The cyborg animal spies hatching in the lab

[WhiteFeather]

Read past history of this, its not science fiction......any more.

WATCH VIDEO off web site

http://technology.newscientist.com/c...n-the-lab.html

The cyborg animal spies hatching in the lab

THE next time a moth alights on your window sill, watch what you say. Sure, it may look like an innocent visitor, irresistibly drawn to the light in your room, but it could actually be a spy - one of a new generation of cyborg insects with implants wired into their nerves to allow remote control of their movement. Be warned, flesh-and-blood bugs may soon live up to their name.

It's not just insects that could be used as snoops. Researchers have already developed remote control systems for rats, pigeons and even sharks. The motivation is simple: why labour for years to build robots that imitate the ways animals move when you can just plug into living creatures and hijack systems already optimised by millions of years of evolution? "There's a long history of trying to develop micro-robots that could be sent out as autonomous devices, but I think many engineers have realised that they can't improve on Mother Nature," says insect neurobiologist John Hildebrand at the University of Arizona in Tucson. Furthermore, animals' sensory abilities far outstrip the vast majority of artificial sensors. Sharks, moths and rats, for example, have amazing olfactory systems that allow them to detect the faintest traces of chemicals. And if you can hide your control system within your cyborg's body, it would be virtually indistinguishable from its unadulterated kin - the perfect spy.

José Delgado at Yale University created the first cyborg animal in the 1950s. Delgado discovered where to insert electrodes in the brains of several species, including bulls, to acquire crude control of their movement. In one dramatic demonstration in 1963, he stood in a bullring in Córdoba, Spain, as one of his cyborg bulls charged at him. With just seconds between him and a good goring, Delgado flicked a switch and the bull skidded to a halt.

The cyborg concept drifted back into science fiction for a few decades, until 2002 when a team announced that they had developed a cyborg rat whose movement could be controlled remotely (New Scientist, 4 May 2002, p 6). The team led by John Chapin at the State University of New York Health Science Center in Brooklyn, implanted electrodes in the rat's brain which apparently mimicked the sensation that its left or right whiskers had been brushed. They then trained the rat to respond to the electrical stimuli. For example, if the rat turned right when the brain region associated with its right whiskers was pulsed, then reward centres in its brain were electrically stimulated.

Linda Hermer-Vasquez at the University of Florida in Gainesville later joined the project to train the cyborg rat to identify specific scents, such as humans or explosives, to demonstrate that it could be used in search-and-rescue missions to find people trapped under rubble, for example, or to sniff out bombs.

To give the animal's operator a rat's-eye view, the most advanced generation of cyborg rats were kitted out with video-camera backpacks. These souped-up rats were trained to board a rolling carrier so that they could be easily transported to the site of their mission.

To test the system, the team allowed a rat to descend from the carrier and remotely steered it to the area they wanted searched for traces of an explosive. Once in the correct area, they switched off their remote control. "When the rat realised that it was no longer being controlled, it went into odour-sniffing mode," says Hermer-Vasquez. Within a few minutes, the rat had successfully identified the source of the scent. They repeated the test several times, with the same result.

"The team steered a rat by remote-control to the area they wanted searched for traces of explosives"It is easy to see if a rat in the lab has found the source of a target scent, but in the field the team would need a way of getting remote confirmation that their cyborg rat had accomplished its mission and found a target. To achieve this, they first identified the neural signals that rats generate when they come across a scent they are trained to find. The researchers began to develop another brain implant to pick up these neural signals and a transmitter attached to the backpack to relay them to "mission control" (New Scientist, 25 September 2004, p 21).

Unfortunately, progress slowed in 2005 when the US Defense Advanced Research Projects Agency (DARPA) stopped funding the research. However, Hermer-Vasquez says the Israeli government has since asked the team to apply for a patent in Israel so the government can license the technology to use cyborg rats in search-and-rescue missions. The patent is pending.

Chapin's team now plans to transfer its technology to birds, where it could be useful for surveillance. They won't be the first, though. Last year, a team led by Su Xuecheng at the Shandong University of Science and Technology in Qingdao, reported implanting electrodes in the brains of pigeons which allowed them to direct the birds up, down, left or right via wireless signals from a laptop. In another, now-defunct DARPA project, Jelle Atema of Boston University, Massachusetts, showed he could control the direction of a small shark - called a spiny dogfish - using a brain implant to stimulate either its left or right olfactory centre. The shark turned to follow the phantom odour (New Scientist, 1 March 2006, p 30).

One benefit of working with animals like rats, pigeons and sharks is that they are big enough to carry off-the-shelf miniature video cameras and computers - and the batteries to power them. But these animals are are too large to blend into the background, which is one reason why DARPA's latest project focuses on insects. Insects' agility in flight is unmatched. The Hybrid Insect Micro-Electro-Mechanical Systems project (HI-MEMS) aims to miniaturise all the technology necessary so that it fits within the body of a flying insect.

During their development from larvae to adults, most winged insects, including moths and beetles, undergo metamorphosis in a pupa stage, during which enzymes dissolve much of the larval tissue and the insect is rebuilt. The HI-MEMS project aims to merge artificial control systems with those of the insect by inserting the devices during the pupa stage. The idea is that as new organs and tissue develop, they will create strong, stable connections between the devices and the insects' neural or muscular tissues. The control devices become part of the adult insect's body.

This is no mean feat. The implants must be small and light enough not to interfere with the adult insect, and must reliably contact the neurons that control flight. HI-MEMS researchers have fabricated ultra-thin neural probes - a few hundred of micrometres across - out of flexible plastic, with traces of metal completing the electrical connections.

Details are hard to come by as the DARPA-funded research teams have been prohibited from speaking about their work. However, two of the groups gave tantalising glimpses of their progress when they presented some results at the IEEE MEMS 2008 conference in Tucson, Arizona, in January.

In a series of video clips shown at the conference and posted online, a tobacco hawkmoth with wires connected to its back lifts and lowers one wing, then the other, then both, in response to signals delivered to its flight muscles. As the researchers ramp up the frequency of the muscle stimulation, the moth's wings beat faster, approaching take-off speed. In another clip, the moth is flying, tethered from above, when electrical impulses applied to muscles on one side or the other cause the moth to yaw left or right.

The clips were filmed at the Boyce Thompson Institute in Ithaca, New York, where a team led by David Stern implanted the flexible plastic probes into tobacco hawkmoth pupae seven days before the moths emerged. They found that inserting them any earlier meant the tissue was too fluid to seal around the probe, but any later and development was too advanced and the probes damaged the moths' muscles. A probe is embedded in each set of flight muscles on either side of the moth and a connection protrudes from the moth's back. This can be hooked up to the tether wires which also deliver control signals and power. According to their paper, the team has also designed and built a battery-powered onboard control system, though there was no mention of whether they had made moths fly using this untethered arrangement.

Meanwhile another DARPA-funded group, led by Michel Maharbiz at the University of California, Berkeley, implanted electrodes into the brains of adult green June beetles, near neurons that control flight (see Illustration). When the team delivered pulses of negative voltage to the brain, the beetles' wing muscles began beating and the bugs took off. A pulse of positive voltage shut the wings down, stopping flight short, and by rapidly switching between these signals, they controlled the insects' thrust and lift. Maharbiz's team found two ways to make tethered beetles turn. In one, they mounted an LED display in front of a beetle's eyes. Lighting up the left or right portion turned the beetle in the opposite direction. In a second, more successful, approach they directly stimulated the flight muscles on one side, causing the insect to turn to the other.

Intriguing as these first steps are, these programmes are a long way from providing the ultimate in stealthy spies. Maharbiz's system uses a battery glued to the outside of the beetle for power, while Stern's moth-control system relies on power provided through wires plugged into the implant. Both would stick out like sore thumbs, and that's before adding the microphones, environmental sensors and transmitters that they would need to be of any use as spies. The challenge now is to shrink the components to hide as many of them as possible inside the insect. They are also looking to harness power from the insects themselves.

DARPA's goal is to create cyborg insects that can fly at least 100 metres from their controller and land within 5 metres of a target, then stay put until commanded to buzz off again. How the insects will be guided to a target is yet another unresolved problem. "There were a bunch of ideas," says Charles Higgins at the University of Arizona, who was involved in DARPA's original brainstorming session for the HI-MEMS project. One was to use radio control to guide the moth, though that would mean emitting radio signals, which could be detected by the enemy. A second was to use GPS signals to guide the insect to its goal, and a third was to point the moth in the right direction and send it off with a pre-programmed series of instructions - for example, fly straight for 50 metres, then circle.

Maharbiz's flight controller works on the latter principle, storing sequences of flight commands which would direct the insect through a preset series of moves. "What you want to avoid is some way of detecting that it's not a plain old insect, or some situation where its signals could be jammed," Higgins notes. The pre-programmed method has its shortcomings, though: a strong wind could easily blow the bug off course.

Another variable is deciding at what level to engage the spy-in-the-sky's nervous system. Stern's probes stimulate the moth's flight muscles directly, while Maharbiz targets the neurons associated with flight in the beetle brain as well as its muscles. A major advantage of using an insect is the ability to tap into its own control system - its ability to steer a straight course, or level itself, for instance - so stimulating each wingbeat manually would defeat the object.

While other teams are trying to stuff insects full of technology, Higgins has gone in the opposite direction by trying to stuff technology full of insects: he wants to recruit insects as high-level sensory controllers for robots. Plugging robots into live insects seemed like a no-brainer once the idea struck him. "I was in the process of designing a new vision chip that would cost about $60,000 to make," he says. "It occurred to me that I was spending $60,000 to mimic a tiny portion of the fly's brain, when I could get better capabilities in a moth for $4."

Artificial vision can't beat living systems, which are honed to recognise objects or detect motion, Higgins says. So far, he has begun experimenting with insects as living sensory-systems within a three-wheeled table-top robot (see "Meet the robo-bug").

Research into cyborg animals has delivered results that have intrigued researchers in other fields. The fact that Maharbiz's group can turn insect flight on and off by stimulating the brain with voltage pulses of opposite polarity is surprising, says Michael Dickinson at the California Institute of Technology in Pasadena. "I have no sensible neurobiological explanation for why that's the case," he says.

If the groups keep making strides, the proverbial fly on the wall may literally become a spy. So, the paranoid out there, beware. It may be time to arm yourselves with a can of bug spray and a good old-fashioned swatter - just to be on the safe side.

[slicknick131]

If this is the case, I hope they market them so I can send one into the Pittsburgh Penguins' locker room.

Oh yeah.