Add idea


«    May 2019    »










DateDate: 26-05-2019, 07:27
The robot was presented at the International Conference on Robotics and Automation (ICRA 2019) in Montreal (Canada).
Created by scientists at the Genoa Research Center (Istituto Italiano di Tecnologia, IIT), HyQReal is the successor to HyQ, a smaller model developed several years ago.
According to the publication TechCrunch, the robot has a length of 1.33 m and a height of 90 cm. Its weight is 130 kg, including a 15-pound battery that provides up to 2 hours of battery life. It is resistant to dust and water and can rise if it suddenly falls or falls over.
At the ends of the legs of the robot is a rubber "sole" that provides traction to the surface.
The video released by the developers showed that the HyQReal robot dragged the Piaggio P180 Avanti aircraft weighing 3 tons over a distance of more than 10 meters. The test took place at Genoa International Airport. The length of this aircraft is 14.4 meters, and its wingspan is 14 meters.
According to the developers, the robot can be useful for farmers and rescuers. Also, the device can be useful when conducting various inspections. At the moment, HyQReal is a research prototype.

DateDate: 25-05-2019, 06:57

Fantastic sound pressure exceeded 270 decibels
Scientists have created the loudest sound: this is the first time in history.
Scientists from the United States, led by Gabriel Blazh, a researcher at the SLAC National Laboratory and Stanford University, have created perhaps the loudest underwater sound in the world.
To do this, experts used the SLAC Linac Coherent Light Source (LCLS) X-ray laser, thus breaking up tiny streams of water to create a fantastic sound pressure above 270 decibels, writes ukrainianwall.
This laser is capable of creating molecular black holes and heating water to 100,000 degrees Celsius in less than one millionth of a microsecond.
According to the researchers, at zero decibels there is no pressure wave, but at the other end the medium through which the sound passes begins to collapse, so it cannot become louder.
This is what happened when the researchers hit a micro-jet of water (with a diameter of 14 to 30 micrometers) with an X-ray laser. When short X-ray pulses fell into the water, it evaporated and created a shock wave.
After this shock wave passed through the jet and formed itself in a “shock wave” consisting of alternating high and low pressure zones. In other words, very loud underwater sound.
Thanks to the experiment, researchers can study in more detail how shock-wave mechanisms work. You can also find ways to protect miniature samples that undergo atomic analysis inside water jets from damage, which could be of great help in the development of more effective drugs and materials.

DateDate: 24-05-2019, 05:55

New Zealand engineers worked with a door mat. Now he can distinguish people by footprints, given the shape of their soles and the weight distribution of the shoe holder.
Smart door mat will remember the trail of the owner
The device creates a scan of a person's shoe when he comes to the door. The proposed development can be one of the stages of fast multifactor homeowner authentication in the future, which will eliminate the use of additional USB tokens, putting fingers on appropriate scanners, etc.
Suranga Nanayakkar and his colleagues from the University of Auckland presented a novelty at the conference AH2019. Inside the doormat there is a 15.3-inch capacitive sensor. He is able to detect an object above the carpet at a height of 4 centimeters. This property allows you to scan the shoes of the person who came to the door.
Unfortunately, the technique proposed by the craftsmen is not perfect and is not 100% accurate. As it turned out, on the second day of development testing, the recognition accuracy dropped from one hundred to ninety-six percent when analyzing the soles of 15 volunteers. But engineers are not losing optimism, they are confident that in the future they will be able to bring the work of the smart carpet to perfection.

DateDate: 23-05-2019, 05:58
Geksapod began its existence without having the slightest understanding of his environment. Thanks to the recursive function developed for self-improvement, the robot can gather information about its environment in order to fine-tune the behavior. The more experience a robot gains, the better it works.
At first glance, nothing complicated, but it is worth considering the fact that the robot must be able to determine not only its location and orientation in space, but also to “feel” the impulses of the sensors located on the machine’s lap.
By optimizing the behavior of the robot and concentrating on the task, scientists from Facebook managed to teach the robot to “walk” in just a couple of hours, not days.
It would seem that further on the hexapod should go and learn everything new. But it’s not so easy to evoke the passion of traveling with robots. However, this is exactly what the Facebook team intends to do in the next research phase.

DateDate: 22-05-2019, 05:57

Forecasts come true with an accuracy of 75% and 92%
The machine learning system has studied thousands of recent solar flares and determined that their occurrence could have been predicted several days before emissions from a sharp increase in the strength of electric currents in the future zone of their origin. This was told by Mark Chung from Stanford University, PNAS reports.
American scientists have suggested that all flares are generated by similar processes in the depths of the sun.
For artificial intelligence training, specialists manually analyzed more than 1000 photos of active regions on the surface of the star that caused flashes in 2010–2016 and were studied using the GOES meteorological probe tools.
The machine learning system rather quickly learned to recognize the “quiet” regions of the Sun and the future centers of flare activity with an accuracy of 75% and 92%.
At the same time, differences between them occurred and persisted for 3 days before and after the birth of the release, which indicates an extremely “long-playing” character of such activity zones.
In their predictions, AI relied on how the electrical activity of the solar photosphere changed in those zones where flashes appeared.
The subsequent analysis of data from solar satellites will help scientists understand what exactly forces solar matter to “eject” from the surface of the star and how such cataclysms can be predicted with an accuracy of 100%.