Scientists and engineers at Florida Atlantic University have presented the new robot dog defined as “intelligent” that “sees and hears.” Unlike other robot dogs, this time the researchers thought they also had an aesthetic side: in addition to being a quadruped, the robot is characterized by a head, printed in 3D, which makes it look like a Doberman.
Precisely the head, in any case, serves to contain the computerized brain whose decisions are based on the techniques of deep learning and in general of artificial intelligence. The computer in the dog’s head is in fact “trained” through a deep neural network, a sort of basic computerized simulation of the brain.
This means that the dog should learn from the experience following all the data it receives through the sensors, cameras and microphones it is equipped with. It around 90 pounds, it is still considered by its creators a puppy in training.
For the moment the robot is able to respond to commands such as “sit down,” “get up,” and “lie down.” According to the creators themselves, however, it should also be able to understand and respond to manual signals and voice signals in various languages. It might even recognize the hands and faces of other dogs.
However, it is not a mere toy: the main missions it can carry out are related to safety. It will be able to detect the presence of guns, explosives, and anything else dangerous to help police and security personnel. It could also be used as a service dog for the blind or as a rescuer for search and rescue missions following natural disasters, such as hurricanes or earthquakes.
The robot has in fact been designed to be able to move even in uneven terrain and could make autonomous decisions, therefore not simply being dependent on the commands entered remotely.
The various sensory inputs are processed by a special Nvidia Jetson TX2 graphics card with four combined teraflops of computing power.
What happened to the cave bear, one of the largest species of mammals extinct by the end of the last ice age? This is a question that has never had definite answers but now a team of researchers from various European institutions, including the University of Zurich, suggests what could be the solution to the riddle.
In a study published in Scientific Reports, the researchers deduce that it was not the climatic effects of the ice age itself that made these bears extinct thousands of years ago, but it was humans.
Analyzing the mitochondrial DNA of 59 remains of cave bears from various European areas, the researchers learned that the populations of these animals began to fall in number before the beginning of the last ice age, or 40,000 years ago.
Furthermore, the same researchers deduced that the bears of previous generations had already managed to overcome the other ice ages without significant population decreases. At the same time we know that modern humans began to populate the areas in which these bear populations lived about 40,000 years ago, all information that suggests that it was precisely the arrival of modern humans to decree its end.
The same researchers also confirmed that in these times even Neanderthals lived in these areas. However, the latter had lived with cave bears for tens of thousands of years and it is therefore unlikely that they were responsible for extinction. This may also mean that modern humans had much more effective and lethal hunting techniques than Neanderthals.
Cave bears were hunted not only for food but also for furs or to eliminate potential competitors in the use of the caves themselves as a dwelling.
A group of researchers has discovered eumelanin, a natural pigment that is also found in human eyes, in the fossilized eyes of crane flies (Tipulidae) dating back 54 million years ago. It was a surprise for the researchers themselves as it was believed that in arthropods there were no melanin pigments in the visual system.
To underline the surprise of the discovery is Johan Lindgren, lead author of the study and professor in the Department of Geology of the University of Lund: “We were surprised by what we discovered because we weren’t looking for it or we didn’t expect it.” After making the discovery in the fossils, the researchers then examined the eyes of today’s crane flies and found that substance also in these species.
Then comparing the fossilized eyes with the eyes of today’s species, the researchers found that the fossilized eyes were characterized by the presence of calcified homatatid lenses. According to Lindgren, it was this mineral that replaced the original chitinous material in the fossil.
And this has led the researchers themselves to reconsider a widespread hypothesis concerning the evolution of visual systems in animals. Previous research had, in fact, suggested that trilobites, a group of extinct sea arthropods, possessed mineralized lenses during their lifetime.
As Lindgren explains, “the general opinion was that the trilobites had lenses made from single crystals of calcium carbonate. However, they were probably much more similar to modern arthropods because their eyes were mainly biological.”
Analyzing half a ton of fresh snow taken from Antarctica, a group of scientists discovered traces of a particular form of iron that is not naturally produced in nature and that most likely comes from space.
Dominik Koll, a physicist at the Australian National University of Canberra and lead author of the study, discovered the rare iron-60 isotope in the ice and snow of the Antarctic. This isotope boasts four more neutrons than the common form and is thought to have settled on the earth’s surface millions of years ago.
However, what has been found in the snow in Antarctica seems to have accumulated over the last two decades. According to the researchers, these particles come from outside the solar system since all the objects that are in the solar system are made more or less from the same materials since they were formed from the same huge cloud of gas and matter.
It could have been the impact of an interstellar meteor, a space body coming from outside the solar system, an event that is very rare. The researchers also ruled out that it could have come from nuclear reactors, tests or accidents. They also excluded that it could prevent from cosmic rays that generate iron-60 when they interact with space dust and meteorites.
The most probable hypothesis, according to Koll himself, is that these particles come from a supernova, “not so close to kill us but not too far away to be diluted in space.” The Earth must have captured these particles as it traveled through the so-called Local Interstellar Cloud, a thirty-year-old light region that the solar system is currently passing through and is expected to continue to traverse for the next 10,000-20,000 years.
A biosensor that relatively quickly detects even small amounts of salmonella in food has been developed by a group of researchers at the University of Missouri. This is a device designed especially for food producers to identify traces of salmonella in the most efficient way possible.
Currently, the tests that are used to understand if there is salmonella in food go back to cooking samples or extracting DNA to understand the presence of pathogens. These methods are accurate but unfortunately, they lose a lot of time (from one to five days), as pointed out by Mahmoud Almasri, professor of electrical and computer engineering at the College of Engineering of the aforementioned university, as well as one of the authors of the study.
The impedance-based microfluidic biosensor uses a specific fluid that can be mixed with food to detect the presence of salmonella bacteria, primarily those of salmonella B and D, both in raw foods and ready-to-eat foods. In a few hours, the duration of a worker’s shift, as is also emphasized in the press release, it is possible to know the presence of harmful bacteria.
“Our device will allow us to monitor and verify that food products are safe for consumers and to reduce the amount of food recalls that occur,” says Shuping Zhang, a professor at the Veterinary Medical Diagnostic Laboratory of the American University and another author of the study.
To understand how any intelligent alien civilizations could observe the Earth and eventually understand that life exists on our planet, a group of researchers has decided to transform and adapt the images we possess of the only planet with the presence of life we know.
We speak of course of the Earth and of the images taken from satellites to other space vehicles, images that the same researchers have analyzed and modified to make them appear possibly similar to those that any alien astronomers would see observing us.
They have worked on several thousand images of the Earth captured by the American satellite Deep Space Climate Observatory, a satellite that is located in an interesting observation point where there is a sort of gravitational balance between the Earth and the Sun. The images, which are taken at 10 specific wavelengths, they were modified by the researchers who worked on the light curves and were able to understand which parameters of these curves corresponded to the ground and which to other details such as cloud cover.
After understanding these relationships, they adapted them to the Earth’s rotation. The result is a sort of terrestrial map that presents various approximate contours of the main continents, contours represented by a black line that roughly follows the coastlines, and various areas of different colors that indicate the ocean and the emerged lands.
A map that the same researchers define as “the first two-dimensional (2D) surface map of the Earth reconstructed from light curve observations without hypotheses on its spectral properties” in the abstract of the study presented for the time being Su arXiv.
The same study could be useful to reconstruct the surface features of Earth-like exoplanets in future observations.
Duckweed, considered one of the fastest-growing plants in the world, could represent one of the key solutions to fight hunger in the world, especially in the coming decades as it is expected that the world population should reach a peak of 9.7 billion people in 2050.
Duckweed is an aquatic plant that reproduces very quickly so that often the freshwater mirrors in which it grows can be completely covered by a carpet of leaves. It is mainly used to produce starch which in turn can be used for the production of ethanol. However, it represents a source of traditional and primary food for different populations living in south-east Asia.
According to Eric Lam, a professor in the Plant Biology Department of Rutgers University-New Brunswick, this small plant, relatively easy to grow, could be the solution. It is more nutritious than salad, has an excellent content of vitamin fibers and offers different nutritional benefits so that some species (the family includes 37) are also used in traditional folk medicine in various regions of Southeast Asia.
It tastes not very strong and chopped or smoothed it can be mixed with other ingredients. Lam himself claims to have tried it also in a sandwich with a hamburger. The same researcher has calculated that the variety of lentils of water that grows faster can produce up to 20 grams (once it has been dried) per square meter every day, a quantity 50 times higher than what can be obtained from corn.
The same seedling, not being very large, can also be cultivated at home or in small gardens with very limited extensions. The same researchers in Lam’s laboratory are trying to maximize culture techniques even further and are also working on automated harvesting methods to reduce reproduction costs even further.
In 20 years these little seedlings, which are almost unknown here in the West, could be one of the typical dishes or side dishes on our tables.
A study published in the Astronomical Journal focused on calculating the probability of the existence of Earth-like planets not only with regard to the size and type of planet but also as regards its distance from the star.
Knowing more precisely the level of this frequency could help to make future observations more profitable, in particular those that will be carried out with future space telescopes, and in general for the search for planets in the habitable zone and therefore also for extraterrestrial life.
The research group, led by Eric B. Ford, professor of astrophysics and astronomy at Pennsylvania State University, used the substantial data set collected over the years by the Kepler space telescope. The main difficulty that the researchers had to overcome, however, lay in the method that Kepler himself used over the years to identify the planets, ie the transit method.
This method makes it easier to find the larger planets closer to the star, but the data is not very useful for finding the smallest and most distant planets. This is precisely why researchers have designed a new study method that simulates universes of stars and planets and then “observes” these simulated universes to determine the number of planets that Kepler would have discovered in each simulated universe.
Danley Hsu, the first author of the study and a student at Penn State, also explains this: “We used the final catalog of the planets identified by Kepler and improved the stellar properties of the European Space Agency’s Gaia spacecraft to build our simulations. Comparing the results with the planets cataloged by Kepler, we have characterized the rate of planets per star and how this depends on the size of the planet and the orbital distance. Our new approach allowed the team to explain several effects that had not been included in previous studies.”
Based on the data, they were able to perform a statistical analysis to estimate the rate of terrestrial-sized exoplanets in the habitable zone around Sun-like stars. The results they obtained seem to show that planets very close to the Earth, from three quarters to one and a half times the size of the Earth, with orbital periods ranging from 237 to 500 days exist around one star every six.
“Knowing how often we should expect to find planets of a certain size and a given orbital period is extremely useful for optimizing exoplanet surveys and designing upcoming space missions to maximize their chances of success,” says Ford.
The discovery made by a group of researchers from Japan and the United Kingdom could prove to be an advantage for rice crops. In their study, published in the Journal of Biological Chemistry, genetic techniques are described to counteract Magnaporthe oryzae, a species of fungus that makes the rice plant sick.
This is a very serious disease that can cause serious losses in terms of rice harvesting (it is believed that one-third of the total rice harvest is currently lost due to this fungus). The various strategies used to counteract this parasitic fungus have not proved to be very sustainable in terms of costs and the environment.
Even attempts to produce new varieties of rice to withstand this fungus more have led to undesirable genetic effects. Today’s new genetic modification technologies can be used “to accurately insert genes into rice plants,” a technique that could overcome the problem of binding drag, a phenomenon during which unwanted genes are transferred along with those desired during the production of other plant varieties.
However, we must first identify the most effective genes that increase the immunity of rice to this disease. This is what the researchers did behind this study: according to them a particular immune receptor in rice could improve resistance to fungal disease in rice by triggering important monetary reactions in response to two distinct fungus proteins. The genes encoding this receptor could then be used to design new rice plants that more effectively counteract fungal proteins.