From the archive, originally posted by: [ spectre ]
DRAGONFLIES NEVER FLY IN A PACK
Are We Being Watched by Flying Robot Insects?
BY John Mahoney / 10/2007
Now, this is pretty weird. Rumors have been floating around the Net
for a while now speculating on whether or not tiny, dragonfly-like
robots have been covertly monitoring recent political demonstrations
and protests around Washington, D.C., and New York. Numerous
protesters at multiple events have reported seeing the helicopter-like
insectoid entities, fueling suspicion that something sneaky was afoot.
Yesterday the Washington Post brought the story mainstream in the
interests of solving the case. What did they discover?
Basically, if the claims are true, someone has made great (and secret)
strides in the field of robots capable of mimicking insectoid flight-
something that’s currently incredibly difficult to do. Research teams
at universities across the country including Caltech, Vanderbilt and
Harvard, are all trying to reproduce insect-like flight in a man-made
robot-some of them even for the intended purpose of surveillance.
None, however, have gotten anywhere near the sophistication required
to engineer the minuscule, agile critters described by the protesters.
The Post also mentioned one of Darpa’s creepiest programs: the race to
embed microchips in the brains of moths and butterflies in the pupal
stage that will eventually fuse to the more developed adults’ brains,
enabling control over their actions. PopSci reported on this and other
ways that bugs are being enlisted for defense earlier this year.
Thankfully, said program is many, many years from realization.
As of now, no one has netted one of these mysterious creatures for
study. Here’s hoping a Homeland Security robo-bug pilot gets sleepy
sometime soon and allows one to be captured. Until then, protesters,
keep your eyes to the sky.
Bugging Out on Homeland Security
Wings, antennae and scales may be our best weapons yet against toxins
By Abby Seiff / March 2007
Annoying as they are, you may want to think twice before you crush a
cockroach or swat a fly-you could be killing a future foot soldier in
the war on terror. Increasingly, scientists are turning to insects and
other creatures for better ways to identify biohazards.
“Cockroaches can detect all kinds of things, from anthrax spores to
DNA,” says Karen Kester, an entomologist at Virginia Commonwealth
University. With $1 million in funding from the Pentagon’s Defense
Advanced Research Projects Agency (Darpa), Kester is studying ways to
use roaches and houseflies as toxin sentinels inside contaminated
buildings or subways. This, of course, spares humans the job, and it
may prove more effective than mechanical sensors, which often lack the
range and sensitivity of their living counterparts.
Bees and fish are also in demand. A small British biotechnology firm
called Inscentinel is employing the finely tuned olfactory system of
bees to sniff for explosives. And New York, California and Maryland
are exploiting the highly sensitive nervous system of bluegill fish to
test for toxins in municipal water supplies. Bill Lawler, co-founder
of Intelligent Automation Corporation, the California company that
sells the bluegill-monitoring system, says living sensors are “the
wave of the future.” So go easy on the Raid.
Micro Inertial Navigation Technology (MINT)
Program Manager and PoC: Dr. Amit Lal
Document Type: Presolicitation Notice
Solicitation Number: BAA07-53
Posted Date: Jun 29, 2007
Original Response Date: Sep 18, 2007
Current Response Date: Sep 18, 2007
Original Archive Date: Jul 03, 2008
Current Archive Date: Jul 03, 2008
Classification Code: A — Research & Development
Naics Code: 541710 — Research and Development in the Physical,
Engineering, and Life Sciences
DARPA is soliciting innovative research and development (R&D)
proposals in the area of Micro Inertial Navigation Technology (MINT).
Proposed research should investigate innovative approaches that enable
revolutionary advances in science, devices, or systems. Specifically
excluded is research that primarily results in evolutionary
improvements to the existing state of practice.
The overall goal of this program is to create navigation sensors that
use secondary inertial variables, such as velocity, to enable long
term precise navigation that a traditional IMU equipped only with
accelerometers and gyros cannot accomplish. This program seeks the
creation of micro and nano scale low-power navigation sensors that
allow long term (hours to days) GPS denied precision navigation. The
sensors should be small enough to be placed in small compartments,
such as the shoe sole or small UAVs, where zero relative velocity
measurement may be accomplished with high accuracy. The power
consumption of the sensors should be small, ideally compatible with
energy harvested during locomotion, so that the weight load due to
batteries powering the sensors is minimized. The sensor technology
should work over wide temperature range and shock environment that is
typical in DOD applications
DARPA Broad Agency Announcement (BAA) No. 07-53, entitled “Micro
Inertial Navigation Technology (MINT),” is provided as an attachment
to this solicitation notice and includes information on the specific
areas of interest; the submission process; abstract and proposal
formats; evaluation and selection/funding processes; as well as all
other pertinent administrative and contractual information. The BAA
may be obtained from the FedBizOpps website: http://www.fedbizopps.gov/,
Grants.gov website: http://www.grants.gov/, World Wide Web (WWW) at
URL http://www.darpa.mil/mto (go to “solicitation” area) or by fax,
electronic mail, or mail request to the administrative contact address
given below. Proposals not meeting the format or following the
submission instructions described in the BAA may not be reviewed.
A “Proposer’s Questions,” website will be posted for BAA 07-53 on the
DARPA, Microsystems Technology Office solicitations page
(www.darpa.mil/baa/#eto). If you would like to have a question
answered and posted on this site, please send your question to the
following address: BAA07-53 [at] darpa [dot] mil [dot]
Amit Lal, DARPA Program Manager,
Phone 571-218-4682, Fax 703-248-1933,
Email amit [dot] lal [at] darpa [dot] mil
Microsystems Technology Office (MTO)
T: (571)218-4682 / F: (703)696-2206
Michael Blackstone, Contracting Officer,
Phone (571) 218-4804, Fax (703) 696-2208,
Email michael [dot] blackstone [at] darpa [dot] mil
Dragonfly or Insect Spy? Scientists at Work on Robobugs
By Rick Weiss / October 9, 2007
Vanessa Alarcon saw them while working at an antiwar rally in
Lafayette Square last month.
“I heard someone say, ‘Oh my god, look at those,’ ” the college senior
from New York recalled. “I look up and I’m like, ‘What the hell is
that?’ They looked kind of like dragonflies or little helicopters. But
I mean, those are not insects.”
Out in the crowd, Bernard Crane saw them, too.
“I’d never seen anything like it in my life,” the Washington lawyer
said. “They were large for dragonflies. I thought, ‘Is that
mechanical, or is that alive?’ ”
That is just one of the questions hovering over a handful of similar
sightings at political events in Washington and New York. Some suspect
the insectlike drones are high-tech surveillance tools, perhaps
deployed by the Department of Homeland Security.
Others think they are, well, dragonflies — an ancient order of
insects that even biologists concede look about as robotic as a living
creature can look.
No agency admits to having deployed insect-size spy drones. But a
number of U.S. government and private entities acknowledge they are
trying. Some federally funded teams are even growing live insects with
computer chips in them, with the goal of mounting spyware on their
bodies and controlling their flight muscles remotely.
The robobugs could follow suspects, guide missiles to targets or
navigate the crannies of collapsed buildings to find survivors.
The technical challenges of creating robotic insects are daunting, and
most experts doubt that fully working models exist yet.
“If you find something, let me know,” said Gary Anderson of the
Defense Department’s Rapid Reaction Technology Office.
But the CIA secretly developed a simple dragonfly snooper as long ago
as the 1970s. And given recent advances, even skeptics say there is
always a chance that some agency has quietly managed to make something
“America can be pretty sneaky,” said Tom Ehrhard, a retired Air Force
colonel and expert in unmanned aerial vehicles who is now at the
Center for Strategic and Budgetary Assessments, a nonprofit Washington-
based research institute.
Robotic fliers have been used by the military since World War II, but
in the past decade their numbers and level of sophistication have
increased enormously. Defense Department documents describe nearly 100
different models in use today, some as tiny as birds, and some the
size of small planes.
All told, the nation’s fleet of flying robots logged more than 160,000
flight hours last year — a more than fourfold increase since 2003. A
recent report by the U.S. Army Command and General Staff College
warned that if traffic rules are not clarified soon, the glut of
unmanned vehicles “could render military airspace chaotic and
But getting from bird size to bug size is not a simple matter of
making everything smaller.
“You can’t make a conventional robot of metal and ball bearings and
just shrink the design down,” said Ronald Fearing, a roboticist at the
University of California at Berkeley. For one thing, the rules of
aerodynamics change at very tiny scales and require wings that flap in
precise ways — a huge engineering challenge.
Only recently have scientists come to understand how insects fly — a
biomechanical feat that, despite the evidence before scientists’ eyes,
was for decades deemed “theoretically impossible.” Just last month,
researchers at Cornell University published a physics paper clarifying
how dragonflies adjust the relative motions of their front and rear
wings to save energy while hovering.
That kind of finding is important to roboticists because flapping
fliers tend to be energy hogs, and batteries are heavy.
The CIA was among the earliest to tackle the problem. The
“insectothopter,” developed by the agency’s Office of Research and
Development 30 years ago, looked just like a dragonfly and contained a
tiny gasoline engine to make the four wings flap. It flew but was
ultimately declared a failure because it could not handle crosswinds.
Agency spokesman George Little said he could not talk about what the
CIA may have done since then. The Office of the Director of National
Intelligence, the Department of Homeland Security and the Secret
Service also declined to discuss the topic.
Only the FBI offered a declarative denial. “We don’t have anything
like that,” a spokesman said.
The Defense Department is trying, though.
In one approach, researchers funded by the Defense Advanced Research
Projects Agency (DARPA) are inserting computer chips into moth pupae
— the intermediate stage between a caterpillar and a flying adult —
and hatching them into healthy “cyborg moths.”
The Hybrid Insect Micro-Electro-Mechanical Systems project aims to
create literal shutterbugs — camera-toting insects whose nerves have
grown into their internal silicon chip so that wranglers can control
their activities. DARPA researchers are also raising cyborg beetles
with power for various instruments to be generated by their muscles.
“You might recall that Gandalf the friendly wizard in the recent
classic ‘Lord of the Rings’ used a moth to call in air support,” DARPA
program manager Amit Lal said at a symposium in August. Today, he
said, “this science fiction vision is within the realm of reality.”
A DARPA spokeswoman denied a reporter’s request to interview Lal or
others on the project.
The cyborg insect project has its share of doubters.
“I’ll be seriously dead before that program deploys,” said vice
admiral Joe Dyer, former commander of the Naval Air Systems Command,
now at iRobot in Burlington, Mass., which makes household and military
By contrast, fully mechanical micro-fliers are advancing quickly.
Researchers at the California Institute of Technology have made a
“microbat ornithopter” that flies freely and fits in the palm of one’s
hand. A Vanderbilt University team has made a similar device.
With their sail-like wings, neither of those would be mistaken for
insects. In July, however, a Harvard University team got a truly fly-
like robot airborne, its synthetic wings buzzing at 120 beats per
“It showed that we can manufacture the articulated, high-speed
structures that you need to re-create the complex wing motions that
insects produce,” said team leader Robert Wood.
The fly’s vanishingly thin materials were machined with lasers, then
folded into three-dimensional form “like a micro-origami,” he said.
Alternating electric fields make the wings flap. The whole thing
weighs just 65 milligrams, or a little more than the plastic head of a
Still, it can fly only while attached to a threadlike tether that
supplies power, evidence that significant hurdles remain.
In August, at the International Symposium on Flying Insects and
Robots, held in Switzerland, Japanese researchers introduced radio-
controlled fliers with four-inch wingspans that resemble hawk moths.
Those who watch them fly, its creator wrote in the program, “feel
something of ‘living souls.’ ”
Others, taking a tip from the CIA, are making fliers that run on
chemical fuels instead of batteries. The “entomopter,” in early stages
of development at the Georgia Institute of Technology and resembling a
toy plane more than a bug, converts liquid fuel into a hot gas, which
powers four flapping wings and ancillary equipment.
“You can get more energy out of a drop of gasoline than out of a
battery the size of a drop of gasoline,” said team leader Robert
Even if the technical hurdles are overcome, insect-size fliers will
always be risky investments.
“They can get eaten by a bird, they can get caught in a spider web,”
said Fearing of Berkeley. “No matter how smart you are — you can put
a Pentium in there — if a bird comes at you at 30 miles per hour
there’s nothing you can do about it.”
Protesters might even nab one with a net — one of many reasons why
Ehrhard, the former Air Force colonel, and other experts said they
doubted that the hovering bugs spotted in Washington were spies.
So what was seen by Crane, Alarcon and a handful of others at the D.C.
march — and as far back as 2004, during the Republican National
Convention in New York, when one observant but perhaps paranoid peace-
march participant described on the Web “a jet-black dragonfly hovering
about 10 feet off the ground, precisely in the middle of 7th
avenue . . . watching us”?
They probably saw dragonflies, said Jerry Louton, an entomologist at
the National Museum of Natural History. Washington is home to some
large, spectacularly adorned dragonflies that “can knock your socks
off,” he said.
At the same time, he added, some details do not make sense. Three
people at the D.C. event independently described a row of spheres, the
size of small berries, attached along the tails of the big dragonflies
— an accoutrement that Louton could not explain. And all reported
seeing at least three maneuvering in unison.
“Dragonflies never fly in a pack,” he said.
Mara Verheyden-Hilliard of the Partnership for Civil Justice said her
group is investigating witness reports and has filed Freedom of
Information Act requests with several federal agencies. If such
devices are being used to spy on political activists, she said, “it
would be a significant violation of people’s civil rights.”
For many roboticists still struggling to get off the ground, however,
that concern — and their technology’s potential role — seems
“I don’t want people to get paranoid, but what can I say?” Fearing
said. “Cellphone cameras are already everywhere. It’s not that much
e-mail : fir [at] epfl [dot] ch
Micromechanical Flying Insect (MFI) Project The goal of the
micromechanical flying insect (MFI) project is to develop a 25 mm
(wingtip-to-wingtip) device capable of sustained autonomous flight.
The MFI is designed based on biomimetic principles to capture some of
the exceptional flight performance achieved by true flies. The high
performance of true flies is based on large forces generated by non-
steady state aerodynamics, a high power-to-weight ratio motor system,
and a high-speed control system with tightly integrated visual and
inertial sensors. Our design analysis shows us that piezoelectric
actuators and flexible thorax structures can provide the needed power
density and wing stroke, and that adequate power can be supplied by
lithium batteries charged by solar cells.
The MFI project started in May 1998. In the first 3 years of this MURI
grant, research concentrated on understanding fly flight aerodynamics
and on analysis, design and fabrication of MFI actuators, thorax and
wings. In August 2001, our MFI prototype (with 1 wing) showed thrust
forces on a test stand. In September 2002, we switched our fabrication
from folded stainless steel to carbon fiber. In March 2003 we
demonstrated 500 microNewtons of lift from a single wing on a test
stand. Since March 2003, we have been working on reducing weight,
increasing actuator power density, increasing air frame strength, and
improving wing control.
Overview of MFI project from UCB Public Affairs office:
Currently work on MFI flight control and sensors is sponsored by NSF
“Robotic Insect Flight Stabilization Using Biomimetic Sensors”, Jan
2005-. NSF Disclaimer: “This material is based upon work supported by
the National Science Foundation under Grant No. IIS-IIS-0412541. Any
opinions, findings and conclusions or recommendations expressed in
this material are those of the author(s) and do not necessarily
reflect the views of the National Science Foundation (NSF).”
Formerly sponsored by ONR MURI Biomimetic Robotics and DARPA
Controlled Biological Systems Program 1998-2003.
Military Builds Robotic Insects
by David Hambling / January 24, 2007
Published by Wired
If you feel something crawling on your neck, it might be a wasp or a
bee. Or it might be something much more dangerous.
Israel is developing a robot the size of a hornet to attack
terrorists. And although the prototype will not fly for three years,
killer Micro Air Vehicles, or MAVs, are much closer than that.
British Special Forces already use 6-inch MAV aircraft called WASPs
for reconnaissance in Afghanistan. The $3,000 WASP is operated with a
Gameboy-style controller and is nearly silent, so it can get very
close without being detected. A new development will reportedly see
the WASP fitted with a C4 explosive warhead for kamikaze attacks on
snipers. One newspaper dubbed it “The Talibanator.”
Fred Davis, technical director of the Assessment and Demonstrations
Division of the Air Force Research Laboratory Munitions Directorate at
Eglin Air Force Base in Florida, confirmed that the United States has
ambitious plans for future micro-munitions, which he says will be
pocket-sized with mission-specific payloads.
“You’re not going to be knocking down walls,” says Davis. “What we’re
looking at is functional defeat.”
This means preventing the target from carrying out its mission, rather
than destroying it, Davis says. A truck, for example, can be put out
of action by destroying its tires; a MAV can do this by squirting them
with few milliliters of a catalytic de-polymerization agent, causing
them to disintegrate rapidly.
Davis sees future MAVs landing and hopping or crawling on the ground
like insects, enabling them to get inside buildings. Once inside, an
entire command center can be disabled by targeting the power supply.
“You could short out the circuit box,” says Davis.
The MAV could do this by physically crawling inside like a wayward
squirrel, or it might release a cloud of metal-coated fibers —
similar to the “soft bombs” the Air Force used to shut down power
stations in Kosovo with a cloud of conductive whiskers. Such fibers
could effectively destroy PCs and other electronic gear as well as
interrupting power to a building.
But what about attacking people? The smallest munitions ever used by
the Air Force were “gravel mines” or “button bombs” dropped by the
millions in the Vietnam war, some weighing just a quarter of an ounce.
A crawling MAV could deliver this type of bomb to the victim’s most
Or, as Davis suggests, the tiny vehicle itself might be the warhead.
“You can make the structure of the craft out of reactive (explosive)
material,” he says. Any unused fuel can add to the blast, a technique
already used in some surface-to-air missiles, and the explosion would
convert the rest of the MAV into lethal shrapnel.
Others have suggested “fire-ant warfare” with tiny robots that can
only do limited damage individually, but have enough cumulative effect
to overwhelm an opponent.
Poison needles or stings have also been proposed. Treaty obligations
would prevent the military from using this approach, but the CIA
developed lethal needles using shellfish toxin in the 1950s, and the
technology is on the shelf.
Of course, the bad guys can use Micro Air Vehicles too.
“After some development time, many countries would produce them,”
warns Juergen Altmann, a physicist at Dortmund University, working in
assessment of new military technologies. Indiscriminate use would
cause many civilian casualties — and they could end up in the hands
“Big dangers can ensue from terrorists,” Altmann says. “For instance,
using MAVs with small explosive charges to assassinate high-level
politicians or to transport biological/chemical agents into protected
To prevent this danger, Altmann advocates an international ban on
armed MAVs, similar to the ban on landmines. Until then, development
will proceed apace.