Here is yet another incremental advance in brain-machine interface (BMI) technology – decoding what someone is saying from their brainwaves using a neural network and machine learning. We are still a distance away from using a system like this to allow someone who cannot speak to communicate, but the study nicely illustrates where the technology is. Here is the BBC’s reporting:

Scientists have taken a step forward in their ability to decode what a person is saying just by looking at their brainwaves when they speak.

They trained algorithms to transfer the brain patterns into sentences in real-time and with word error rates as low as 3%.

Previously, these so-called “brain-machine interfaces” have had limited success in decoding neural activity.

Now here are all the caveats from the paper. First, the technology used electrocorticography (ECoG), which is an EEG with brain surface electrodes. So this requires an invasive procedure, and persistent electrodes inside the skull and on top of brain tissue. Also, in order to get the best performance, they used a lot of electrodes – resulting in 256 channels (a channel is a comparison in the electrical activity between two electrodes). They simulated what would happen with fewer electrodes by eliminating many of the channels in the data, down to 64, and found that the error rates were about four times greater. The authors argue this is “still within the usable range” but they consider usable range up to a 20-25% error rate. What this shows is that – yes, more electrodes matter. You need the very fine discrimination of brain activity in order to get good (usable) results.

Continue Reading »

This is an interesting idea that will probably not be actually implemented (although not impossible) but does raise some important points. A paper explores the viability of using urea from human urine as an agent in lunar concrete. Why is something like this even being considered?

The overwhelmingly dominant factor of building anything on the Moon is that it costs about $10,000 to put one pound of anything into Earth orbit, and more to take it to the Moon (although most of the energy would be used just getting into orbit). This is why it is a high priority for NASA to reduce the cost of getting stuff into space. Elon Musk has also made this a priority and SpaceX is geared mainly toward this purpose. Even if they reach their goal of reducing the cost by 10 fold, to about $1000 per pound, that still adds up when you are trying to build an entire Moon base. One solution is to use as much native material as possible.

Let’s talk a bit about the lunar regolith. The term regolith just refers to any loose material on top of the rocks on a world’s surface. The Earth has regolith, we call it dirt, sand, or soil. The lunar regolith is the result of micrometeors pulverizing the lunar surface for billions of years. In most locations the regolith extends down 4-5 meters, but can be as deep as 15 meters in places. Because of the absence of natural erosion from wind, water, or biological activity, the lunar regolith remains sharp and pointy. So the Moon is basically covered with a deep blanket of fine but jagged dust.

Continue Reading »

One extremely exciting technology that is in development is the brain-machine interface (BMI). This technology allows for communication between a biological brain and a computer chip. Once perfected, the implications are incredible. Perhaps most exciting is the possibility of  robotic prosthetics that can be controlled with the mind. There are many medical conditions that impair the ability to move, from spinal cord injury to strokes that can literally cause people to be “locked in”. In many conditions the brain is working, but the peripheral nervous system is damaged. With a sufficiently functioning BMI non-functioning limbs or blockages in communication can be bypassed. Amputees could also have fully robotic limbs to replace what’s missing.

From a theoretical perspective, all of the necessary proofs of concept have been done. The brain can learn to control machines, even computers. The brain can map itself to new limbs, and it can incorporate new sensory feedback. With sufficient sensory feedback, control is enhanced, and the sensation that the new artificial limb is part of one’s body can be complete. We can even be made to feel as if we occupy virtual avatars.

Computer hardware and software technology are already powerful enough meet the demands of any such BMI application. Robotics are also functional enough to work, although there is certainly a lot of room for improvement here. All these components are good enough to use right now, and continued incremental advances will just make them better.

The main limiting factor for BMI applications right now is the interface itself – how do we connect the brain to silicon? Scalp surface electrodes are the safest and most convenient. Electrical signals from the brain do make their way through the skull and scalp and can be recorded, but they are greatly attenuated. Only relatively large parts of the brain firing at the same time produce a sufficient signal to produce a detectable wave at the surface. Still, even with this method there are BMI applications, but the discrimination is limited.

Continue Reading »

We have many current solutions to the energy challenges we face – solar, wind, geothermal, hydroelectric, nuclear. I and others have previously argued that if we are going to have the best chance to minimize climate change as much as possible, we need every option on the table. That is partly because no one of them is perfect, and they get worse the more you try to push penetration into the energy infrastructure. So best to pick the low-hanging fruit from each. Also, if we try to solve our energy problems with one solution we would likely run into shortages of raw material and optimal locations.

In fact, it would be nice to add more options. Researchers are aggressively trying to work out the challenge of fusion energy, for example. Even if that succeeds it won’t be the one optimal solution for all our energy needs, but it will help a lot. We also need better grid storage options. You may also notice that I did not have biofuels on the list, because I think it’s unlikely they will make a major contribution, at least not if they need to use up arable land that could be used for food production. But if we can develop a biofuel process that does not require premium arable land, that could fill a niche also.

There is another possible technology on the horizon that occasionally grabs headlines when some incremental advance is made – the artificial leaf. This term is used to refer to any process that uses light in a photosynthesis type process. So it is not a photovoltaic, directly creating electricity, but rather is using light to split water and combining the hydrogen with CO2 from the atmosphere to make fuel. Essentially, this is artificial biofuel. Then why not just use biological leaves to make biofuel? That gets back to the land issue – you could theoretically have a biofuel plant using an artificial leaf process in the desert. I suppose theoretically you could also use salt water for the process, and would therefore not be a drain on our fresh water supply (but I am not sure about this).

Continue Reading »

I’m a sucker for perpetual motion machines. I don’t mean that I think they work – they don’t – but they are often intriguing contraptions out of some cyberpunk fantasy. They are also often a bit of a puzzle. How are they supposed to work, and why don’t they? That free energy or perpetual motion machines don’t work is a given, because of the laws of thermodynamics. Energy has to come from somewhere, so for each such claim it’s a fun game to figure out where the energy is actually coming from. This game also helps dispel any notion of continuous or free energy.

A new perpetual motion claim is revealed in an article in the Rob Report. The claims is for an electric plane that will fly mostly with the energy generated by the friction of the flying itself. The idea is that the plane will have rechargeable electric batteries that are used for take-off and landing. But while in flight, the batteries will be recharged by vibrations and the flexing of the wings. The inventor, Michal Bonikowski, who calls his project Eather One, hopes this will yield enough energy to keep the plane flying indefinitely.

The problem with this concept, as with all perpetual motion concepts, is the second law of thermodynamics. Every time you change energy from one state to another, at least a little bit is lost. You can never have 100% efficiency. So the energy you use to propel the plane forward will have to be greater than the energy you harvest from pushing through the air. If you design a mechanism (as in the concept art) for harvesting air friction, the extra resistance from the mechanism will cause the plane to slow by more than using that energy to propel it will increase its speed. The entire process will be a net negative. You would be better off optimizing aerodynamics.

Continue Reading »

This is definitely one of those “Holy Sh*!” technology breakthroughs; a game-changer that will likely have many more implications than you can take in immediately. Researchers have demonstrated that with the “judicious design of a multi-level diffractive lens (MDL)” they can create a single flat (one thousandth of an inch thick) thin lens with an extreme depth of focus – four orders of magnitude greater than traditional lenses. Let that sink in.

The depth of focus (or depth of field for objects not at infinity) is the range over which objects are in focus. You can adjust the depth of field on a camera by changing the aperture, with smaller aperture settings having a deeper depth of field. But you still need to focus the camera to bring the desired image into sharp focus. This requires that the camera can change the distance between the lens and the sensor, and modern cameras may use multiple lenses. Good cameras also use multiple lenses for different colors (wavelengths of visible light) to make sure all the colors are focusing the same way.

Now imagine if all this could be replaced with a single very thin flat lens. That is what the researchers have done. They accomplish this by using nanostructures on the glass to control the path of the light. The lens can simultaneously focus objects at different distances, and also light of different colors. What we have now is a proof of concept, and of course we need to see what an an actual commercial camera using this technology will be like. But if the published results pan out, there are several immediately obvious implications.

The first is much smaller, cheaper, lighter camera lenses. This will be great for cell phones and other tiny electronic devices with cameras. Medical devices such as those used for endoscopic surgery would also benefit from smaller lenses. Any situation where size and weight are at a premium would benefit – such as drone cameras.

Continue Reading »

A new study published in Nature details the use of a neural network on a 2-dimensional computer chip that by itself can be trained to recognize specific images within nanoseconds. This is more of a proof of concept than something with direct immediate applications, but let’s talk about that concept.

To back all the way out – evolution represents hundreds of millions of years of tinkering with multi-cellular structures, and even longer when talking about biochemistry. This is a natural laboratory that has developed some elegant designs, and at the very least can serve as a useful source of inspiration for modern technology. That is the concept of neural networks, designing computers to work more like a vertebrate brain. Specifically, the “neurons” in a neural network are not just binary, on or off, but rather can fire with various degrees of strength. Further, their firing affects the activity of those neurons they are connected to. Computer hardware with networks designed on these basic principles are called  artificial neural networks (ANN). They hold the promise of not only faster and more powerful computing, but are designed to learn (which is why they are so often associated with artificial intelligence).

Another principle at work here is top-down vs bottom-up processing, another concept that has increasingly been incorporated into AI. If we go all the way back to the early days of AI the basic idea was to create high level computer intelligence that could solve problems with the top down, with deep understanding. That goal, now referred to as general AI, is still a ways off. But meanwhile AI has advanced considerably through more of a bottom-up approach, using algorithms to sift data in increasingly sophisticated and adaptable ways. We now have deep learning AI and other specific processes that can produce impressive results without any general AI “understanding” what it is doing.

One question is – will we be able to build a general AI out of these limited AI components? Is it just a matter of building in enough sophistication and complexity? We won’t know until we do it, but if living organisms are any guide, I think there is reason to be positive. Specifically – that is basically how our brains work.

Continue Reading »

It’s important to recognize that we currently do not know with any confidence the path forward that our energy infrastructure will take. This is why we have to spread our bets out on as many technologies as possible – we don’t know which ones will be the most successful. Many people place their hopes on battery technology, and there is no doubt that batteries are a great energy storage medium and will play a critical role in our energy future. But batteries are not a simple panacea, and we may run into important limits. This is why we need new battery technology.

The demand for batteries is likely to increase significantly. Electric cars depend on batteries, and therefore putting millions of EVs onto the road means necessarily putting millions of batteries on the road as well. Also, batteries are one possible solution to home and grid energy storage, which will be necessary if we want to maximize renewable energy sources like wind and solar. Current lithium-ion battery tech is great, and is getting incrementally better all the time, but it has limitations. One significant limitation is the availability of lithium and cobalt which are necessary for their manufacture.

Cobalt, for example, comes mostly from the DRC, an unstable country, and it comes mostly as a byproduct of copper and nickle mining. Global supplies are expected to fall short of global demand, and if there is a surge in Li-ion batteries this will only get worse. Lithium is more complicated, and we are not really sure what the worldwide supply is. For now there is no problem, but there is widespread concern that lithium supply will not keep up with demand as EVs take to the streets. We also do not currently have the ability to recycle lithium into a pure enough state to reuse in batteries.

What battery tech is on the horizon that will potentially change the game for batteries? For now, continued incremental improvements in Li-ion battery technology are important. We need to squeeze as much function out of the raw materials as possible, with greater capacity, and longer charge-discharge lifespans. Right now Tesla boasts million-mile batteries for its EVs. Increasing the lifespan further will decrease the need for new batteries as replacements. Batteries from retired EVs can also be repurposed for grid storage, where it wont’ matter if their range has decreased.

Continue Reading »

The use of artificial intelligence in the drug discovery process is not new, but it is advancing in significant ways. Several weeks ago the BBC announced the first AI developed drug to be taken to human trials. Now they are announcing the discovery of a new antibiotic using AI. Let’s talk about drug development to see how advances in AI are impacting this process.

Finding a drug that is useful medically is tricky, because it has to have a lot of properties simultaneously, and any one property can be a deal-breaker. A useful drug needs to get into the body, get to the target tissue, survive long enough to have the desired effect, it needs to have a desired effect at a dose that is lower than doses that cause significant side effects, and it needs to lack significant toxicity, such as liver or kidney damage. Will the compound be stable on the shelf? The same needs to be true, at least in lack of side effects and toxicity, for all the metabolites of the drug that may be created before everything is eliminated. On top of that we have to worry about drug-drug interactions, and even interactions with certain foods.

For this reason there is no perfect drug. Every pharmaceutical is a trade-off. Being “natural” is also not a magic wand that bypasses all these concerns. Substances that occur in nature did not evolve for our benefit. They generally evolved to be poisons to creatures that might eat them, including us. Drugs derived from plants are basically poisons that we have purified, usually altered, and then discovered a dose range that can be safely exploited.

Continue Reading »

Jeff Bezos, founder of Amazon, is the richest person in the world, with a net worth of around $115 billion. He recently announced that he is pledging 10 billion of those dollars to the Bezos Earth Fund, the primary objective of which is to fight climate change:

⁣⁣⁣”Climate change is the biggest threat to our planet. I want to work alongside others both to amplify known ways and to explore new ways of fighting the devastating impact of climate change on this planet we all share. This global initiative will fund scientists, activists, NGOs – any effort that offers a real possibility to help preserve and protect the natural world.”

Bezos reports that he will begin dolling out money this summer. Here is the big question – how will he spend this money? And how should he spend it? How would you spend it? This is a complex question. His statement suggests that he wants to primarily fund research, which I think is a good place to start – but research into what? I think this question is answered by simply looking at the sources of human CO2 release:

“87 percent of all human-produced carbon dioxide emissions come from the burning of fossil fuels like coal, natural gas and oil. The remainder results from the clearing of forests and other land use changes (9%), as well as some industrial processes such as cement manufacturing (4%).”

Clearly the largest culprit is the burning of fossil fuels. Land use is also a big chunk, and then 4% make up other industrial processes. Every bit helps, but let focus on these top two, starting with the smaller portion, land use. I think there are three significant efforts that could help reduce CO2 loss from improper land use. The first is to stop cutting and burning forests that represent major carbon sinks. I am not talking about logging for lumber, which can be done sustainably, but mostly the reduction of the Amazon rainforest and other old-growth forests. This will require regulations, but also we need to seek ways to reduce the incentive for farmers to clear forest to grow more crops.

Continue Reading »