Grossmann: The Bee Rescue – Some Old Solutions to Some New Problems

13 March 2014

In an effort to maintain the population of bees and other pollinators, the United States Department of Agriculture has budgeted $3 million.  Most of the money will go to ranchers, farmers and beekeepers in a conservation effort to preserve and expand pollinator habitat.


Bee populations have been declining for over 7 years now. First, termed a “disappearance,” then, a “die-off.” the continuing depopulation is, now, formally referred to as “Colony Collapse Disorder.” The continuing decline has been both rapid and widespread affecting perhaps the entire world.

Bees get a lot of scientific attention because they are vital to American agriculture, which is vital to the American economy. Without bees, production of some of our most profitable crops would be impossible. Every few weeks, a news article announces the discovery of “the cause” of the threatened bee “extinction.”  In fact, there probably isn’t a single cause. The current die-off seems to be the result of several factors working together.

The puzzle goes like this. A bee (1) has a parasite like varroa mites; (2) is exhausted by transport over long distances; and (3) is exposed to a particular pesticide. Alone, none of these factors would kill a bee. Even all of these put together wouldn’t kill a bee. However, all of these put together might weaken the bee’s immune system. Then, with a compromised immune system, the bee contracts, and dies from, a completely unrelated disease. That disease is the final cause the bee’s death. However, the underlying cause is an immune system compromised, not by one factor, but by a particular combination of several factors. For now, that combination remains a mystery.


Modern agriculture has come to be dominated by a particular style called monoculture.  The modern farm is a study in intensive land use with about every square foot of available soil used for the continuous cultivation of crops – or more precisely a signal crop.  This modern style has little in common with the traditional agriculture of even a generation ago.

In the past, the typical farm included a fair number of fallow (unplanted) tracts of land in which wild brush and unmown grass were allowed to grow.  These tracts served several purposes.  They provided “breaks,” uncultivated buffer areas between cultivated fields of crops.  First, breaks slowed or prevented the spread of disease from field to field.  And, second, breaks prevented the seeds of one kind of crop from creeping into fields planted with another.  The third purpose of keeping some land fallow (unused) was to prevent soil depletion.  The practice of letting some fields “rest” for an a season was called crop rotation, which helped prevent a loss of, or restore,  fertility to tracts of land.

Traditional agriculture had always avoided modern monoculture’s practice of planting only one kind of crop.  The traditional reason for planting several different kinds of crops was, again, a sort of insurance against the spread of disease.  While one kind of crop might fall victim to disease, another would be less susceptible and survive to produce a much-needed yield at harvest.

What happened to traditional agriculture?  Advances in chemical fertilizers, herbicides and pesticides have dramatically reduced the need for crop rotation and fallow tracks of land as buffers.  But this created another problem.  The modern farm needs bees just as much as the traditional farm it replaced.  And bees need habitat.


When we think of bees, we tend to think of the hive-dwelling honeybee.  The honeybee seemed to fit in perfectly with modern monoculture.  Like everything else needed by the modern industrial farm, when you need bees, you just order them “brought in.”  Beekeepers truck bees, sometimes hundreds of miles, to various locations during pollination season.  Then, the bees are trucked out when pollination is over.  At least, that was the plan before CCD and honeybee depopulation became a reality.

But, with or without depopulation, what’s with “habitat?”  The only thing honeybees need is a hive, a beekeeper, and the beekeeper’s truck.  Right?  Well, not quite.  Honeybees aren’t the only pollinators.  Worse, honeybees can’t pollinate some cash crops including certain varieties of tomatoes, cranberries, almonds, apples, zucchinis, avocados, and plums.  For these crops you need bumblebees.

So, why not truck-in some bumblebee hives?  And there’s the problem. Bumblebees don’t live in hives.  The plump bumblebee is the nearest thing to a loner within its social species.  Bumblebees don’t build permanent hives.  They build nests that are deserted for a new location on a yearly basis.  The bumblebees don’t forage (search for and find food) in swarms, but wander alone from flower to flower in open grasslands.

On the traditional farm, these wild bees made their nests in fallow tracks of grass lands or break areas between cultivated fields.  Because the bumblebee’s service as a pollinator is only needed seasonally, these bees survived during the rest of the year by foraging in the same wild grasslands in which they built their nests.


Monoculture changed all that.  Fallow tracts, breaks, and buffers vanished with every yard of available soil planted with a crop.  Even the small islands of wild grass along the farms paths and roadways were pressed into service.  And the bumblebees left.

What did we lose?  A lot.  The bumble’s unique style of pollination is required, and accounts, for about 3 billion dollars in produce each year.

Fresh off the farm, the bumblebee made its way to the city or, at least, to more populated areas to find the welcome mat missing.  Modern urban and highway landscaping favors a neatly manicured look that requires the elimination of the wild grasslands required by the bumblebee’s lifestyle.  In parks and even around highway overpasses, that great enemy of bumblebee habitat, the lawn mower, doesn’t destroy the grass, but prevents the appearance the blooms and blossoms on which the bumblebees depend for food.  And worse, the lawn mower is the arch-enemy of bumblebee nests.

When the habitat vanished, so did the bumblebee.  Beginning in the late 1990’s, these bees all but disappeared from a vast area of their range extending from the Pacific Coast of California north into British Columbia.  Only recently have there been sightings of even a single bumblebee in several states that once supported an enormous population.


It is said that those who felt uncomfortable in “civilization” used to become trappers and wander into the mountains — earning the name “mountain men.”  Well, maybe bumblebees did the same.  As these bees almost completely disappeared from their lowland range, their numbers were, and are, unaffected in the North American Rockies where they continue to live and thrive.  Mountains are not favored for agriculture and the rough beauty of mountainous areas is only enhanced by wild growing grasslands.  The mountain habitat is well within the bumblebees comfort zone.


With all the developments in the efficiency of modern agriculture, it is a little surprising to read of a USDA spokesman discussing the use of cover crops, rangeland, pasture management and other practices that dropped out of modern agriculture decades ago.  But the purpose behind the reintroduction of crop rotation, breaks, and buffers makes sense if the purpose is to preserve native pollinators, most prominently the often forgotten bumblebee.

Without effective pollinators, there will be no harvest in spite of the most intensive and efficient use of the available land.  The USDA spokesman explained that these “new” practices “are expected to provide quality forage and habitat for honey bees and other pollinators, as well as habitat for other wildlife.”

1 a a a a a-a a a a a-a wwwww-wwwww-wwwm

m-mwmmw-wwwww 2 a a a a a-a a a a a-a wwwww

wwwww-wwwww-mmmmm-wwwww n

1 b b b b b-b x o o ollo x x mm-mmmm j j w 2 b b b b b-

b x o o o-x x x mm-mmmmww w 3 b b b b b-b wwww-

wwwmm-mmmmwww n

1 c c c-c c c c c-c c c c c-c c c o o-o o o o i o-

o o o o o-o o o o o-o o o w w w w w w w 2 c c-c c c c -

c c c c c -u u u u o-o o o o o o-o o o o o- o o o

o o-o o o w w w w w w

c00000f00000i00000  j00 k 0 0 0 0 0

b00000e00000h00000j00 k 0 0 0 0 0

a00000d00000g00000j00 k 0 0 0 0 0

p C B A s i ii 3 2 1 + + + + +



Grossmann: Toy Robot Spiders — As If the Real Things Weren’t Enough 

6 March 2014

“The only excuse for making a useless thing is that one admires it intensely.”

Oscar Wilde

Before we go, we have to get some definitions out of the way.

A robotic purist will explain that there’s no such thing as a toy robot.  The words “toy” and “robot,” used together, form an oxymoron.  In other words, by definition, a toy isn’t a robot, and a robot isn’t a toy. A robot is a machine that “does work.”  A toy is a machine, but not a machine that does work.

An animatronic device is a machine that moves like a living creature.  Animatronic devices are used for entertainment.

But these aren’t robots. Right?

Is entertainment work?

Well, uh . . . .   Let’s get back to robots.

No one can play with a robot. Right?

Well, I have to admit that children can play with anything including (and especially) the cardboard box their “toy” came in.

So, if a child plays with a robot, does it become a toy? Well, if a tree falls in the forest . . .

Let’s forget the purist definitions.

There are toy robot spiders. They are really cool.

Inside Adam Savage’s Cave: Awesome Robot Spider!


In addition to the animatronic spider, the Robugtix line includes a hexapod (6-legged) robot for those who are not “spider purists” demanding the full 8-legs of the “octopodal” arachnid.

[video] iitsii the Hexapod Robot

These animatronic devices are produced by Amoeba Robotics Ltd., a research, engineering, and design company.  Founded in 2010, this Hong Kong based concern focuses on “providing innovative robotics systems for professional and educational use.”  I can’t resist including another video of the “T8.” [video]

Watching these animatronic devices, you might pause to wonder what their working counterparts, the “robots,” must look like.  And there you might get a surprise.  Working robots, like their animatronic/entertainment counterparts, are being designed to resemble animals and even people.


As soon as engineers began developing sophisticated robotics, they ran into some problems.  You may have seen those sleek glass and metal robots from those 1950’s sci-fi movies.  In those days, there was an idea that robots would have to be, somehow, completely different from organic life forms.  And this idea carried over into early, “real-world” technology.  But there were problems.  These “unlife-like” robots didn’t work so well.

The reason was obvious.  Most often, we don’t need robots to do weird, strange, or superhuman tasks.  We really need robots that do exactly what human beings (and a variety of common animals and even insects) do. What’s more, the tasks we want robots to do aren’t necessarily complicated. Often we need robots that do common, everyday tasks. Tasks that are simple, but time consuming and repetitive,

So, for about the past decade, most robots have been developed to imitate animals and human beings.  And, not surprisingly, these robots are becoming more animatronic – life-like — in their movements and, even, appearance.

Sometimes, this is intended as in the Army Research Laboratory’s Robo-Raven. This aerial drone is designed to fly and maneuver with movements so much like a bird that it actually fools real birds. [image] [video]

The “animatronic” appearance and movement aren’t the result of idle tinkering.   Instead, it’s part of this aerial drone’s camouflage.  This particular “application” of camouflage is called mimesis or “masquerade.”  The goal is to create an aerial drone that the observer mistakes for — just a bird flying by.  But the bird is a flying drone relaying sound and video back to another, concealed observer. [video]. So, the “bird-watcher” is the one being watched.


Grossmann: Robo-Spider to the Rescue

6 March 2014

About the creepiest of all the land-roaming creatures is the spider.  But it only makes sense that, with 8 legs, the spider would be among the most sure-footed animals on earth.  And sure-footedness was just what researchers were looking for in a new search and rescue robot.  So, it’s no surprise that they picked the spider as their model.

Researchers at the Fraunhofer Institute for Manufacturing Engineering and Automation, in Stuttgart, Germany, have developed a new weapon in the search and rescue arena: a robotic spider.  In stark contrast to the real eight-legged arachnid’s creepy reputation, its robotic incarnation is not just your friend, but your potential rescuer.  These spiders can, for example, creep into the smallest spaces in collapsed buildings and provide location information about trapped victims as well as damage and air quality assessments.

These ‘bots are about as friendly and helpful as real spiders are not.  But, personality aside, in terms of the mechanics of operation, these robots are really a lot like the actual spider.  Research groups throughout the world have turned to biomimicry in designing the last few generations of robots.  That is, the robots of the future are being designed not just to imitate, but to function just like, plain old biological organisms.

In old sci-fi movies, robots of the future were visualized as almost anti-human and anti-organic.  In other words, they were made to look like everything that plain old living creatures weren’t — sporting sleek metal and glass surfaces with an enormous bulk and weight carried hither and yon on wheels.  However, when it came to building the real thing, these sleek-looking innovations turned out not to be . . . innovations.

We “organics” got the last laugh, most recently, watching Mars Rovers continually getting stuck because, even on slightly rough terrain, wheels don’t work as well as feet.  In terms of design, the bulky, weighty robots have been left behind because their very bulk and weight made it difficult for them to move freely and to perform flexibly enough to accomplish a wide variety of tasks.

For most current robotic applications, biomimicry, imitation of real-life organic creatures, is the order of the day.  And it makes sense.  Every time we see a really creepy spider, we worry that one of those creepy things will find its way into our home.  And they do — in spite of considerable obstacles.  What better creature to imitate if your goal is to rescue someone in a collapsed building?  Why try to figure out a new way to do what spiders have been able to do since . . . forever.

Even better, this spider can be produced inexpensively using a 3D printer. The resulting Robo-Spider is disposable (just in case you don’t want these critters crawling around the house after they’ve done their job).  Rather disturbingly, the Robo-Spider moves just like its biological counterpart and, if it has to, it can even jump — leaving me with vaguely disturbing mental image.

Creepy Robot Spider Crawls Out of 3-D Printer

3D-printed spider robot skitters where humans can’t

Also, BAE Systems, a British defense company, is working on robotic spiders (and dragonflies and snakes) to aid soldiers in combat zones.  These robo-spiders would play the role of scouts crawling through potentially dangerous areas and relaying precise reconnaissance information in situations too dangerous for human beings.

[image] [video]
The project has been taken seriously enough to garner a $38 million contract with the U.S. Army for the development of these robotic “octopeds.”

Robot Spiders, Dragonflies, Snakes to Aid Soldiers in War Zones


Grossmann: Crows – the Organic Aerial Surveillance Vehicle

26 December 2013

A very few of us can brag that we never forget a name, but a lot of people will tell you they never forget a face.  Crows don’t forget faces either.  You won’t remember their face, but they’ll remember yours . . . if, of course, you are a “person of interest” in their eyes.

Long before we started worrying about being watched by the intelligence community and long before the intelligence community was even planning to start watching anyone, “someone else” was watching both them and us and . . . just about everything else.  They were crows.

DARPA has, and will, invest millions of dollars in aerial surveillance and reconnaissance drones, like the Maverick and the developing Robo-Raven.  The agency hopes to “add” surveillance and reconnaissance equipment to these flying drone.  And, these drones will simulate the flight, movements, and appearance of actual birds, so well,  that they will be dismissed by viewers as passing birds.  However, neither DARPA nor their contractors may suspect that they aren’t adding anything to their mechanical birds that the real things don’t already have.

A group of birds of the genus. corvid, including crows and ravens, have every surveillance and reconnaissance ability of their mechanical imitators.  But the crows can do something more.  They interpret the data “on board” as it’s received.  And their abilities to recognize, interpret and communicate what they observe would be the envy of any and every DARPA contractor.

In other words, a pervasive group of flying eyes has been systematically watching us (and the rest of the world) since before the dawn or recorded history.  Crows?  Yes, just those ever-present large black birds.  They can recognize your face in a crowd of humans and, then, remember your face, communicate what they’ve seen and recognized to each other, and even plan and conspire.  What are they planning?  What are they conspiring about?  We don’t know.  Actually, it’s all pretty weird when you think about it.

The FBI is working on a type of biometrics called facial recognition.  Biometrics, generally, is the computer recognition of human characteristics such as voice, gate and facial features.  DARPA’s Robo-Raven may, someday, relay surveillance footage back to a computer, which will, then, attempt a facial identification of any human beings in the video feed.  But this robotic bird will be doing nothing the real thing wasn’t doing when the Egyptians were building the pyramids.

The crows’ ability to recognize and remember faces remained unproven until John M. Marzluff, a wildlife biologist at the University of Washington, decided to test the crows’ facial recognition abilities.  Why ?  After 20 years of working with these birds, which included capturing and tagging, it seemed that particular birds were “wary” of particular scientists and tried harder to avoid capture.  He had long wondered whether the birds could identify particular researchers.  Although the attempts to evade capture were little more than a minor annoyance, he decided it was worth a test.

Marzluff performed the test on the University’s Seattle campus.  He used Halloween-type masks, which were worn by researchers when they captured, tagged and released seven crows.  The birds certainly didn’t enjoy the capture and tagging.  The question: Would the birds remember the features of the masks worn by the taggers.

Later, two groups of volunteers strolled through the campus, one group wearing the same masks worn by the taggers and another group of wearing the same type of mask, but one with different features.  Did the crows remember?  Oh, did they.  While the volunteers wearing the different masks were ignored, those wearing the taggers’ mask were, first, yelled at.  The crows squawked and “scolded” the wearers of the offensive mask with obviously hostile calls.  However, as if to dispel all doubt about their displeasure, the crows dive-bombed a few of the volunteers but, again, only those wearing the offensive mask.

Let it be known that crows are not only vocal birds, but quite aggressive.  As David Dietle put in his excellent article on the amazing abilities of these birds, crows “hold a grudge.”  If you read Dietle’s full article you’ll realize that if you are on a crow’s bad side, . . . well . . . you may have something to worry about.

But Marzluff’s test revealed a few things no one expected.  When volunteers wore multiple types of masks and strolled through campus, in a group, the wearers of the taggers’ mask were singled out for scolding and dive-bombing by crows other than the original seven that had been captured and tagged. So, how did the other crows find out about the faces of the taggers?  Well, apparently, not by imitating or joining in the attack after the original seven began the festivities.  As amazing as it sounds, it’s likely that the attackers “heard’ about the facial features of their victims from the original seven.  Crow behavior implicates an amazing system of vocal communication that has never been well researched and is not well understood.

Apparently, crow calls are extremely diverse and demonstrate clear regional variation.  As David Dietle explains it, these birds have “dialects” – something almost inseparable from language.  Some crow calls have been interpreted to mean certain things in certain contexts, but sufficient studies have never been done to determine the extent of crow language.  Therefore, it’s impossible to estimate the degree of articulate communication between and among crows

Crows have usually large brains for their body size.  In fact, these birds have unusually large brains — period.  The size the crow brain is about the same as that of a chimpanzee.  Nathan Emery and Nicola Clayton, from the departments of animal behavior and experimental psychology at Cambridge University in England, have recently published a study in Science discussing evidence suggesting that the crow and its fellow corvids may have cognitive abilities that match those of chimpanzees and gorillas.

So, can crows not only remember your face, but describe you to other crows to the extent that these others will recognize you on sight?  Well, there’s a high probability that this amazing suggestion is true.  By the way, when the Seattle experiment was repeated with more normal-looking masks, the birds performed just as “well.”

The reader may be thinking, “If I offend a crow, maybe I can ‘lay low’ until all the crows, alive now, are dead.”  Sorry, that plan won’t work.  In the Seattle test, subsequent generations of crows, birds that had never seen the offending masks, recognized them and attacked the wearers in the complete absence of the original seven birds that had been tagged.  I don’t know about you, but this makes even me just a bit nervous . . . and I feed wild birds regularly.

In fact, certainly crows could give any elephant a run for the money when it comes to “never forgetting.”  There are numerous documented reports of whole flocks of crows avoiding homes, locations, even communities in which even a single crow has been killed.  It’s hard to believe that literally thousands of birds, for generations, would avoid a specific location on account of single death of one of their own, but they apparently do.

In evaluating these reports, David Dietle made an interesting observation.  If you mess with a crow, thousands of crows will remember your address for generations.  In other words, as many as a few hundred thousand crows will “know where you live.”

I can imagine some readers thinking, “Yes, crows do remember.”  “But if they don’t like me, they’ll just scold me and dive at me.”  “I can deal with that.”  And you’re probably right.  However, you should know some of the things crows can do – if they want to.

Consider a certain group of crows that loved nuts, but couldn’t crack the shells themselves.  These birds took the nuts to an intersection with a traffic light.  Spreading the uncracked nuts on the road, they waited until cars ran over the nuts and cracked the shells.  However, the crows didn’t fly out after each nut was cracked.  They waited and watched the traffic light.  When it displayed a signal that would stop traffic, the birds flew into the road and retrieved the nuts.  Then, took them to a safe location and ate their meal at their leisure.

Imagine what crows could do to their enemies if they really wanted to.  These birds are intelligent planners and communicators.  Also, they hold grudges and have long memories.  Oh, I forgot to add that these birds display great ingenuity and they are very, very patient.  Nervous yet?

As I said, I feed wild birds regularly.  I used to do this out of an empathic affection for wildlife.  But now, I look at these feedings as something more like payments of “protection money.”

A “must read”:

David Dietle, “6 Terrifying Ways Crows Are Way Smarter Than You Think”


Grossmann: The Bee’s Brain — the Green Brain Project

19 December 2013

Any new project intended to deal with decreasing honeybee populations should probably focus on CCD, Colony Collapse Disorder.  In case you’ve been shipwrecked on a desert island for about seven years or have been really busy, during that period of time, you’ve probably heard that honeybee populations throughout the world are declining at an alarming rate.  The cause of the current decline is unknown, but has a name, Colony Collapse Disorder.

There have been mass bee die-offs since the beginning of recorded history.  So, for the first few years of the current die-off, there was concern, but not alarm.  What’s special about CCD is, first, that it’s worldwide.  Historical bee die-offs have been local affairs.  Second, historical bee die-off’s have been brief.  Our present CCD isn’t stopping, and the bee numbers keep declining.  The one thing that CCD has in common with historical die-offs is that no one knows exactly what’s killing the bees.

It’s fashionable to blame pesticides, and almost everyone does, with a new article announcing the discovery of “the cause” appearing every few weeks.  It’s sort of like those monthly magazines that feature a “new” diet every month.  Each month, it’s “the last diet you’ll ever need.”  Then, the next month, a “new” last diet you’ll ever need . . . and on and on.  If you believe all this . . . well, we’d all be thin as rails and bees would have stopped dying off years ago.

Here’s the actual puzzle.  Bees are weakened, not by one thing, but a number of things happening at the same time.  Today’s bee suffers with fatigue from long distance transport, parasites and infections, exposure to pesticides, and these insects are even “drugged” for better performance — something like what’s done to athletes with steroids.

The problem is that no single one of these factors, alone, would kill a bee.  Worse, even all these factors, together, wouldn’t kill a bee.  Unfortunately, all these factors together will weaken a bee’s immune system to the point that the insect will contract and die from a completely unrelated disease.

So, finding the cause of CCD is a bit like solving the old puzzle called Rubrics Cube.  You have to line up combination after combination.  Except, unlike the puzzle, after you line up a particular combination, you have perform lengthy tests, again and again, until you find the “right” combination.  Sadly, long before the combination is found, the bees may be gone.  Of course, our current honeybees could be replaced, on the one hand, with a less efficient species imported from somewhere else or, on the other, with an efficient, but really mean and dangerous species (Africanized).

Strangely, the only thing that would take longer than finding the cause of CCD would be to build a robotic bee.  However, that’s what a lot of people are trying to do.  Robo-Bee is most definitely a thing of the future.  The best current prototype has just mastered a few seconds of controlled flight – then it crashes.  Prior to this, the old prototype would just take off and crash into the nearest wall without any maneuvers at all.

Robo-Bee’s crashes are even more discouraging when you realize that the current prototype is stabilized by a fixed wire.  Also, Robo-Bee “needs a cord.”  Figuratively speaking, you have to plug it into the wall, because there is no battery both powerful enough and lightweight enough that Robo-Bee can lift into the air.  But that’s not all it will have to lift.  Robo-Bee will also need an on-board flight computer to direct its flight.  Right now, no computer small enough and light enough exists.  And, if it did, there’s not even a prototype of a flight program that could successfully direct the small robotic insect in flight.

Discouraged yet?  Well, to their credit, the would-be developers of Robo-Bee aren’t the least discouraged.  And, as modest as Robo-Bee’s current performance may be, it’s an incredible achievement.  Only with the persistence of the project engineers have a host of seemingly impossible challenges and problems been met and solved.  However, it will be a long, long time before the first Robo-Bee rolls off the assembly line, flies into the fields, and begins pollinating.

And there will be even more challenges. We’ve only covered a few of the issues.  Even with an on-board computer to direct its flight, how will the Robo-Bee pollinate flowers?  Think about it.  To do so, these robots would have to see and smell.  They’d have to master the varied challenges of the pollination of each individual bloom.  To do that, these ‘bots would almost have to be able to . . . think.  How are they going to do that?  Well, the members of the Green Brain Project “are glad you asked them that question.”

I don’t know how long people have been thinking about the answer to that question but, about a year ago, researchers in Great Britain, specifically, at the Universities of Sheffield and Sussex decided to do something about it.

In an article describing the project, George Dvorsky, reports that, late last year, the Engineering and Physical Sciences Research Council (EPSRC) put up £1 million (USD $1,614,700) for the development and creation of the “first accurate computer simulation of a honey bee brain.”  However, when you consider the challenging goal of the project, even this “front money” is not so very much. After all, the project couldn’t afford the kind of computer muscle that would seem to be needed to tackle a job like this.  However, a creative solution to the computer problem has been provided by California’s NVIDIA.  That corporation will provide the project with a number of high-performance graphical processing units called GPU accelerators. This will allow the researchers to simulate aspects of a honey bee’s brain by using a large group of paralleled desktop PCs.  In other words, put together enough desktops and you can approximate some of the functions of a cluster of supercomputers, but at a fraction of the cost.

However, no matter how much or little money and equipment are involved, this part of the Robo-Bee project, building the bee’s mind, is an even more formidable challenge than building a robotic insect that just flies.  The mind of even an insect is breathtakingly complex, but the Green Project researchers are not trying to tackle the replication of the honeybee’s entire brain.  Instead they are focusing on only two functions: vision and the sense of smell.

Researchers are attempting to develop cognitive models of sight and smell.  To duplicate even part of an actual bee’s brain, you need to study an actual bee or, at least, work with someone who has.  That someone is Martin Giurfa of Toulouse, “an expert in all aspects of bee brain anatomy, physiology, and bee cognition and behavior.”  The ultimate goal is a robotic bee that can detect particular odors or particular flowers.  But, more immediately, the researcher are hoping to develop computer models of these processes that, someday, will be downloaded directly into the “brain” of a robotic bee.

However, the description above understates and ambition of one aspect of this project.  The researchers are attempting to develop models with true artificial intelligence.  That is, they are attempting to develop a computerized intelligence that will allow a robotic honeybee to act autonomously.  Put yet another way, these robotic bees would have the cognitive ability to perform certain basic tasks without pre-programmed instructions.

Such cognitive models are several steps beyond simple programming.  Successfully modeling the cognitive processes associated with vision and speech is essential to the development such artificial intelligence.  But why?

What do we think about?  We think about what we see, smell, hear, feel and taste.  Could a human intelligence ever be “designed” without senses and sensory input?  No.  So, in order to develop a real artificial intelligence — an intelligence that thinks, that intelligence must be “embodied” with those senses that provide the necessary sensory input (something to think about).

Simply put, the concept termed “embodiment,” as applied to robotics, “holds that any true artificial intelligence is impossible unless the robot has sensory and motor skills that connect it to the world.”  In other words, without sensory input, cognitive intelligence, as we know it, wouldn’t exist.

The envisioned final version of Robo-Bee will be able to think.

These, and other projects, reach far beyond our current technological abilities and promise innovations that are scarcely imaginable.  A thinking robotic bee is just one step away from more sophisticated social integrations leading, perhaps, to a thinking hive that would autonomously send certain robotic bees to certain locations, monitor honey and pollen reserves and so on.

There’s something both fascinating and frightening about thinking machines.  I must admit my mind wanders to sci-fi.  I can’t help thinking of the 1984 film, Terminator, in which the artificially intelligent Skynet initiates a planned extermination of the human race to allow intelligent machines to take over the world.

Imagine a bee version of Skynet, maybe, “Buzznet,” coordinating all activities of all beehives throughout the world.  Of course, scientists design “Buzznet” to “help” the few remaining organic honeybees.  Sure.  We know where this is going.

“Buzznet” promptly tries to wipe out all the remaining biological bees.  The few survivors will be herded into special detention hives, and come to depend on John . . . , no, Jane Connor.  (All active bees are female).

Of course, Buzznet will develop a robotic terminator bee and send it back through time to kill Jane’s queen and mother.  Because bees are not easy to find in crowded hives, the Terminator bee will just start killing them all.  In our film, this is the cause of CCD  — Terminator bees from the future.

As silly as all this sounds, I wonder . . . .

As technology advances to almost unimaginable frontiers including the development of artificial intelligences that operate independently of biological intelligence (in other words, independently of us), perhaps, a cautionary note is in order.  Maybe all scientists, technologists and engineers should be forced to take a course featuring 10 to 20 selected sci-fi movies in which good science goes bad.   The collection would include more than one film illustrating “what not to do.”   Or, at least, what not to do when developing artificial intelligence that operates independently of human intelligence.


Grossmann: The Bee’s Brain — the Zombie Apocalypse, but with Bees?

19 December 2013

The New York Times broke the story in late 2012.  There are zombie bees.  So, Night of the Living Dead might be a true story!?  Yeah, but with bees instead of people and . . . substantial script revisions.

If zombie bees were going to “appear” somewhere, California does seem like the most appropriate place.  Then, the zom-bees spread to Washington state.   But why did they avoid Oregon?   When they suddenly arrived in a third state, North Dakota – that seemed odd.   The zombie horror genre just hasn’t conditioned me to think of North Dakota as a sort of hotspot for zombie anything.  Still, the bees can go where they will.  If, as zombies, they still have their own will.

Anyway, the short answer: zombie bees are with us.

When you say you’re going to talk about zombies, the next question is, “What kind of zombies?”  It’s not so much that there are different varieties of zombies as different versions.  There are horror movie zombies, the zombies in folklore, and the real zombies – or at least “real” in the sense that a lot of people alive today absolutely believe in the reality of zombies.

On the top of the heap, in terms of popularity, is the Hollywood horror version of the brain-eating zombie.  However, many of the characteristics of these, oh, so familiar, zombies were made up by Hollywood writers.

Digging deeper, we reach the cultural folklore of zombies together with anthropological explanations of that folklore.  Many believe that what are taken to be zombies are persons who are drugged with a special concoction that, either by its very nature or through precision dosing, so depresses vital functions that the victim is mistaken for dead and buried.  The perpetrator, then, digs up the depressed, but still living body of the victim and either fools or drugs them into a life of servitude.

However, the true believers in zombies will tell you that specially trained and/or gifted “voodou” (voodoo) practitioners have the ability to reanimate a dead body and control it like a robot.  They believe that the victim’s soul, consciousness, or spirit has permanently departed, but their body remains as a biological robot under the complete control of its “bokor.”

What about our bee zombies?  Well, actually their zombification resembles none of the above.  However, the result is so reminiscent of the zombies of folklore that, perhaps, there no better and readily understandable term to describe what’s happening to the poor bee victims.

Unlike the zombie of actual tradition, the zombie bee falls victim to a parasitic fly, Apocephalus borealis.   The fly lays its eggs physically inside the bees body, the eggs affect the bees behavior not too unlike what was presented in the 1982 film, StarTrek: The Wrath of Khan, in which “indigenous eels” of Ceti Alpha V are introduced into the brains of the crew members, characters Chekov and Terrell, by the character Khan — maddened by his years in exile.  The film’s eels enter the ears of their victims and, reaching their brains, render them susceptible to mind control.

However, unlike Star Trek’s eels, the eggs and larvae of the Apocephalus borealis fly actually control the bee’s “mind” only briefly before causing its death.  Then, they consume the bee’s physical remains.  From another angle, the action of larvae in “eating their way out” of the dead bee’s body reminds one of another Hollywood creation, the mythical earwig.

The earwig is a real and mean-looking insect, but it doesn’t enter the human ear, burrow into the human brain and lay its eggs.  All of that was an old and almost forgotten “urban legend,” until it was featured in a 1972 episode or Rod Serling’s Night Gallery (Season 2, Episode 60, “The Caterpillar”).  However, even the apocryphal earwig had no ability to control the mind of the host.  So, zombification was not part of the earwig repertoire.

But the New York Times story that revealed zombie bees to the world, asked, “Whose in charge in [the bee’s] head”? Because the fly larvae, inside the bee’s body, directly affect the honeybee’s behavior in disturbingly zombie-like ways.  Under the influence of the developing fly larvae, the honeybee abandons its exclusively daytime routine and does something bees don’t do  — flies at night.  Just before, and during, this “last flight” into the night, the bee begins to move erratically.  And it ends its flight in death.  Only then, do the fly larvae eat their way out of the dead bee to continue their growth to maturity.

Hollywood has never quite dealt with this specific kind of zombification.  Of course, the zombie bee might be a good subject for a (not so) new and (not so) different kind of zombie movie.  Maybe the zombie creating flies enter the hive of apiarist, Ms. Red Queen, owner of Raccoon Apiary.  Realizing the problem, she uses an insecticide to kill all the possibility infected bees in a particular hive.  However, these particular flies are “mutants” and have laid mutant eggs in the bees’ bodies.  The larvae don’t grow to eat the infected bees, but reanimate them into murderous zombie bees worthy of any respectable (or not so respectable) Hollywood production.

One bee, Alice, is accidentally outside the hive (or something) during the spraying of insecticide.  She survives and re-enters the hive to discover zombified bees trying to escape and infect the apiary’s other hives.  She engages in a heroic struggle to contain the zombie bees and the infection they carry only to awaken from a coma outside the hive days later.  She sees only a single obviously dysfunctional bee flying repeatedly into a tree while repeating a message: “The dead buzz.”

Many sequels could follow.

Grossmann: The Nano Hummingbird – The Original Bird ‘Bot

12 December 2013

[Nano Hummer Video]

On 17 February 2011, DARPA announced the development of the first fully functional robotic bird. [1]  The “Nano Hummingbird” or, as it is also less imaginatively called, the “Nano Air Vehicle” (“NAV”), was the successful result of a project started in 2006 by AeroVironment, Inc. under the direction of DARPA. [1] Robots, by definition, must “do work.”  And the Nano-Hummer was the first fully functional bird-drone designed and able to perform surveillance and reconnaissance missions.

This robotic hummingbird can remain aloft for 11 minutes and attain a speed of 11 mph. [1]   With a skeleton of hollow carbon-fiber rods wrapped in fiber mesh, coated in a polyvinyl fluoride film, [5] and carrying “batteries, motors, and communications systems; as well as the video camera payload,” the robo-hummer weighs just .67 ounces. [1]

Designed to be deployed in urban environments or on battlefields, this drone is can “perch on windowsills or power lines” and even “enter buildings to observe and its surroundings” while relaying a continuous video back to its “pilot.” [video] [1]

In terms of appearance, the Nano-Hummer was, and is, quite like a hummingbird.    Although larger than the typical hummingbird, Nano-Hummer, is well within the size range of the species and is, actually, smaller than the largest of real hummingbirds. [1]   With a facade both shaped and colored to resemble the real bird, the Nano-Hummer presents the viewer with a remarkable likeness of a hummingbird. [1]

The Nano-Hummer isn’t stealth in the sense of evading radar.  Nor is it “cryptic,” that type of camouflage that blends, or disappears, into the surrounding terrain.  Rather, with the appearance of a hummingbird, the designers used a type of camouflage called “mimesis,” also termed “masquerade,” as concealment.  A camouflaged object is said to be “masqueraded” when the object can be clearly seen, but looks like something else, which is of no special interest to the observer.  And such camouflage is important to a mini-drone with the primary purpose of surveillance and reconnaissance. [1]

Designing this drone on the “hummingbird model,” however, was not done only for the purpose of camouflage.  The project’s objective included biomimicry, that is, biologically inspired engineering. [8] With the hummingbird, its amazingly diverse flight maneuvers were the object of imitation.  However, UAV’s head researcher, Matt Keennon, admits that a perfect replica of what “nature has done” was too daunting. [5]  For example, the Nano-Hummer only beats its wings 20 times a second, which is slow motion compared to the real hummingbird’s 80 beats per second. [video] [5]

Whatever the technical shortfalls, this bird-bot replicates much of the real hummingbird’s flight performance. [5]  Not only can it perform rolls and backflips [video] but, most important of all, it can hover like the real thing. [video] [5]  Part of the importance of the ability hover relates to its reconnaissance and surveillance functions.  Hovering allows the video camera to select and observe stationary targets.  However, the “hover” of both hummingbirds and bees attracts so much attention from developers of drone technology because it assures success in the most difficult flight maneuver of all — landing.  In fact, landing is the most complex part of flight, and the maneuver most likely to result in accident or disaster.

When landing, a flying object must attain the slowest speed possible before touching down.  Hovering resolves the problem neatly by assuring that the robot can stop in midair and, therefore, touch the ground or perch as zero speed.  Observe the relatively compact helicopter landing port in contrast to the long landing strip required by an airplane which must maintain forward motion when airborne.

The drone has a remarkable range of movement in flight much like the real hummingbird. [1] Nano-Hummer “can climb and descend vertically; fly sideways left and right; forward and backward; rotate clockwise and counter-clockwise; and hover in mid-air.” [1]  Both propulsion and altitude control are entirely provided by the drone’s flapping wings. [video] [1]

This remote controlled mini-drone can be maneuvered by the “pilot” without direct visual observation using the video stream alone. [1] With its small camera, this drone can relay back video images of its location. [1] The camera angle is defined by the drone’s pitch.  In forward motion, the camera provides a continuous view of the ground.  Hovering provides the best camera angle for surveying rooms. [video] [5]

To DARPA, it was particularly important that this drone demonstrate the ability to hover in a 5 mph side-wind without drift of more than one meter. [1]  The CIA’s “insectothopter,” a robotic dragonfly was developed in the 1970’s. [image] [3] This unmanned aerial vehicle “was the size of a dragonfly, and was hand-painted to look like one.” [3]  Powered by a small gasoline engine, the insectothopter proved unusable due to its inability to withstand even moderate wind gusts. [video] [3]

The Nano-Hummingbird was named by Time Magazine as one of the 50 best inventions of 2011 [4] and has paved the way for the development of a whole generation of bird inspired ‘bots, including Prioria’s “Maverick,” [image] [video] and, the even more “bird-like,” Robo-Raven, which is still in development by the Army Research Laboratory. [image 1] [video] [video] Also, the development of this first small bird-bot brought the U.S. Air Force one step closer to one of the goals on its wish list: “flocks of small drones.” [7]

A flock of small drones sounds really cool – as long as the flock isn’t after me.