Looking into the Future, one day at a time https://www.eitdigital.eu/newsroom/blog/ EIT Digital Blog by Roberto Saracco 1 en Moving on - From EIT Digital to IEEE FDC https://www.eitdigital.eu/newsroom/blog/article/moving-on-from-eit-digital-to-ieee-fdc/ Sun, 30 Apr 2017 01:31:00 +0200 I started blogging back in 2009 when I was the director of the Telecom Italia Future Centre. Then in 2014 I moved the blog to the EIT Digital (at that time EIT ICT Labs). Now, after having written 1185 posts hosted by EIT Digital I am moving on to...

Moving on ... Image credit: Sumacwasi hummingbird]]>
to blog on the IEEE Future Direction Committee website, and will continue to keep an eye on the activities and results of EIT Digital to which I wish success in its endeavour, because its success is the success for everyone of us.

Moving on ... Image credit: Sumacwasi hummingbird]]>
The thin line separating privacy from social life https://www.eitdigital.eu/newsroom/blog/article/the-thin-line-separating-privacy-from-social-life/ Sat, 29 Apr 2017 01:07:00 +0200 I spent a few days in Rotterdam to see my youngest son, studying there at the university. As I walked around I noticed a car with a strange apparatus on its roof and my son explained that it was checking that parked cars have paid the parking...

Vehicles scanning parked cars to check payment. Plate recognition is made as the vehicle travels in the traffic at normal speed. Credit: ScanGenius]]>
a car with a strange apparatus on its roof and my son explained that it was checking that parked cars have paid the parking fee. 
Indeed, the strange apparatus on its roof is actually a set of digital cameras connected to a computer that in turn is connected to the park meters in the area. When you park your car in Rotterdam you are typing your car identification plate in the park meter. This information is passed on to the computers in these guardian cars. They drive along the streets, at normal traffic speed, and check that you paid your dues. Clearly, such cars can control larger areas much more effectively than wardens walking around. In Singapore they don't need a ScanCar. They are requiring all cars to have a transponder. This allows a central system to identify the position of each car and charge its owner accordingly. Again this is simplifying life to Singapore drivers since they don't need to look for a parking meter. The fee will be charged automatically to a monthly bill (that includes also the driving tax based on how much you drove and where). Singapore is probably the most controlled place on Earth. This provides its citizens with several advantages, an easier life, better services, a safer environment. And a complete loss of privacy. Our social life consists of a set of obligations. In turns these obligations turn co-existence, on average, in a better life. If we know more about you, we can serve you better. At the same time, you may not like to have "the system" to know more about you. Yet, this is the thin line we will have to walk in the coming decades.  On the one hand there is more and more possibility to know everything, and I mean everything, about you, including your heartbeat, and that will enable a host of services to make your life simpler and safer. The internet of things, including sensors in our own body, in our homes, in our car... will just make this even more "feasible" from a technology point of view.  At the same time this knowledge, complemented by the knowledge about everybody else's life will make Society more reactive and useful. But with that privacy goes away. And the question is: how much of our privacy are we willing to trade for a better life and a better environment?

Vehicles scanning parked cars to check payment. Plate recognition is made as the vehicle travels in the traffic at normal speed. Credit: ScanGenius]]>
Leveraging from paper cutting art https://www.eitdigital.eu/newsroom/blog/article/leveraging-from-paper-cutting-art/ Fri, 28 Apr 2017 01:56:00 +0200 In September 2015 I posted a news on work done at the University of Michigan to create solar cells that could track the Sun. The tracking was made possible by the particular structure of the cell surface that got inspiration from Kirigami, the...

A paper-based device inspired by the Chinese and Japanese arts of paper-cutting that can harvest and store energy from body movements. Credit image: American Chemical Society]]>
a news on work done at the University of Michigan to create solar cells that could track the Sun. The tracking was made possible by the particular structure of the cell surface that got inspiration from Kirigami, the ancient art of paper cutting (Origami is the art of paper folding). Now I have stumbled onto another news, coming from China, where researchers at the University of Beijing in cooperation with teams at the National University of Taiwan and Georgia Institute of Technology have leverage on Kirigami to create self charging power units for self powered electronic and medical systems. They used paper as a substrate to create a power unit that can both harvest and store energy leveraging on body movement. The team in China has been working for several years on triboelectric nano-generators (TENGs), micro devices that can transform energy of movement into electricity (similar to piezo-electric devices that transform pressure in electricity). The problem is the low power being generated with this approach. They turned to Kirigami and created a a structure made with rhombic paper cut a few inches in length that gets distorted by movements. These distortions generate electricity thanks to a layer of fluorinated ethylene propylene film overlaid on the paper. Additionally they covered the paper with gold and graphite sand paper to create a super-capacitor that accumulate the electrical energy produced. A single device can accumulate 1mF at 1V in few minutes, that is hundreds of time better than previous devices based on polymers. This quantity of energy is sufficient to operate a remote control, a watch or a temperature sensor. The scavenging of energy (harvesting energy that is around in the ambient, like the one produced by our walking) is getting more and more important as sensors are getting less and less power hungry. The tiny energy that can be harvested through scavenging becomes sufficient to power many types of sensors and in turns this makes their use practical. In the next decade the number of sensors will skyrocket, both on our bodies (in our dresses) and in the ambient, like in urban environment. Finding a solution to power them is essential.

A paper-based device inspired by the Chinese and Japanese arts of paper-cutting that can harvest and store energy from body movements. Credit image: American Chemical Society]]>
An army of robots to sort out packages https://www.eitdigital.eu/newsroom/blog/article/an-army-of-robots-to-sort-out-packages/ Thu, 27 Apr 2017 02:24:00 +0200 Last year, 2016, Hikrobotics, a subsidiary of Hikvision, made the headlines by showing (in action) a robot that can park cars. A car rolls over the robotic platform, the driver gets out of the car and the robotic platform takes over, parking the car...

Hikvision robots in STO sorting centre in Hangzhou, China. Credit: Handout, South China Morning Post]]>
Hikrobotics, a subsidiary of Hikvision, made the headlines by showing (in action) a robot that can park cars. A car rolls over the robotic platform, the driver gets out of the car and the robotic platform takes over, parking the car among other cars with a precision of 5mm. In this way it is possible to park cars minimising the space. Several robots can be at work at the same time coordinated by a central system. Now much smaller robots are at work at STO Express, a Chinese company in Hangzhou, managing 200,000 packages a day at half the cost of a human worker and with no glitches (take a look at the video). The robots (quite similar to the ones used by Amazon in its warehouses) are loaded with a single package or a parcel and move under a bar code reader that wirelessly instruct the robot of the destination of that package. The robot moves along a grid till it gets to the delivery point where it downloads the package. The robots are controlled in their movements by a central system that make sure there is no collision, not an easy feat given the huge numbers of robots active on the grid. Clearly sorting of packages is a job that lends itself quite well to automation, many sorting activities around the world are already automated. In many sectors it is not a matter of using automation to reduce the workforce but as the only way to sustain the volumes involved. Nevertheless, a decrease in workforce is happening and it is not balanced by the increase in workforce needed to manage and maintain the robots (that by the way autonomously look for the recharging station when power runs low). These robots have been designed to operate within a warehouse but I expect, and several trials are under way, to see them (a modified version) to move out of the warehouse and reach (in different forms) the customer home. There are many hurdles to overcome, first of all moving in a not controllable environment, but I can see an evolution where robotic vehicles (autonomous vans) will be loaded automatically at the warehouse by robots, will drive to the vicinity of the delivery point and other robots will take care of the last leg (may be in form of drones). We should expect quite a change in the whole delivery chain in the next decade.

Hikvision robots in STO sorting centre in Hangzhou, China. Credit: Handout, South China Morning Post]]>
Getting a little help ... from AI https://www.eitdigital.eu/newsroom/blog/article/getting-a-little-help-from-ai/ Wed, 26 Apr 2017 02:42:00 +0200 I don't know about dogs or other animals but humans have an uncanny ability to recognise "things" in shapes found everywhere. Ever saw a cloud reminding you of a face, of a horse of a... whatever. Yes I am sure you have. Yet recognising...

This was my drawing. It's a cat, isn't it? ]]>
a service from Google to help us drawing images on our computer, tablet, phone...You draw a sketch of what you have in mind (see the masterpiece I draw, first figure) and Google recognise your chicken scratch and suggests some interpretations. You click on the one you want and ... voilà, there is your nice drawing (see the second figure). The magic is in the capability to interpret a few lines as a representation of an object. This is done using some sort of artificial intelligence. More specifically, Google is saying that they use machine learning, which is a great trick because the application will keep learning, and getting better, the more it is used. Google started releasing some games in November 2016 that actually helped in training the application. This is really smart if you think about. You create a learning machine and let it go free to learn. Potentially all of us become teachers, and the more teachers there are the faster and the better the application becomes. It never tires out of learning (it is not like us that very soon would lose drive and focus). If you try it out I am sure you will love it, but as you get entertained just pause a bit to think at the amazing intelligence that is supporting this apparently simple application. There is nothing so difficult to recognise that I meant to draw a cat, isn't it. And yet it was simply impossible for a computer to do that just few years ago. This is marvellous and scaring at the same time. We are entering a time where computers can be as good as we are in many practical, everyday things ... just way better.

This was my drawing. It's a cat, isn't it? ]]>
A canary in a chip https://www.eitdigital.eu/newsroom/blog/article/a-canary-in-a-chip/ Tue, 25 Apr 2017 02:12:00 +0200 Long time ago, but it was just in the last century, miners used canaries to detect poisonous gases (carbon monoxide) in mines. Now there are much better detection systems (and no canary needs to suffer anymore...) but these are quite bulky and...

A drone carrying the new chip able to detect traces of gases. Notice the separatino between the sensing level and the processing level that amplifies the variations in the sensing level and comnunicate the result to a receiver. Credit: images by Hossain Fahad/UC Berkeley]]>
sensor on a chip that works at ambient temperature and can detect a variety of gaseous molecules with high sensitivity. The chip consists of FET (Field Effect Transistors) in a 3.5 nm layer. These transistors have very thin junctions covered by noble metal ions. These latter respond to the presence of gaseous molecules of a certain type by changing the resistance of the silicon layer (the FET) below that in turns changes its amplification resulting in a signal that is specific for that gas molecule. By decoupling the amplification from the sensing the researchers have been able to keep power consumption very low and by changing the composition of the sensing layer they can create sensors that can detect,at the same time with a single chip, a variety of molecules. The chip is not ready for sale, quite a bit of engineering is still needed, in particular to ensure stability. You don't want to get false signals, do you? Eventually, these chip may find a place in our smartphones, providing early detection and generating alarm. You don't have to go down the shaft of mines to be exposed to dangerous gases, sometimes the carbon monoxide is produced by a stove in the kitchen...

A drone carrying the new chip able to detect traces of gases. Notice the separatino between the sensing level and the processing level that amplifies the variations in the sensing level and comnunicate the result to a receiver. Credit: images by Hossain Fahad/UC Berkeley]]>
Marrying Memory with Storage https://www.eitdigital.eu/newsroom/blog/article/marrying-memory-with-storage/ Mon, 24 Apr 2017 01:58:00 +0200 If you think that Memory and Storage is one and the same, think again! Storage is for keeping our data safe for quite a long time with very little cost associated to it (which makes sense since we are talking about long period of time) in...

Intel’s Optane storage device is a fast solid-state drive. Next, the technology will be used to create memory modules. Credit: Intel]]>
Optane, that is leveraging on the best of both worlds, providing a chip that is as fast as a memory chip and as power savvy as a storage chip. In March Intel has announced they will start to sell these new chips to data centres. They are non volatile like flash memory (hence they do not need power to keep the data stored) and are fast as DRAM. Unfortunately they don't come cheap: a 375GB drive cost 1,520$ three times as much as a SSD and 30 times as much as a magnetic hard drive. However, we know the trend: they will get cheaper and cheaper. Today they are targeted to data centres where speed is important and the cost of energy is huge. Tomorrow, they may reach a price point where they will be affordable in other applications and eventually they will find the way to our homes. If Optane will deliver as promised it will indeed be the holy grail of storage, marrying memory and storage in a single chip. So if you thought memory and storage were one and the same just stick to it a few years more and you'll be correct!

Intel’s Optane storage device is a fast solid-state drive. Next, the technology will be used to create memory modules. Credit: Intel]]>
Transparent biosensors https://www.eitdigital.eu/newsroom/blog/article/transparent-biosensors/ Sun, 23 Apr 2017 02:31:00 +0200 In Italian there is a saying: "the eyes are the mirror of the soul". And it is not just in Italian. “The eyes are the mirror of the soul and reflect everything that seems to be hidden; and like a mirror, they also reflect the person...

Transparent biosensors in contact lenses (made visible in this artist’s rendition) could soon help track our health. Credit: Jack Forkey/Oregon State University]]>
The eyes are the mirror of the soul and reflect everything that seems to be hidden; and like a mirror, they also reflect the person looking into them.” (quote Paulo Coelho, Manuscript found in Accra). Literature aside, our eyes can indeed reveal quite a bit about our health. The retinal blood vessels provide tell tale signs of a variety of diseases. Tears carry with them molecules that mirror the ones in the blood, like glucose. Researchers at the Oregon State University have worked on creating contact lenses with sensors that can detect the presence of glucose in the tears. The problem, of course is that these sensors have to be transparent to light. It turns out that the material used in displays, indium gallium zinc oxide (IGZO), that is transparent, can also be used to create sensors. The sensor is made by placing an enzyme, glucose oxidase, in contact with a field effect transistor made with IGZO. When glucose enters in contact with the enzyme the ph of the solution changes and this affect the conductivity of the transistor providing a sign of the presence of glucose. Note that the concentration of glucose in tears is way lower than the one in blood, so the sensors in the contact lenses need to be much more sensitive than the ones used to detect glucose in blood. The sensor developed indeed has been proven to work nicely. Interestingly, the way the sensor is made opens the door to sensing many other types of molecule, one needs to couple the transistor with the appropriate enzyme. The team has already experimented with different types of sensors, like one placed on a catheter to detect uric acid.   According to the researchers it would be possible to include 2,000 different sensors in a patch of just one square mm on a contact lens, transforming it into a full lab on our eyes!  However, it will take over one year to have a safe contact lenses prototype that can be tested on animals, and several more before we car wear those contact lenses.

Transparent biosensors in contact lenses (made visible in this artist’s rendition) could soon help track our health. Credit: Jack Forkey/Oregon State University]]>
Excitement from Neural Laces https://www.eitdigital.eu/newsroom/blog/article/excitement-from-neural-laces/ Sat, 22 Apr 2017 01:15:00 +0200 The recent launch of Neuralink by Elon Musk, a company whose business is connecting brains to computers, has created quite an excitement in the scientific community, not because scientists feel that they will succeed anytime soon, the problems that...

A stentrode can be injected in an artery and pushed to find home in a capillary of the brain. Credit: Melbourne University]]>
recent launch of Neuralink by Elon Musk, a company whose business is connecting brains to computers, has created quite an excitement in the scientific community, not because scientists feel that they will succeed anytime soon, the problems that remain to be solved are huge, but because Elon Musk has a knack for creating innovative, and successful, business. His strive in this area is a signal that technology is reaching a maturity point, it is no longer a lab affair. Elon Musk, in a conference speech in 2016, spoke about the possibility of creating an interface made by Neural Laces, overlaid on the brain cortex but it did not provide any details on how to do it nor on what a Neural Lace might be. An interesting article on IEEE Spectrum is making a point on the most advanced technologies to interface with a brain. I already blogged on a few of them in the past.  The basic problem is how to be able to capture the activity of the brain with accuracy and with minimal noise (generated by activities in nearby areas...) and remain as less invasive as possible from the point of view of the brain. Work is progressing in finding materials that are not perceived as foreign bodies by the brain, that can gently move as the brain moves and yet keep their position. Notice that even if this is achieved the brain is a dynamical structure, continuously reshaping itself. Hence, if you have a probe in just the right place to capture the activity of that precise neural circuit related to -say- the analyses of depth in an image, in a while that activity may shift to a different part of the brain and your probe is no longer useful.  The more you are interested in high level activity, thinking, the more dynamical structures are. On the contrary, if you are interested in some motor action you can pinpoint places that are responsible for instructing muscles to move and these are much more stable. Anyhow, research has resulted in several ways to create neural interfaces.  One that I found really interesting is the use of stentrode, a probe invented at the University of Melbourne, funded by DARPA (watch the clip).  The stentrode is a ... stent. It can be inserted in an artery and pushed to a specific location, a capillary in the brain. Being a stent it does not block blood circulation, quite the contrary, it slightly widens the capillary to let the blood flow through. It can detect electrical activity in the surrounding and can transmit the data gathered to an external monitoring device. In a way it is much less invasive than a normal probe in the brain since it does not touch the neurones nor cause any damage in being positioned, but is as effective as a needle like probe in terms of detecting electrical activities is nearby neurones. The way is still winding and there is not arrival point in sight but clearly several, important, steps have been made and the commitment of Musk to this area will likely bring an acceleration.

A stentrode can be injected in an artery and pushed to find home in a capillary of the brain. Credit: Melbourne University]]>
Is virtual taking us a bit too far? Virtual lemonade https://www.eitdigital.eu/newsroom/blog/article/is-virtual-taking-us-a-bit-too-far-virtual-lemonade/ Fri, 21 Apr 2017 02:17:00 +0200 Our capability of transforming atoms into bits has progressed tremendously, we have been able to digitalise almost everything, including emotions. There are applications able to "look" at a face and tell if that person is happy or sad and...

The pH sensor on the left is used to capture the flavor of a drink and transmit it to the tumbler on the right. Credit: NSU]]>
applications able to "look" at a face and tell if that person is happy or sad and quantify, in bits, the level of happiness.  Much more difficult is the transformation of bits into atoms. Of course, from the very beginning of the digitalisation era we have had bits transformed into photons (in video screens), into sound waves (loudspeakers) and in ink on paper (printers). But these transformations, probably because they are so natural, are not really perceived as transforming bits into atoms. In the last ten years the development of 3D printers has been impressive, and 3D printing is clearly the epitome of transformation of bits into atoms. They will progress even further and this may lead to a disruption in the way (some) products are designed, made and retailed. There is another way of transforming bits into atoms, that is transforming bits into "virtual atoms", atoms that are not real but "feel" real. This goes though cheating our senses. Haptic interfaces cheat our sense of touch, leading us into feeling something that is not there, giving the feeling of texture, softness, even providing the sensation of wetness; holographic interfaces cheat our sense of vision, sound surround cheats our sense of hearing. Cheating our sense of smell and taste has proven more elusive. In 2014 a UK inventor, Charlie Harry Francis, launched the Edible Mist Machine, a device that was able to re-create 200 flavours mimicking the ones of bacon, cheese... It was not particularly successful, yet it was an attempt to cheat our sense of smell.  There have been several other attempts (even before 2014) to re-create smell sensations but so far there have been no killer application in this area. At NSU, the National Singapore University a team of researchers are studying ways to cheat our sense of taste by stimulating the tongue with electrical pulses that in turns activate the taste buds detecting sour, salty, sweet, umami and bitter.  Recently they performed an experiment to serve virtual lemonade. The goal was to cheat the taste buds of people drinking water so that they would think they were drinking a lemonade.  The experiment took into account also the visual perception of a lemonade glass (and of the one pretending to be). In fact, the visual aspect plays a significant role in our perception of taste (think about a dish being served in a top notch restaurant where the visual impact is high). To assess the effectiveness of tricking the taste buds of tested subjects the experimenters used a special glass with LED on the bottom to colour the water and provide a visual feeling of lemonade. It worked. Actually, by changing the colour, hence the visual perception, they changed the "taste" of the lemonade (in terms of perception, of course, but that is what really matters!). In the case of the lemonade the most important "taste" is the sour one. Researchers are working to improve their wares with the goal of teleporting taste feeling over the web. 
Not sure if there is a business for this. As they say, there is nothing like the real thing....

The pH sensor on the left is used to capture the flavor of a drink and transmit it to the tumbler on the right. Credit: NSU]]>