From The Gift Guide: 8 Perfect Gifts To Learn With

If your goal is to give the gift of knowledge this year, we’ve got the perfect selection for you. 

Read more on MAKE

The post From The Gift Guide: 8 Perfect Gifts To Learn With appeared first on Make: DIY Projects and Ideas for Makers.

Searching for the Perfect Neuron for AI

Researchers tried out several new devices to get closer to the ideal need for deep learning and neuromorphic computing

Disrupting Reproduction: Two New Advances in Tech-Assisted Baby-Making

Last week, news of CRISPR-engineered babies launched a firestorm of debate on the future of human reproduction: Is it safe? Is it ethical? Do we now have the ability to “play God”?

But even as scientists, ethicists, and the general public struggled with the implications of a fundamentally-altered reproductive future, other teams released results that also have the potential to disrupt reproduction—not for genetic treatment or enhancement, but to help those who cannot get pregnant produce healthy, living babies.

In one study released in Nature, a team engineered a placenta-like structure inside a test tube. An ephemeral and often forgotten tissue, the placenta is the critical link between the mother and fetus, providing oxygen and nutrients for the developing baby. Failure of the tissue can lead to miscarriages and stillbirths. The “mini-placenta” mimicked its biological counterpart so well that it fooled an off-the-shelf pregnancy test, and it’s now survived for a full year inside its petri dish.

In another case published in The Lancet, scientists transplanted a uterus from a deceased donor into a woman born without one. She carried a living, healthy baby to term, who is also about to celebrate her first birthday later this month.

Here’s the skinny on the whats and whys.

A Pregnant Placenta-in-a-Dish

Other than reports of celebrities eating their placentas after birth, the tissue doesn’t get much love. Placentas are often trashed as a side-product, no longer needed for their support function following birth.

But now it’s their turn in the spotlight.

“The missing link between complications during pregnancy and development of the fetal brain has been hiding in plain sight for a long time,” said Dr. Daniel R. Weinberger at the Lieber Institute for Brain Development in Baltimore. “It’s the placenta.”

During pregnancy, the placenta develops from a few cells into an organ of about a pound—and many of those cells are from the fetus, not the mother. Early after the fertilized egg implants into the uterine lining, it sends out a team of “intruder” cells that tunnels into her blood vessels and coaxes them to redirect towards the embryo, providing oxygen and nutrients.

But it’s not just a supportive organ: the placenta is also the barrier against infections, a recycling plant that pumps out waste, and a secretion factory that secretes hormones and other proteins. Basically, it’s a life-support system for the fetus, and failure—perhaps unsurprisingly—leads to stillbirths, miscarriages, and other pregnancy complications.

This happens more often than expected, roughly 20 percent of the time, and damages the mother’s health later in life. Heart disease, stroke, and gestational diabetes are all problems that can eventually lead to the mom’s obesity and metabolic disease. What’s more, mental stress on the placenta is linked to a higher chance of neurodevelopmental conditions like autism in baby boys.

So why hasn’t the placenta gotten much love? Mainly, it’s really, really hard to study. Scientists normally can only dissect placentas after birth (for obvious reasons), which is akin to an autopsy rather than a live feed.

Human placentas are also critically different from those of other species, said Dr. Ashley Moffett at the University of Cambridge who led the study, so animal models and cells lines don’t work.

To get around these issues, Moffett and her team turned to a recent trend in biological models: organoids. Using ethically-obtained tissue, the team isolated cells called trophoblasts from discarded placentas at roughly six to nine weeks of gestation. They then grew the cells inside a 3D gel structure to provide shape, and supplied it with the necessary nutrients and proteins. Within two weeks, the team had placenta-like organoids in the tissue dish.

Careful examination found that these mini-placentas acted like the real thing around the first trimester. They popped out structures normally found within biological placentas, including finger-like, wavy projections called villi that grab onto oxygen and nutrients. They also produced gonadotropin, a hormone that’s measured by drug store pregnancy sticks. In fact, when the team stuck a test into an organoid it registered as “pregnant.”

Slightly creepy? Sure! But the team thinks their placenta-like organs have much to offer. “These ‘mini-placentas’ build on decades of research and we believe they will transform work in this field. They will play an important role in helping us investigate events that happen during the earliest stages of pregnancy and yet have profound consequences for the life-long health of the mother and her offspring,” said study author Dr. Graham Burton, adding that they may also reveal why certain viruses—such as Zika—can pass through this defense barrier but not other similar viruses.

Going forward, the team hopes to use these shelf-stable placentas to study the flow and ebb of hormones as they grow, in order to suss out early biomarkers that may indicate pregnancy problems. Eventually, the “mini-placentas” could also be used to test the effects of drugs or vaccines taken during early pregnancy, or even to provide stem cells to treat failing pregnancies.

Transplanted Womb Births Healthy Baby

Meanwhile, work on the clinical side is tackling a separate problem: is it possible to birth a baby from a transplanted womb, placenta and all?

If the idea sounds completely out there, there is in fact precedent. So far, roughly a dozen babies have been born from uteri transplanted from living donors—generally the recipient’s mother, sibling, or close friend. But so far, doctors haven’t had much luck using uteri donated from the recently deceased.

Now, Dr. Dani Ejzenberg at the University of Sao Paulo in Brazil reported the first birth of a healthy baby from a transplanted uterus from a deceased donor. The recipient mother is a 32-year-old woman born without a uterus, though otherwise healthy, who received a matching uterus from a 45-year-old mother of three who died from stroke.

“With a deceased donor, you reduce the risk because you don’t have the risk to the donor—and you reduce the costs, too, because you don’t have the hospitalization and the very long surgery of the donor,” said Ejzenberg.

It’s a special case. The recipient was born with healthy ovaries, meaning that her eggs could be collected. Back in 2016, the team performed the transplant—an intricate procedure lasting over 10 hours—and added a stump of the donor’s vagina to that of the recipient, while connecting all the blood vessels, ligaments and other tissues. The recipient took immunosuppressant drugs to keep her immune system from attacking the new organ.

Seven months later, following stringent monitoring of signs of immunorejection, the team transplanted a fertilized embryo engineered with IVF into the donor uterus. By 35 weeks, the baby was delivered without complications through a C-section.

The woman was a lucky case. Others who underwent the same procedure experienced immunorejection that prevented them from keeping the womb. The team actually removed the transplanted womb following the successful childbirth, citing that they wanted to currently focus on giving infertile women their first child.

These are still early days for uterine transplants—dead or alive—but the proof-of-concept shows that women who had to previously rely on surrogates may have an alternative way. The team is looking to further refine the protocol, for example, for how much immunosuppressant to give and harvesting the organ as early as possible, to potentially increase success rates.

These successes, in addition to work into lab-generated egg and sperm cells, suggest that human reproduction is poised for the ultimate disruption. Could a future gay couple use a deceased uterus or an artificial womb and an edited sperm-turned-egg fertilized embryo to give birth to genetic children? Could menopause be a thing of the past? And how would that reality alter the fabric of society?

Image Credit: Magic mine /

Use Peg Board and Rubber Bands To Make Your Own Pinball Machine

Inspired by a picture of a reconfigurable pinball table, Beth Sallay set out to build her own. As you can see in the video above, she has pulled off the goal quite admirably. The gameplay seems fun and it is fully reconfigurable. The most difficult part of this is getting […]

Read more on MAKE

The post Use Peg Board and Rubber Bands To Make Your Own Pinball Machine appeared first on Make: DIY Projects and Ideas for Makers.

Sensitive robots feel the strain

Flexible skin for soft robots, embedded with electrical nanowires, combines conductivity with sensitivity within the same material.

Reinforcement learning shows promise for industrial robotics

Industrial robots deployed today across various industries are mostly doing repetitive tasks. The overall task performance hinges on the accuracy of their controllers to track predefined trajectories. The ability of robots to handle unconstructed complex environments is limited in today’s manufacturing.

Two examples are flexible picking of previously not encountered objects or the insertion of novel parts in assembly tasks. There are numerous examples of spectacular robot demonstrators exhibiting dexterity and advanced control, e.g. robot Fanta challenge, or robots playing ping pong. However, these applications are hard to program and maintain, usually they are the output of a PhD thesis, and they haven’t made the leap into manufacturing.

Endowing machines with a greater level of intelligence to autonomously acquire skills is desirable. The main challenge is to design adaptable, yet robust, control algorithms in the face of inherent difficulties in modeling all possible system behaviors and the necessity of behavior generalization.

Reinforcement learning (RL) methods hold promise for solving such challenges, because they enable agents to learn behaviors through interaction with their surrounding environments and ideally generalize to new unseen scenarios.

reinforcement learning

Figure 1: Reinforcement learning loop for robot control. (Credit: Siemens)

Reinforcement learning

RL is a principled framework that allows agents to learn behaviors through interactions with the environment. As opposed to traditional robot control methods, the core idea of RL is to provide robot controllers with a high-level specification of what to do instead of how to do it. Thereby, the agent interacts with the environment and collects observations and rewards.

The RL algorithm reinforces policies that yield high rewards, see Fig. 1. RL can be distinguished in value-function-based methods and policy search. In policy search, robots learn a direct mapping from states to actions. In value-function-based approaches, robots learn a value function, an intermediate structure that assesses the value of an explicit state, and derive actions from the value function.

Both policy search and value-function-based approaches can either be model-based or model-free. Model-free methods do not consider the dynamics of the world. Model-based methods incorporate a model of the world dynamics, which is learned from data as well.

Reinforcement learning for industrial applications

As we can see, robot control methods can be grouped along a continuum where on one end we find “rigid” feedback control laws, which are hand-engineered, incorporate domain knowledge and the control structure is not adapted by data. On the other end of the spectrum we have RL methods, which allow learning control policies purely from observed data. Both methods have advantages and disadvantages.

Traditional feedback control methods can solve various types of robot control problems very efficiently, such as trajectory tracking in free space, by capturing the structure with explicit models, such as rigid body equations of motion. However, many control problems in modern manufacturing deal with contacts and friction, which are difficult to capture with first-order physical modeling. And if higher-level reasoning is required (where to pick in bin picking problems, for example) current robot controllers lack flexibility. Applying feedback control design methodologies to these kinds of problems often results in brittle and inaccurate controllers, which have to be manually tuned for deployment.

RL, on the other hand, can, in principle, learn any control structure. However, for real-world robots, the continuous exploration space is large and, hence, large amounts of data and, therefore, long training times are required. Moreover, unlike conventional feedback control, convergence and stability statements are difficult to derive for RL methods.

Just to name two recently popularized use cases for both control methods: Boston Dynamics is known for deploying conventional feedback control laws (more precisely Funnel Control) for all its well-known demonstrations. Google, on the other hand, has shown that RL is capable to arrive at a robot controller for bin picking simply through trial and error. However, several months of training on a robot farm were required to achieve the required control performance.

After realizing that robot control methods comprise a continuum, where the underlying dimension is how much influence online data has on shaping the control algorithm, it seems that best control performance for flexible manufacturing has to combine both traditional control theory and data-driven RL. Traditional control can provide guarantees in safety and performance, while RL can bring flexibility and adaptability, if tuned correctly. In a way, RL removes the specificity needed at the engineering stage, where controls are designed. It targets to achieve the same performance than a carefully engineered feedback control algorithm, but without the need of tedious programming and rules.

We suggest decomposing robot control pipelines, which consist of perception, state estimation, control etc, into sub-problems that can be explicitly solved with conventional methods and sub-problems, which are approached with RL. The final control policies are then superpositions of both data-driven components and control policies from first-order models. Our approach combines the benefits of traditional control theory (e.g. data-efficiency) with the flexibility of RL. For example, position control is taken care of by a PID controller, and RL contributes the control part that deals with friction and contacts. We have conducted studies on different industrially relevant use cases, which amongst others include robots to perform real-world assembly tasks involving contacts and unstable components.

Figure 2 illustrates two assembly use cases, where conventional feedback control was combined with RL to solve complex assembly tasks in a flexible manner. Subfigures (a) and (b) show how a gear wheel is placed on a shaft. The use case is part of the Siemens Robot Learning Challenge. The robot required less than seven iterations to learn the required control policy. Subfigures (c) and (d) show a different use case for which the same control algorithm was used as for (a) and (b). Again, after less than seven iterations, the robot learned the control policy.

reinforcement learning

Figure 2: Insertion use cases solved with a combo of conventional control and reinforcement learning. (Credit: Siemens)

A challenge persists in this approach. Seven iterations may seem reasonable for lab setups, but they entail an inherent risk, as every iteration in a friction-rich environment has the danger to damage the part in contact with the gripper. Accurate sensors and adequate constrain management can alleviate the problem. Those are better handled in the pipelines that use traditional control, and can filter the output of the RL commands. Note that a certain amount of engineering is still needed to ensure that the robot is not in a lock position, unable to move because of the constrains. In these situations, calling a human for help may be the best course of action. In addition, in order to reduce the number of real world iterations, novel approaches in simulation to reality gap (sim2real) have been proven to accelerate the learning.

As a conclusion, we believe the current hype of reinforcement learning around robotic applications has a valid motivation; however, it is not the main ingredient to guarantee success. End-to-end learning approaches have shown poor performance in tasks that require precision. In an analogy that we like to make, if you want to make a chocolate cake, chocolate (reinforcement learning in this case) is not the main ingredient. You still need eggs, flour, etc. These “less-sexy” ingredients are in our case traditional control approaches. They are the base to build a successful flexible robotics application.

reinforcement learning

Figure 3: Siemens Robot Learning Challenge. (Credit: Siemens)

Robot Learning Challenge

We strongly believe that to accelerate robot learning research and its adaption in industry, we need a benchmark for the research community. We have seen that the ImageNet benchmark, which was introduced by Fei Fei Li in 2009, became the catalyst for image classification with deep learning. Machine performance for classification surpassed human capabilities in 2015. Benchmarks accelerate research because they facilitate reproducibility and allow comparison of research.

In the case of robot picking, this work goes in the right direction. In the case of robotic assembly, there is still need for globally accepted benchmarks. Therefore, we introduced the Siemens Robot Learning Challenge at the first Conference for Robot Learning in 2017. The challenge consists of a gear assembly task as seen in Fig. 2 (a) and (b) and Fig. 3. Details and CAD models for 3D printing can be obtained here.

Since the inception of our challenge, we have seen a variety of research work being published that is based on the Siemens Robot Learning Challenge – see examples here and here. We would like to encourage the community to try the challenge and help us refine it to cover as many cases as possible. Only with a common, easily reproducible benchmark can the robot learning community start building pipelines and tools that built on top of each other. If you have tried it, and want to contribute with your results, feel free to email the authors.


About the Authors

Juan Aparicio is the Head of Advanced Manufacturing Automation at Siemens Corporate Technology in Berkeley, CA. Aparicio has extensive experience managing complex projects, involving hardware and software; and bridging the technology gap between universities and businesses. His areas of interest include advanced manufacturing, advanced robotics, connected cars, Industry 4.0, and cyber-physical systems.

Aparicio is member of the Technical Advisory Committee for the Advanced Robotics in Manufacturing (ARM) Institute in the US and the Project Manager of the Open Process Automation Forum.


Dr. Eugen Solowjow is a Research Scientist specialized in robotics and machine intelligence at Siemens Corporate Technology. He has received his PhD from Hamburg University of Technology (TUHH), Germany. From 2012 to 2017 he was employed as a Research Associate at TUHH and as a visiting researcher at U.C. Berkeley.

Eugen was the technical lead in multiple government funded projects at TUHH in the field of robotics. He co-authored 20+ peer-reviewed publications (IROS, ICRA, RA-L, AuRo etc.) and has received multiple scholarships, fellowships, and academic awards.

The post Reinforcement learning shows promise for industrial robotics appeared first on The Robot Report.

Robot Gift Guide 2018

Over a dozen robots that we promise will make fantastic holiday gifts

12 Days of Hobby Project Tutorials

Death Ray Designs offers 12 tutorials on texturing, painting, finishing, and basing gaming miniatures and terrain.

Read more on MAKE

The post 12 Days of Hobby Project Tutorials appeared first on Make: DIY Projects and Ideas for Makers.

Automation needs to strike balancing act

Earlier this month, I crawled into Dr. Wendy Ju‘s autonomous car simulator to explore the future of human-machine interfaces at Cornell Tech’s Tata Innovation Center. Dr Ju recently moved to the Roosevelt Island campus from Stanford University. While in California, the roboticist was famous for making videos capturing people’s reactions to self-driving cars using students disguised as “ghost-drivers” in seat costumes.

Last January, Toyota Research published a report on the neurological effects of speeding. The team displayed images and videos of sports cars racing down highways that showed spikes in brain activity. The study states,”we hypothesized that sensory inputs during high-speed driving would activate the brain reward system. Humans commonly crave sensory inputs that give rise to pleasant sensations, and abundant evidence indicates that the craving for pleasant sensations is associated with activation within the brain reward system.”

The brain reward system is directly correlated to the body’s release of dopamine via the ventral tegmental area (VTA). The findings confirmed that higher levels of brain activity on the VTA “were stronger in the fast condition than in the slow condition.” Essentially, speeding, which most drivers engage in regardless of laws, is addicting as the brain rewards such aggressive behaviors with increased levels of dopamine.

Autonomous vehicles could lead to a marketing battle for in-cabin services pushed by manufacturers, software providers, and media/Internet companies. As an example, Apple filed a patent in August for “an augmented-reality powered windshield system,” This comes two years after Ford filed a similar patent for a display or “system for projecting visual content onto a vehicle’s windscreen.”

Both of these filings, along with a handful of others, indicate that the race for capturing rider mindshare will be critical to driving the adoption of robocars. Strategy Analytics estimates this “passenger economy” could generate $7 trillion by 2050. Commuters who spend 250 million hours a year in the car are seen by these marketers as a captive audience for new ways to fill dopamine-deprived experiences.

I predict at next month’s Consumer Electronic Show (CES) in-cabin services will be the lead story coming out of Las Vegas. For example, last week Audi announced a new partnership with Disney to develop innovative ways to entertain passengers. Audi calls the in-cabin experience “The 25th Hour,” which will be further unveiled at CES. Providing a sneak peak into its meaning, CNET interviewed Nils Wollny, head of Audi’s digital business strategy. According to Wollny, the German automobile manufacturer approached Disney 18 months ago to forge a relationship.

Wollny explains, “You might be familiar with their Imagineering division [Walt Disney Imagineering], they’re very heavy into building experiences for customers. And they were highly interested in what happens in cars in the future.” He continues, “There will be a commercialization or business approach behind it [for Audi] … I’d call it a new media type that isn’t existing yet that takes full advantage of being in a vehicle. We created something completely new together, and it’s very technologically driven.”

When illustrating this vision to CNET’s Road Show, Wollny pointed to Audi’s fully autonomous concept car design that “blurs the lines between the outside world and the vehicle’s cabin.” This is accomplished by turning windows into screens with digital overlays that simultaneously show media while the outside world rushes by at 60 miles per an hour.

Self-driving cars will be judged not by speed of their engines, but the comforts of their cabins. Wollny’s description is reminiscent of the marketing efforts of social media companies that were successful in turning an entire generation into screen addicts. Facebook’s first president, Sean Parker, admitted recently that the social network was founded with the strategy of consuming “as much of your time and conscious attention as possible.” To accomplish this devious objective, Parker confesses that the company exploited the “vulnerability in human psychology.” When you like something or comment on a friend’s photo, Parker boasted “we … give you a little dopamine hit.”

The mobile economy has birthed dopamine experts such as Ramsay Brown, co-founder of Dopamine Labs, which promises app designers with increased levels of “stickiness” by aligning game play to the player’s cerebral reward system. Using machine learning, Brown’s technology monitors each player’s activity by providing the most optimal spike of dopamine. The New York Times columnist David Brook’s said it best, “Tech companies understand what causes dopamine surges in the brain and they lace their products with ‘hijacking techniques’ that lure us in and create ‘compulsion loops’.”

The promise of automation is to free humans from dull, dirty, and dangerous chores. The flip side many espouse is that artificial intelligence could make us too reliant on technology, idling society. Already, semi-autonomous systems are being cited as a cause of workplace accidents. Andrew Moll of the United Kingdom’s Chamber of Shipping warned that greater levels of automation by outsourcing decision making to computers has lead to higher levels of maritime collisions.

Moll pointed to recent spat of seafaring incidents. “We have seen increasing integration of ship systems and increasing reliance on computers.” He elaborated that “Humans do not make good monitors. We need to set alarms and alerts otherwise mariners will not do checks.”

Moll exclaimed that technology is increasingly making workers lazy as many feel a “lack of meaning and purpose,” and are suffering from mental fatigue, which is leading to a rise in workplace injuries. “Seafarers would be tired and demotivated when they get to port,” cautioned Moll. These observations are not isolated to shipping.

In the Pixar movie WALL-E, the future is so automated that humans have lost all motivation to leave their mobile lounge chairs. To avoid this dystopian vision, successful robotic deployments will have to strike the right balance of augmenting the physical while providing cerebral stimulation.

The post Automation needs to strike balancing act appeared first on The Robot Report.

Antenna evaluation method could help boost 5G network capacity and cut costs

Researchers have developed a method for evaluating and selecting optimal antenna designs for future fifth-generation (5G) cellphones, other wireless devices and base stations.