Smart prosthetic hand combines both human and robot control

Zhuang et al. Shared human–robot proportional control of a dexterous myoelectric prosthesis. Nature Machine Intelligence, September 2019.

Holding an object in your hand might seem easy, but it’s actually a very complicated and challenging task; if, for instance, an object starts to slip, you typically have a couple of milliseconds to react. Scientists have been trying new approaches for improved control of robotic hands, particularly for use by amputees. A recent technology was able to combine individual finger control and automation for improved grasping and manipulation by successfully merging the fields of neuroengineering and robotics. This interdisciplinary approach was tested on three amputees and seven non-amputee subjects. The neuroengineers achieved the intended finger movement from muscular activity on the amputee’s stump, allowing for individual finger control of a prosthetic hand, which had never been done before. The robotics team enabled the robotic hand to take hold of objects and maintain contact with them for robust grasping. The amputee first performed a series of hand movements in order to train the algorithm through a machine learning paradigm. This taught the algorithm to decode user intention and translate it into finger movements of the prosthetic hand. Concurrently, sensors placed on the amputee’s stump detected muscular activity, which trained the algorithm to learn which hand movements corresponded to which patterns of muscular activity. Once the user’s intended finger movements were acquired, this information could then be used to control individual fingers on the prosthetic hand. When the user tried to grasp an object, the robotic automation initiated. The algorithm told the prosthetic hand to close its fingers when an object was in contact with sensors on the hand’s surface. This automatic grasping was designed to infer the shape of objects and grasp them based on tactile information alone, without any help of visual signals. The robotic hand has the ability to react within 400 milliseconds, and it is equipped with pressure sensors all along the fingers: it can react and stabilize the object before the brain can actually perceive that the object is slipping. While this promising technology can be used in in several neuro-prosthetic applications such as bionic hand prostheses and brain-to-machine interfaces, there are still many challenges remaining to implement this technology in a commercially available prosthetic hand for amputees. It is currently being tested and improved.

Cats securely bond with people, too

Vitale et al. Attachment bonds between domestic cats and humans. Current Biology, September 2019.

Dogs have long been regarded as man’s best friend. They’re sociable, faithful, and obedient. Cats, on the other hand, are often described as more aloof, mysterious, and independent. But a new study suggests that cats actually bond with their owners in similar ways to how humans and dogs bond with companions. The most established way to study human attachment behavior is to observe an infant’s response to a reunion with their caregiver following a brief absence in a novel environment. When a caregiver returns, secure infants quickly return to relaxed exploration while insecure individuals engage in excessive clinging or avoidance behavior. These tests had been previously run with humans, primates, and dogs; researchers decided to run the same test with cats. 79 kittens and 38 adult cats and their caregivers were recruited. During the test, an adult cat or kitten spent two minutes in a novel room with their caregiver followed by two minutes alone. Then, they had a two-minute reunion. The cats’ responses to seeing their owners again were classified into attachment styles. The results show that cats bond in a way that’s surprisingly similar to infants. In humans, 65% of infants are securely attached to their caregiver and domestic cats and kittens mirrored this, as about 65% of them securely bonded to their people. After the first round of tests, the researchers enrolled half the kittens used in the study in a training and socialization course. The other half served as a control group. Researchers then found the same results, suggesting the training did not have an effect on kittens’ attachment behavior toward their owners. This indicates that once a cat forms a bond, it seems to remain stable over time. This social flexibility may have helped facilitate the success of the species in human homes. It is still not clear what the factors are that shape the caretaker relationship, but it’s likely a miraculous complex mix of genetics, personality, and experience.

An efficient and green way to convert heat into electricity

Zheng et al. Paramagnon drag in high thermoelectric figure of merit Li-doped MnTe. Science Advances, September 2019.

A recent discovery could help scientists to develop more efficient ways to generate electricity from heat that would have been otherwise wasted, such as heat coming from car exhaust, industrial processes, and interplanetary space probes. In principle, magnetic fields can be used to generate electricity. If we move a magnet through a coil or wire, the magnet pushes and pulls electrons that create an electrical current. Magnets themselves don’t have energy, but they can control energy currents through the created magnetic field. The main problem with magnets is that when a magnet is heated up, it loses most of its magnetic properties and becomes a so-called paramagnet. Until this discovery, scientists believed that paramagnets couldn’t be used for generating electricity. In the new study, researchers found a way of designing thermoelectric semiconductors that can convert heat to electricity. The tiny particles in paramagnets, so called paramagnons, ended up producing enough spin to push an electron, for only a billionth of a millionth of a second – apparently long enough to make paramagnets viable energy-harvesters. This breakthrough in the conventional understanding of magnetic properties could lead to more research into how magnets and energy interact to potentially facilitate electricity production from heat that is otherwise wasted and oftentimes harmful to the environment.

How about quantum computers? Quantum computers make use of a quantum mechanical phenomenon, so-called quantum superposition (being in different states simultaneously). Classically, voltage across a circuit element can be either positive or negative when measured by a voltmeter, but not simultaneously negative and positive. Could it somehow be possible to be in both states simultaneously?

Quantum superposition

For electrical circuits, the answer is obviously no. In microscopic scales of single atoms, or photons (i.e., single quantized packets that constitutes the light beam), however, the answer is yes. Consider an optical component, for instance, that splits an incoming light beam into two beams of equal intensity. In optics, such a device is called a 50/50-beam splitter. You can ask what happens when a single photon is sent to such a beam splitter. Since a single photon cannot be split in this simple experiment, you might expect that it would either be transmitted or reflected with equal probability . Experiments, however, show that this is not actually true in the single photon level. The single photon is indeed simultaneously reflected and transmitted.

Once microscopic quantum superposition is brought into our macroscopic world, we can imagine many interesting phenomena. Simultaneously occupying many different places and being dead and alive at the same time are only two of them. Of course such technology, especially applied to humans is highly science fiction, given current experimental and theoretical challenges. Nevertheless, quantum superposition has a strikingly interesting similarity with the spiritual states already achievable by saints, such that they can be available in more than one place at a given time or become dead and alive, in the sense that they live both in the future and in the past.

It is not known exactly why quantum superposition exists, but what we know is that it is a necessary ingredient for our complex universe to perform its vital functions in a finite amount of time. Quantum superposition principle reflects the great wisdom and power of the Omnipotent. Similar to the single photon example above, with this principle, God gives the underlying particles of the universe an immense power to achieve many tasks simultaneously. Otherwise, regarding the finite age of the universe (about 15 billion years), our physical universe and the events taking place all around us would not come into existence. The Quantum superposition principle has also inspired researchers to build unprecedentedly fast computers to solve the problems that are intractable with any classical computing method. In this article, we introduce this new strategy to computing.

Quantum computing with superposition

Having provided some background about the quantum superposition, we ask the question “How could we exploit quantum superposition for fast computing?” Below we will give a glimpse of that power. Consider a three-bit register. It can only store one out of eight numbers in the set, {0, 1, 2, 3, 4, 5, 6, 7}, in a given moment of time. For example, number 5 is stored in a three-bit register as “101.” Now suppose that these three bits are replaced by their quantum cousins, so-called qubits (short for quantum bit). You can imagine, for example, a quantum register consisting of three rubidium (Rb) atoms. These individual atoms can be prepared in the 0 and 1 logical states simultaneously by shining a laser beam for a certain amount of time. Then it is possible for three atoms combined to be prepared in a superposition of eight numbers, which is impossible classically. In other words all those eight guys physically exist in the same room, although it doesn’t allow more than one guy to fit classically. If we want to make operations on all of these numbers, we don’t need to perform serially; instead, we can achieve that in only one computational step on a single hardware. Thus, quantum superposition leads to a massive parallelism, which renders the computational complexity (i.e., a measure of how efficiently a given problem could be solved) highly reduced for various difficult problems in computer science.

For example, let’s consider RSA, a well-known algorithm (i.e., a set of instructions to solve a problem on a computer) for secure communication that was invented by Rivest, Shamir, and Adelman, hence the name, in 1977 at MIT. It is widely used in electronic commerce protocols. The details of RSA are out of scope in this article (See the FAQ section of the RSA Laboratories’ web site in Ref. [1] for a brief introduction to RSA). Here, we only want to mention its vulnerability to quantum computers if they were to exist. The security of the RSA cryptosystem relies on the difficulty of factoring large numbers, which is intractable with classical computers. Factorization for small numbers, say 15, is quite simple. When the number of digits increase up to a few 100s, for example, enormous computational resource is required. RSA Laboratories publish the RSA challenge numbers (see Ref. [2] for the list of challenge numbers and the prize) on their web site to test the security of their algorithm at various key lengths. The largest integer, RSA-640, which has 193 decimal digits (640 bits), was factorized recently by F. Bahr, et al. The next challenge number in turn is RSA-704, and the prize is $30,000. Imagine factorizing a 1000-digit number. You would probably be a considerably rich person in just a few minutes, if you had a moderate quantum computer and the RSA Laboratories kept feeding you with new challenge numbers, because the factorization of such a large number with current computational resources takes forever, perhaps even more than the estimated age of the universe. Of course, the RSA Laboratories will not let you be very rich, by simply quitting posting new challenge numbers. They would be interested in your quantum computer, though.

How does the quantum computer crack the world’s most secure cryptosystems with little effort? One can construct new algorithms for quantum computers based on above described principle of superposition. These algorithms can take the outcome of previous calculations and input them as a superposition to the next stage of the instructions, which results in a highly efficient form of computing (please consult Ref [3] to have for a simple explanation of quantum superposition for fast computation). In 1994, Peter Shor from AT&T’s Bell Labs in New Jersey just did that. He developed the world’s first quantum algorithm, which efficiently performs factorization. In 1996, Lov Grover also at Bell Labs invented the unstructured database (i.e., a disordered list such as a list of city names not in alphabetical order) search algorithm for quantum computers, so-called Grover’s algorithm.

Suppose that there is a basket with ten balls in it. You are now asked to find a specific one with your eyes closed, say red. It is known, however, beforehand that there is only one red ball in the basket. All you, or your smart digital friend, can do is just pick one randomly and see if it is red. If you are lucky enough, the first ball you pick might be red. In the worst case, however, you will be successful at your last choice. So, classically you have to repeat the process on average at half times the number of balls. If you made a quantum friend rather than classical, however, your life would be smoother. You would be able to find and manage your stuff easily, no matter how messy you are. Quantum computers speed up such unsorted database searches quadratically. You can find, say your favorite socks, in a number of trials that is about the square root of the total number of your stuff. You may think that you don’t have that much stuff. But consider identifying a specific element in a considerably large pool of unsorted data. As the number of elements in the set increases, it quickly becomes intractable to find what exactly you are looking for. In that case the significance of quadratic boost cannot be denied.

Motivated by the above mentioned factorization and unsorted database search algorithms, the power of quantum computing has inspired great attention, since their invention, among many disciplines including physicists, computer scientists, mathematicians, engineers, and material scientists.

Quantum computer today

Despite promising developments in theory, progress in the physical realization of quantum circuits, algorithms, and communication systems have been extremely challenging to date. There are many approaches for quantum information processing. Major model physical systems include nuclear spins, ions, neutral atoms, solid state nanostructures, superconductors, and optical circuits. In optics, for example, the qubit can be represented by the polarization (i.e., direction of oscillation of electric field) state of a single photon. So that the instructions described by the algorithm could be implemented by manipulating the polarization states of single photons. Unfortunately, all the models for quantum computing have their own drawbacks besides their advantages.

Given the trends, nobody knows whether or not a sufficiently scalable (i.e., large enough to harvest its potential power) quantum computer would be available in the decades to come. Nonetheless, D-Wave Systems, Inc., The Quantum Computing Company, was eager enough to unveil the “world’s first commercially viable quantum computer” (see Figure 1, and Ref [4] for the story.). D-Waves’ 16-qubit quantum computer makes use of superconducting element niobium, which operates at an extremely low temperature. It can search for molecular structures that match a target molecule, create a complicated seating plan, and fill in Sudoku puzzles. Although the device is very slow compared to an inexpensive PC, D-Wave intends to develop a 1000-qubit quantum computer.* The goal is to scale the quantum computer to about 10 thousand qubits to solve the most challenging problems outright, which are simply intractable with classical computers. The researchers, however, are not very optimistic. Prof. Lloyd of Massachusetts of Institute of Technology, a pioneering scientist in superconducting approach for quantum computing that underlies the D-Wave’s quantum computer, says “It’s too good to be true.”

Once quantum computers of reasonable power are built, the world will be unimaginably exciting and perhaps scary too. When the first commercial computer, Universal Atomic Computer I (UNIVAC I) (see Figure 2), was shipped to the United States Air Force in 1952, nobody was indeed aware of what this fat guy would lead to in our social, economical, political, and psychological life. Its descendants, however, are now inevitable parts of our lives. They are helping us in many aspects of daily life. Controlling machines, sending electronic mail, scheduling our plane tickets, communicating with our best friends, playing games, making our payments are only some of them.

In this article we only sketched the quantum superposition principle as an important ingredient for quantum computation. This is certainly not the whole story. “Entanglement” [6], for example, is another complementary resource for quantum computing and communications, as well as quantum mechanics to test its foundations.

Contrary to its classical counterparts, the power of quantum computers indeed comes directly from our granted capability of tailoring and mimicking the amazing design hidden in the microscopic world of atoms, photons or other quantum particles. Quantum computers sooner or later will bring the most science-fiction into reality. They will play a significant role especially in the development of ultra-intelligent machines and robots superior to classical ones, and communication systems whose ultimate security is guarantied by the nature’s architecture which was lay down by God. Quantum computers will reveal to us the deepest secrets of our Creator embedded in our universe, which cannot be explored using conventional computers. That day, the future will only be lacked by our limited imagination.

Acknowledgment

This article was produced in MERGEOUS [7], an online article and project development service for authors and publishers dedicated to the advancement of technologies in the merging realm of science and religion.

Omer D. Ikramoglu is a freelance writer in optics and quantum physics.

References

1. RSA Laboratories, http://www.rsa.com/rsalabs/

2. RSA Challenge Numbers, http://www.rsa.com/rsalabs/node.asp?id=2093

3. A short introduction to quantum computation by A. Barenco, A.Ekert, A. Sanpera and C.Machiavello from La Recherche, November 1996. http://cam.qubit.org/articles/intros/comp.php

4. J. R. Minkel, “First “Commercial” Quantum Computer Solves Sudoku Puzzles”, Scientific American, Feb 13 (2007).

5. UNIVAC I, http://en.wikipedia.org/wiki/UNIVAC_I

6. S. Candaroglu, “Quantum Entanglement: Illusion or Reality?” Fountain, Issue 61 (January-February 2008).

7. http://www.mergeous.com/

* At the time of writing D-Wave Systems had only 16-qubit quantum chip and they were intending to develop a 1000-qubit quantum computer by the end of 2008. Although they couldn’t meet their goal, they now have a design of a 128-qubit most powerful ever quantum chip which awaits the tests (see http://www.dwavesys.com for up to date information).

As is the case in many scientific arenas, the first studies in the field of algorithms were conducted by Muslim scientists. The concept of algorithm was used for the first time by the worldwide renowned Muslim scientist Al-Khwarizmi, who lived between 780–850 AD in Baghdad. Al Khwarizmi put his name to historic studies in the fields of mathematics, astronomy, and geography. His work entitled Hisāb al-jabr wa-l-muqābala (Calculation by Completion and Balancing) constitutes the first collection of algorithms. The Latin translation attracted great attention in Europe. The Europeans used the term algoritmi, rendered from “Khwarizmi,” to refer to rules for solving arithmetical problems by using Arabic numerals.

Having first developed as a branch of mathematics, algorithms have been described as a sequence of instructions that is used to solve a certain problem. These finite number of steps,begin at a certain point and come to an end with a result. But, today when we talk about algorithms, what comes to mind are the methods pursued in the operations of sequencing in computer programming; though algorithms are used in many fields, including physics, chemistry, biology, and music.

When different algorithms are used for solving the same problem, this enables us to reach the same solution through different means. There are, for instance, a great number of sorting algorithms (selection sorting, merge sorting, quick sorting) that have been developed for the purpose of incrementally sorting a given set of numbers. The same also applies to the search algorithms which find a given number in a set of numbers. What does this imply for us? Different people can attempt to solve the same problem with different algorithms or approaches that are better, truer or more “beautiful”. Thus, it is always important not to reject an idea without a prior examination or understanding, to welcome proposals of different opinions and to try to benefit from better alternatives or solutions. Thus, we can understand from an algorithmic approach that establishing dialogue and showing mutual respect are essential moral virtues in transforming differences into wealth.

Computer algorithms solve certain problems according to pre-defined parameters. Thus, we should never forget that computers cannot carry out a spontaneous operation, but can only carry out functions that have previously been programmed into them by human beings. Although some researchers, by relying too much on recent developments in computer sciences, and particularly in fields like artificial intelligence, attribute human characteristics to computers. They even estimate that they can produce human-like things in the near future, which seems rather unfeasible when we consider countless physical and spiritual features of humans. What does this imply for us? Pursuing a systematic method when solving problems in both computer sciences and daily life would considerably increase one’s success rate in solving any problems encountered. The most critical stages in problem solving are obviously a correct diagnosis and full description of the problem. The algorithms are developed and the most coherent and suitable are then chosen.

After an algorithm is designed in a flowchart, it is translated into software and tested against verifiable data prior to its usage in real operations. This test is necessary to prevent probable flaws in the software. What does this imply for us? By utilizing all the talents and especially the intelligence granted to us, human beings are always able to find what is better or truer. Research has been conducted in scientific studies by keeping in mind that there is always a next step in any development that has been achieved; the experience of the history of mankind demonstrates that a method that has been developed today might prove to be inadequate tomorrow.

When analyzing a number of different algorithms that are used to solve the same problem, the cost is assessed, with the least costly and most suitable one being selected. The cost of an algorithm is determined according to the number of operations carried out in solving any problem. An algorithm which solves the same problem with a greater number of steps has a lesser efficiency and a lesser performance.

Sorting algorithm by insertion

Sorting algorithms are one of the most frequently used algorithms in computers. As sorted data is more easily processed and used, data are usually sorted first with a sorting algorithm before it is processed within more costly operations.

In insertion sort (i.e., sorting by insertion), numbers to be sorted are inserted into a new “sorted” set. The new sorted set is empty at the beginning, and the numbers are inserted one by one into the right position of this set. During each insertion, the number at hand is compared to the numbers of the new set from the smallest to the largest. During a comparison, if the term (number) at hand is smaller than the term in the new sorted set, then the term is inserted right before the term in the sorted set. As each term is generally compared with all the terms which precede it, sorting of n numbers is done by about n2 comparisons. Thus, the cost of such a sorting algorithm is O(n2).

The quick-sort algorithm

This algorithm is used to find the shortest way between two points. Let us suppose that we have a limited number of points, some of which are inter-related. During this type of calculation, the shortest way is systematically found by checking all the combinations that have been able to be established among given points. The web sources that are provided for planning routes and mapping services with PNDs (Portable Navigation Devices) for travelers can be given as examples; here the minutest details of every highway and road have been recorded to the virtual media. In accordance with certain criteria, by taking into consideration all the probable routes between the two addresses provided by the user, the optimum driving route, including road and street names, is suggested.

Use of algorithm in medicine

Any research or inquiry used in the diagnosis and treatment/curing of a disease is defined as a medical algorithm. The logic of a decision tree is used in diagnosing, curing and following up diseases with these algorithms. Guiding algorithms, for example, “if symptoms A, B, C are observed in patient, then, the patient is probably suffering from the disease D, so, use the treatment E” are used in medical expert systems. The purpose of such algorithms is to standardize the medical services provided and thus minimize potential uncertainties and errors that likely to arise. Moreover, such algorithms are also used for educational and training purposes and as a guide to approaching the patients, diagnosing and curing diseases and towards solving their problems with greater ease. A great number of medical information that has been published has been transformed into numerous algorithms, ranging from those that use simple calculations to those that make highly complicated decisions. For instance, in the algorithm that was developed to measure the Body Mass Index (BMI), age, sex, height and weight are among the questions that are asked of the patient. The data obtained from the patient is later applied to a formula developed from the experiences of medical science, and the said index can thus be calculated. However, doctors should compare the results obtained from such algorithms with their existing knowledge, for even such simple algorithms as BMI become undependable in different cases, for example an athlete or an expectant mother. Algorithms only provide guidance for physicians, but the final decision should be taken by the physician after having individually assessed the status of each and every patient.

Algorithms and music

The algorithmic composition has been used in music for centuries to produce music with a methodological approach. Canon, for example, is a form of music produced with an algorithm in which a melody is replicated by one or more imitations of it played after a given duration.

Human brain, nature, and algorithms

Although the secrets of brain have not yet been fully explained, vital information has been compiled about its biological structure and functioning, thus proving that different sections of the brain fulfill different roles. Such sophisticated tasks as controlling the bodily organs, mental discernment, functions corresponding to the use of tongue and other sensory activities, the perception of colors, calculation of actions and the perception of physical conditions, are all carried out by use of brain mechanisms and faculties. All these tasks are accomplished by algorithmic processes which have not yet been fully explained. The issue of the unification of the data or images received by the eyes is, for instance, one of the matters that scientists do not yet fully understand. Although many have tried to imitate the human brain, such simple cerebral functions as walking and communicating have, to date, not been convincingly imitated.

What algorithms imply for us

As science and technology develop more and more, new inventions are being made, algorithms are being proposed for unsolved problems, while algorithms of already solved problems are being further developed. For example, many ciphers, which in the 1970s were believed to be unsolvable within an acceptable span of time, are being easily solved today thanks to the algorithms that have been developed and the ever-increasing operational power of computers. Consequently, developments in algorithms suggest to us that we should always search for that which is better and more effective; also we should remember that there are a number of different ways that lead to the very same target. Thus, we should always be open to the ideas of others and willing to consult with them before making a decision about any subject.

Ahmet Isik has a PhD in mathematics and is a freelance writer.

References

• Cormen, T. H., C. E. Leiserson, R. L. Rivest, Clifford Stein. Introduction to Algorithms, MIT Press, 24th Edition, 2000.
• http://www.nist.gov/dads/termsType.html#P