Category Archives: Technology

A Tricoper from the Saar

Flugroboter!

My German is basically non-existent, so I am constantly discovering wonderful new words in that language.  This week I learned the word “Flugroboter” – literally, “flying robot”, but generally applying to what the English literature calls “drones”.

Honestly, Flugroboter is so much cooler name than “drone” or eve UAV!


I was actually trying to grok an interesting looking tricoper, from Universität des Saarlandes. These copters—Flugrotoers!—have three instead of the now common four rotors.

This seems to stem from earlier work reported in a poster [1].

The new videos demonstrate the that it certainly works, especially the automatic stabilizing.

Of course, what catches the eye is the fact that the researcher is wearing protective gear, which is rarely seen in UAV demonstrations.

Is this device especially dangerous, or is this lab especially cautious?

Anyway, I’m happy to learn a new word, Flugroboter, which, by the way, is an excellent name for a band!


  1. Marius Jank, Simon Luckas, and Karsten Weber, Entwicklung eines Greifers für einen Flugroboter. Universität des Saarlandes Mechatronisches Projekt, 2016. https://www.uni-saarland.de/en/lehrstuhl/rudolph/research/stud-projects/greifer-fuer-einen-flugroboter.html

 

Robot Wednesday  (Roboter Mittwoch)

Robin Hood Coop: Offshore Finance for Doing Good?

I worry about this one.  It’s probably OK so far, but I still worry.

The Robin Hood Coop calls itself a “coop”, though it operates as an “activist investment fund”, using an opaque algorithm to automagically manage the investments.  Some of the proceeds are raked off to support the operations.

Investing in a magic algorithm should be done carefully in all cases.  so it is a definite red flag that I can’t find any information about the algorithm (called, “The Parasite”, which somehow fails to reassure me very much).  There certainly nothing published or peer reviewed on their web site.

There are additional warning flags in the promotional materials.  This isn’t just a hedge fund, it is “a cooperative that bends the financialization of economy for the benefit of those who are not the financial elite”.  There are management fees plus a rakeoff to “fund projects that expand the commons”.  And, of course, it is implemented using Ethereum, which is “revolutionary technology for decentralised computing” (so what could possibly go wrong?)

Lot’s of fine sentiments, but impossible to really know what they might mean in practice.

I have to admit that my eyebrows rose at the FAQ item, “Is it Legal?” (Answer: Yes, under the laws of Finland.)  I’m glad that this is a relevant consideration, but it’s still troubling that they think people would be in doubt.

Of course, the answer is actually much more complicated than the FAQ indicates, at least for people not living in Finland. You are transferring money to this Finnish organization, which is investing in the US. If you take money out, it’s another transfer from Finland.  As they say, “When you receive payments from Robin Hood, you will be liable for whichever taxes you would pay for investment income in your country of residence and/or Finland.”  In short, who really knows?

Anyway, this is probably OK, though it seems unlikely to be a particularly profitable investment.  It may be a good way to support the “commons” projects they select, though, again, I don’t really know the implications of such international investments.

In the long run, they might be well advised to create a network of RHC affiliates incorporated in different jurisdictions.


This project has moved to a more troubling RHC2.0, which manages all shares as a private cryptocurrency (RobYns) which is used for trading in various currencies and cryptocurrencies.  This may or may not be a cost effective way to run the service, but it certainly does raise issues of trust. (Do I want to invest via a fund that has no humans in charge, and has no legal presence in my local jurisdiction?)

And if I understand their materials, they correctly are aiming to go even farther, with V3.0 called Robin Hood Unlimited.  This is a member owned (I think) “platform, where anyone could develop financial instruments, and launch them as apps for other members to use”. The goal is to offer “every opportunity to devise different investment strategies and different ways of directing profits or other funding to projects”.

Cooperative or not, this is a description of an offshore, extralegal, money hub, which is not a socially positive animal.   I this an “offshore coop”?  Do the benefits of “cooperative” outweigh the negatives of “offshore”?  Does the “Unlimited” cancel out the “member owned”?


Whether you share my own aversion to hot money in any form, you have to agree that RHC seems to take the motto “think globally, act locally” kind of backwards.  The fund is directing funds from everywhere to a few deserving projects, extracting capital and transferring it. These transactions may be ethical (though we have to trust them on that), but they certainly aren’t local for most of the investors.

The blockchain technology they use not only makes this strategic error easy, it is really the only way that blockchain can reasonably be used.  A global, peer-to-peer network is a primary affordance of blockchain technology, which is just plain the wrong model for local economies. In other words, selecting Ethereum technology is leading down the wrong path.

My basic point is that where ever the investors are, they surely could find local enterprises to invest in.  And that is what we all should do.  Shipping funds off to other continents or to untethered Internet projects is not a good way to make your own community better, which should be a top priority. In fact, it tends to move funds away from local social enterprises.

The new “RH Unlimited” project will likely be much worse. Much, much worse. Sure, it will be possible to create local tokens and local cooperatives, though we already can do that with “some guys in Finland”.  But it will also be trivial to sell and buy shares in anything, with unknowable consequences.

Making it easy for anyone to mess about with financial instruments is really not a way to promote community solidarity, trust, or sustainable development.  How will this be policed, if it even can be?  What happens if the coop is turned into a financial tool for criminals, terrorists, corporate trolls, and/or political shenanigans?

I note that even if you are satisfied that the current leadership is ethical, a decentralized organziation–cooperative or not–can be taken over by nefarious forces.  So watch out.


I probably shouldn’t be too hard on RHC.  There are dozens of variations on these “non-extractive investment” ideas and Ethereum is a favorite technology for these concepts.  To me, this looks like a hammer in search of nails.  We have a technology that lets people build their own financial systems, so obviously we should tackle the problem of ethical finance.

My own view is that the harder problems are social, and they need to be solved by talking and working together, face-to-face, in our own communities.  Imagining that faceless, soulless Internet technology will help in this ground campaign is misguided.  In fact, blockchain technology is antithetical to the personal human contact essential to actual ethical economics.

I’m willing to be proved wrong.  I will leave this as a challenge to RCH and other similar projects.  Let’s see what kind of sustainable “non-extractive” economic activity is actually possible.

 

Cryptocurrency Thursday

Festo’s Wondrous Robots

I love Festo’s bioinspired robots!

This summer they released a short video summarizing their “bilologic network” of researchers, designers, and technologists.

The video has some old favorites (why are there not millions of Festo butterflies in atria around the world?) and at least one new one: the chameleon tonge grasper.  Shwapp!

So cool!

 

Robot Wednesday

Blockchain support for local solar power generation

Let’s combine two of my interests, Solar Power and Blockchains. The use case here is a peer-to-peer microgrid, buying and selling power from local small scale solar generation. This idea is scarcely new, power cooperatives and co-generation schemes have been around for a long time, and there are a lot of contemporary variations on this theme (and not just for electricity). <<link>>

 

https://robertmcgrath.wordpress.com/2018/01/30/manzanedo-and-trepat-on-positive-platforms/

 

https://robertmcgrath.wordpress.com/2016/05/31/what-is-platform-cooperatism-real-platform-cooperatives/

 

The blockchain wrinkle is to use a blockchain to track the transactions among the participants.  The idea is that the distributed ledger is an inexpensive and robust way for people to buy and sell power they generate in tiny amounts.  In addition, the decentralized blockchain protocol kind of matches the decentralized group of generators and consumers, so neighbors can share without the overhead or interference of third parties.

This approach has been going for a while in at least a few cases, such as the Brooklyn Microgrid. TBM uses some proprietary technology to connect the solar generation hardware to the software accounting and trading system.  Much of the work is community organizing, recruiting people in the local community to participate and invest.  (This kind of thing is not really a good use of kickstarter or something to dabble in from afar.)

I’ll note that from the start the Brooklyn Microgrid has boasted that it uses blockchain technology.  But you would never know that from the web site, recruiting materials, and mobile apps.  And that is how it should be!  Normal humans should never see a blockchain!  TBM is a good example of just how little the blockchain matters to ordinary users—you could take away the blockchain and no one would know or care.


This summer, the Power Ledger is launching another such effort,  down under [1].  Actually, Power Ledger is more equivalent to the proprietary technology the BM is built on.

PL is actually built on top of Ethereum, which seems like a very plausible technical strategy—Ethereum’s executable contracts are just the thing for this use case.  PL plans to create their own tokens and run markets for electricity.  This would be one of many such token/market systems built on Ethereum, so again, a plausible engineering decision.

(While Ethereum has grievous problems, and is far from idea; for any serious long-term business, it’s better than making up your own blockchain solution from scratch.)

Power Ledger isn’t even in beta yet, so it’s impossible to say how well it might work.  Given the not-particularly-innovative technical path, we can be confident that they will have working software.  But how well will they deliver the dozens of use cases they are excited about, particularly, peer-to-peer solar production?

They seem to believe that what people need and want is some way to sell their excess production in a more or less open market.  Presumably, this is supposed to incentivize installing your our solar panels, and/or help defray the cost.  It’s not obvious how well this will work, or if there would be enough demand to matter.  (Is there really a use for lots of little dribs of electricity?  Is there much use to receiving tiny payments for such dribs?)

It’s not totally clear, but it looks to me like they intend to use the conventional grid, just as previous negative metering systems have done.  This concept depends on political action and, blockchain or no blockchain, is not popular with grid operators.

At this point, their most developed case seems to be aiming to operate charging ports for electric vehicles.  This sounds like a possible business, but doesn’t really require the full generality of the distributed ledger, or create a local microgrid.

As the BM project illustrates so well, the more important part is building the community of users.  And this effort needs well designed services and consumer apps, where “well designed” generally means no visible blockchain.  And, more than anything, it requires community organizing.  PL’s technology is neither here nor there for this part of the solution.

Worse, the project started with an ICO, which raised millions of dollars.   This has absolutely nothing to do with solar power at all.

Reading the tea leaves of Power Ledger’s materials, they call attention to a number of interesting possibilities down the road.  If the distributed ledger can do negative metering and trading among neighbors, then it might also do accounting for a Carbon tax or other offset system.  That would be a neat feature for a “smart meter”, and a blockchain might be a plausible way to do it (maybe).

In the end, though, these are neat ideas but it is far from clear that blockchain technology is the only or best way to implement them.  Yes, the distributed ledger is philosophically simpatico for the desire for peer-to-peer power trading.  But a real peer-to-peer system, a la BM, is about trust, and good service (which is about trust).  Making that happen doesn’t need a blockchain, per se, and, to the degree that the blockchain is “trustless”, it probably doesn’t help.  I’m pretty sure that happy Brooklynites are happy with TBM people and service, not with whatever software lies inside.

So, we’ll see.


  1. Jennifer Bisset, Blockchain helps us take green power into our own hands, in cNet – News. 2018. https://www.cnet.com/news/blockchain-democratise-solar-energy-power-ledger-startup/

 

Cryptocurrency Thursday

More Tensegrity Robots

Tensegrity + robots?!  This must have a coolness multiplier of “infinity times infinity”!

This summer John Rieffel and Jean-Baptiste Mouret report on a cool “soft tensegrity robot”  [2].  As they point out, the general approach called “tensegrity” is frequently seen in biological systems, which are primarily constructed of soft and flexible materials.  However, humans are just learning to design soft robots, and to date, have had limited success with tesegral designs. For one thing, soft, flexible, floppy materials are not easy to design for robotic “gaits”—methods of locomotion.

“[Soft robots] are by their very nature high dimensional dynamic systems with an essentially infinite number of degrees of freedom. The elasticity and deformability that provide their appeal come at the cost of resonances and tight dynamic coupling between components, properties that are often avoided, or at least suppressed, in conventional engineering approaches to robotic design. “ ([2], p. 318)

The research developed a soft (indeed, squishable) tensegrity robot, with vibrating weights as motors on three struts.  The rather mind-blowing idea is that different vibrations will wobble through the structure in complex ways, some of which are useful gaits!  Huh!  What!  Wow!

“[W]e explore the hypothesis that the inherent resonance and dynamical complexity of real-world soft tensegrity robots can be beneficially harnessed (rather than suppressed), and that, if properly excited, it can resonate so that the robot performs step-like patterns that enable it to locomote. ” (p. 319)

To achieve this, the research deploys machine learning to discover these useful settings. Where earlier projects developed a gait through trial and error, the machine learning essentially automates the ad hoc search.

Using machine learning not only speeds up design and optimizes the robot, it makes it adaptive to damage or changed conditions:  another round of learning can discover a new gait that works for the changed situation.  It also makes the design robust, in that the vibrating motors and weights do not have to be placed extremely precisely—the algorithm adjusts for minor variations in the actual robot.  Cool!

Looking at the demo video, it looks like a bit of a bumpy ride, and it’s not obvious how well this would scale up to larger sized robots (e.g., that I could ride in myself : -))  The behavior is defined by the resonance of the structure, which depends on the specific materials and cross sections and so on.  I’m no where clever enough to work out how a big one of these would work.

But obviously, this is an interesting approach for something like NASA’s tensegrity lander concepts (deeloping over the past few years, here, here, here, here, here).  Get the power and control on board, and a sensor payload, and you’ve got yourself a pretty robust planetary explorer that can learn to “walk” on whatever alien terrain it encounters.


  1. John Rieffel and Jean-Baptiste Mouret, Adaptive and Resilient Soft Tensegrity Robots. xarive, 2017. https://arxiv.org/abs/1702.03258
  2. John Rieffel and Jean-Baptiste Mouret, Adaptive and Resilient Soft Tensegrity Robots. Soft Robotics, 5 (3):318-329, 2018/06/01 2018. https://doi.org/10.1089/soro.2017.0066

 

 

Robot Wednesday

Halaburda on the Blockchain: Blockchain not essential

I’ve been saying for years that blockchain technology per se isn’t necessary for many of the use cases advocated.

In the recent issue of Communications of the ACM Hanna Halaburda makes the same point [1].  “Most of the suggested benefits of blockchain technologies do not come from elements unique to the blockchain.” (p. 27)

Blockchain technology is basically a combination of three technologies, “encryption; automated execution of transactions (“smart contracts”); and distributed ledger, a type of a distributed database. “ (p. 27)  These technologies can be used together (as in Bitcoin or Ethereum), but also can be used separately. (The are also, for the most part, not especially “innovative”.)

Halaburda recounts the many confusions about this technology.  Bitcoin is the most famous and arguably the most successful.  But blockchains don’t have to be public and permissionless (a la Bitcoin), and Bitcoin has very little in the way of “smart contracts”.  And executable contracts a la Ethereum can be implemented by a variety of technologies, not just blockchain.  And so on.

She argues that this point is important for assessing the costs and benefits of blockchain technology.  For example, “smart contracts” may reduce costs, delay, and errors.  But they can be implemented on centralized systems—and have been for decades (e.g., “stored procedures” in database systems).

Similarly, many of the benefits of blockchain technology stem from the use of cryptography and digital signatures.  Bitcoin and its extended family have, quite rightly, turned attention to the value of these technologies. But Bitcoin itself uses standard and widely used cryptography.  (And where cryptocurrencies “innovate” in the area of cryptography, it is generally a catastrophe.)

“Currently, encryption is underutilized in business practice. Bitcoin’s blockchain itself uses standard, well-established cryptography tools. But excitement about blockchain’s safety turned more attention to the new developments in cryptography.” (p. 28)

As to the blockchain itself, it is mainly beneficial where “where reconciliation of contradictory ledgers is costly.”  It’s far from clear when this is a problem, and whether blockchain really is cheaper.

“So far it has not been clearly demonstrated in which circumstances the benefits of employing a distributed ledger outweighs the cost of delays and duplicated storage.” (p. 28)

Blockchain distributed ledger technology is supposed to have additional benefits over other decentralized databases (which have been around for decades).  In particular, the decentralization offers disintermediation, which, theoretically might offer cost savings or other advantages (such as resistance to censorship or tax collection).

These benefits have proved to be hard to realize. As I have pointed out many, many, many times, the “trustless” blockchain essentially pushes the problem of trust out to other parts of the system.  As Halaburda says, assets have to get onto the blockchain in the first place.  (And the many failures of blockchain based systems are mainly failures of these other parts of the end-to-end system.)

She points out that for other use cases, such as property records, the problems do not involve storage or publication of transaction records (e.g., property deeds).   The problems are mainly due to lax (or possibly dishonest) record keeping and transcription errors.  Using a blockchain will not improve these processes, and could potentially make things worse because there is no way to delete or correct errors on a blockchain.

I’ll add that even the most successful use case, digital currency, doesn’t actually require a blockchain per se. It can be achieved by establishing a web of trust, along with robust cryptography and distributed databases.

In the end, Halaburda concludes that cryptography and executable contracts will probably be important innovations, distributed ledger technology may well not be. So the “blockchain revolution” may lead to a new world, but one without blockchains.

“The blockchain revolution may give us new tools and change the landscape of some industries. But since the benefits of encryption and smart contracts can be realized without a distributed ledger, the world after the blockchain revolution may well be a world without the blockchain.” (p. 29)


  1. Hanna Halaburda, Blockchain revolution without the blockchain? Communications of the ACM, 61 (7):27-29, July 2018. https://cacm.acm.org/magazines/2018/7/229049-blockchain-revolution-without-the-blockchain/abstract

 

Cryptocurrency Thursday

More Morphin Copters

Apparently, reconfiguring drones is an idea whose time has come.

Earlier I noted an admirably simple folding quad copter, from a French team.  This week I read of a group in Tokyo who see your quad copter and raise you four—a snaky octocopter that can configure in a zillion ways—the flying DRAGON [2] .  So there!

This flying snake thing has modules connected by gimbals, each with  two rotors, also on gimbals.  Altogether, the assembly can bend in 6DOF, just like a robot arm.   A flying robot arm.

The researchers conceptualize this robot as a sort of overactuated flying arm that can both form new shapes and use those shapes to interact with the world around it by manipulating objects.” (from [1])

Reconfiguring in flight is, well, complicated.

A key feature of this design is that the rotors aren’t all in the same plane as in a rigid quadcopter. This is actually a key to stability:  the rotors point in multiple directions and the body is rigid, yielding stable flight and hovering.

“To achieve an arbitrary 6DoF pose in the air, rotor disks cannot be aligned in the same plane, which is the case for traditional multirotors.” ([2], p. 1177)

The control system is modular, featuring “spinal” and “link” controllers, as well as a high level processor.  Indeed, the device looks like nothing so much as a hovering spine.

The demo video shows an impressive maneuver, slinking thorough a small horizontal hole, unfurling while hovering and slipping link by link up through the floor.  Pretty cool.

What’s more, the software autonomously determines the transformation needed. Very impressive.

This flying robot arm has the potential to be used as a flying robot arm:  it can poke and grasp and carry cargo.

It will be interesting to see how this approach compares to swarms of rigid copters.  What are the advantages and disadvantages of a handful of really complicated snakey fliers versus a constellation of many simpler fliers.   (A swarm is probably harder to shoot down.)

I predict that this will soon be a moot question, because there will be swarms that can lock together into spines, and disperse again into drones, as needed.


  1. Evan Ackerman, Flying Dragon Robot Transforms Itself to Squeeze Through Gaps, in IEEE Spectrum – Robotics. 2018. https://spectrum.ieee.org/automaton/robotics/drones/flying-dragon-robot-transforms-itself-to-squeeze-through-gaps
  2. M. Zhao, T. Anzai, F. Shi, X. Chen, K. Okada, and M. Inaba, Design, Modeling, and Control of an Aerial Robot DRAGON: A Dual-Rotor-Embedded Multilink Robot With the Ability of Multi-Degree-of-Freedom Aerial Transformation. IEEE Robotics and Automation Letters, 3 (2):1176-1183, 2018. https://ieeexplore.ieee.org/document/8258850/

 

Robot Wednesday