, Scientists Find Order Among the Chaos - NYTimes.com

Assisted by instruments that can track in fine detail how parcels of fluid move, and by low-cost computers that can crunch vast amounts of data quickly, researchers have found hidden structures beyond Monterey Bay, structures that explain why aircraft meet unexpected turbulence, why the air flow around a car causes drag and how blood pumps from the heart's ventricles. In December, the journal Chaos will highlight the research under way to track the moving skeletons embedded in complex flows, known as Lagrangian coherent structures.

Shared via AddThis