How researchers and engineers are tapping nature’s wealth of well-tested designs.
The animal kingdom is packed with talent surpassing that of humans. A bloodhound’s sense of smell, for example, is so keen that it can detect a fugitive’s scent days later and distinguish it from those of dozens of innocents that may have travelled the same path in the meantime. Rattlesnakes and other pit vipers can discern and home in on the body heat of their prey, even with their other senses masked. A variety of creatures can perceive sounds beyond the normal range of human hearing.Little wonder, then, that engineers often extend into these natural characteristics to solve human problems. After all, evolution has been fine-tuning animals and their senses for countless generations: why not take advantage of a ready-made design?
As technology advances, life-mimicking, or biomimetic, solutions are increasingly the answer to a variety of challenges. In the coming decade, such innovations could yield improvements in everything from hearing aids and sonar to bomb-sniffing sensors.
From the ears of an insect
Scientists have long been impressed by Ormia ochracea, a small parasitic fly whose females deposit their larvae on or near certain types of cricket. Despite having tympanic membranes (the insect equivalent of human eardrums) that are only half a millimetre or so apart, lab tests show the flies can pinpoint the direction to a chirping host within 2° - a sound-finding accuracy comparable to that of humans, who typically have a distance between their left and right ears over 300 times greater.
All things being equal, the further apart a creature’s ears, the larger the differences in sound intensity and arrival time between the left and right ear; both are key clues to help a creature discern where a sound is coming from. But obviously, all things are not equal with Ormia.
The fly’s two trampoline-like eardrums are directly connected to each other with a stiff bridge of tissue that helps amplify directional cues, says James Windmill, an electronic engineer at the University of Strathclyde in Glasgow, Scotland. But that enhancement comes at a price: recent analyses reveal that the directional accuracy is best at only one particular frequency - which for the fly is the 5kHz chirp of its cricket host.
Yet for many applications, such as in hearing aids, designers seek directional accuracy over a wide range of sound frequencies, says Windmill. He and his colleagues are now developing a tiny broadband microphone that’s based on the fly’s hearing system design - one that in hearing aids might help wearers focus only on sounds coming from in front of them and reduce or control unwanted noises arriving from other directions. One of the team’s designs , described in the April 2014 issue of Micro & Nano Letters, doesn’t quite look like Ormia’s ears, but in terms of performance “it’s mathematically close”, says Windmill.
A newer, simpler version of the mike maintains this keen performance.Using the same techniques that manufacturers use to etch circuits in silicon wafers, the researchers created a prototype mike that looks like a large figure ‘8’ straddling a long, slim beam. Sound waves arriving from one side of the beam will strike one disc before the other, causing the beam to twist in that direction. Sound from a source along the beam will deflect both discs at the same time, causing the beam to bend. By measuring the forces at each end of the beam, it’s possible to estimate the intensity of the sound as well as the direction from which it originated, says Windmill. So far, he notes, the team’s tests suggest the prototype mike can pinpoint the source of a sound within a range of 5° or 10°.
Over the next three years - and funded by a £430,000 grant provided by the Engineering and Physical Sciences Research Council - the team will refine the device. “We’re still at the point of figuring out what these sensors are capable of,” says Windmill. Then, the team will have a design that can be turned over to manufacturers for more detailed analyses such as durability tests; whilst making the microphone’s beam thin allows it to twist more freely and therefore boosts its sensitivity, it also increases its fragility, he notes.
Besides its potential use in hearing aids, the technology could also be used in networks of devices deployed to help pinpoint sources of gunfire on the battlefield or in crime-ridden neighbourhoods. Or, possibly more lucrative, tiny fly-based mini-mikes could find their way into laptops, tablets and the ever-popular mobile phones. “At one billion units a year, that’s an awful lot of microphones,” says Windmill.
Different fly, different sense
The fruit fly Drosophila melanogaster is the lab rat of the insect world for many reasons: it breeds quickly, it’s easy to care for, and it has a relatively simple genome. This aptly named creature, also known as the vinegar fly, feeds on fermenting fruit, so it’s no surprise that evolution has finely tuned its olfactory neurons to recognise odours that help it home in on a meal. Recent studies, however, have revealed the fruit fly’s sensors also respond to far more interesting compounds - including a variety of explosives.
Altogether, there are about 28 different types of olfactory neurons in the fly’s antennae and palps, the sensory appendages near its mouth. In living insects, about 20 of those can be easily probed with wires to measure their responses to various odours, says Thomas Nowotny, a computational neuroscientist at the University of Sussex. In lab tests, he and his colleagues assessed the responses of these 20 neuron types to 36 aromatic chemicals often found in the air above wine. All but one of the neuron types - one that is already known to be particularly devoted to detecting carbon dioxide - responded to at least one of the 36 odours, says Nowotny. Conversely, 29 of the compounds caused one or more of the neurons to fire at least 15 times per second, a sign of excitement in the neurological sense.
Then, the researchers exposed the olfactory neurons to 35 odours that the flies typically wouldn’t have encountered in a natural environment and therefore hadn’t specifically evolved to detect. Many of these chemicals are associated with illegal substances or considered noxious or even toxic, says Nowotny. And despite a lack of evolutionary ‘need’ to discern these chemicals - everything from bleach and ammonia to the explosives nitroglycerin, TATP and TNT - the neurons fired at least 15 times per second when exposed to 21 of them. The researchers describe their results in the October 2014 issue of Bioinspiration & Biomimetics.
Usually, one type of olfactory neuron isn’t sufficient to identify the presence or absence of a particular chemical of interest, says Nowotny. “Typically, that’s a problem of pattern recognition,” he notes. In other words, a number of neurons, along with the strengths of their responses, would form a distinctive pattern that could be used to recognise the chemical.
So what use are these findings? Fruit flies can’t be trained to fly into a hazardous environment and then report back with their findings. Nor can their olfactory neurons be easily grafted onto circuit boards to create electronic sensors. Instead, Nowotny suggests, the proteins in the neurons that respond to hazardous chemicals can be engineered to change colour, or even luminesce, when they detect the compounds. Then these proteins could be extracted or produced in large quantities and incorporated into arrays of sensors that can, in turn, be incorporated into devices such as ‘electronic noses’. These sniffers could be used as handheld equipment by inspectors or, for really hazardous environments, sent in via drones or robots.
Nowotny and his colleagues are now working to integrate traditional metal-oxide-?based sensors into electronic noses that could be riding on a drone within a year. To develop and miniaturise the fruit-fly-inspired sensors might take a decade or so, he estimates. In the meantime, though, such sensors, once fine-tuned, could well complement or possibly replace larger and more expensive equipment such as gas chromatograph/mass spectrometers, says Nowotny.
Like a bat, but underwater
Originally inspired by nature, sonar systems use sound to detect objects and determine their range. Some merely listen for the sounds emitted by their targets, but other systems emit pulses of sound and then listen for the echoes.
In either case, the receiving portions of typical sonar arrays can be made up of a large number of simple microphones, says Rolf Müller, a mechanical engineer at Virginia Tech in Blacksburg, USA. Underwater sonar receivers can measure several metres across and include hundreds of microphones, he notes. But those arrays could shrink dramatically if they were based on a horseshoe bat’s anatomy.
Horseshoe bats have only two receivers (their ears, of course), which typically lie only a few centimetres apart. But those ears aren’t simple microphones, says Müller. Besides having a small flap of cartilage in front of each outer ear, or pinna, and a notch in one edge, the inner surface has a central ridge and a washboard pattern of grooves and ridges - all of which influence how sound bounces off the pinna and then makes its way into the ear canal. Further complicating the matter, sudden contractions of the 20 or so muscles in or attached to each pinna can warp the ear in as little as 100ms, about one-third of the time that a human takes to blink an eye, says Müller.
All of these machinations help the bats process and make sense of incoming sound waves - but exactly how the bats do that is still a mystery. To study the process in the lab, Müller and his team have built a simple model of the bat’s ears, as well as the structures around its nose that help shape its outgoing ultrasonic chirps. Rather than building a complicated cone of cartilage, each faux ear is cut from a thin, deformable sheet of rubber. And instead of twitching the ears with muscles, the tip is moved back and forth with a single rod.
The team’s analyses of their robotic sonar system are shedding light on how bats perceive the world, says Müller. Once we have a better idea, he and his colleagues reported at this year’s meeting of the Acoustical Society of America held in Washington DC in May, it may be possible to build sonar systems that are smaller, use less power and require less computer processing than current sonar arrays.
Once developed and fine-tuned, the team’s design could be used on aerial drones to help the craft navigate, avoid obstacles and to track or home in on targets - essentially, the same tasks for which a real bat uses sonar. For underwater systems, the faux pinnae would need to be housed inside compartments that were separated from the surrounding water, says Müller: Otherwise, motion through the seas would cause turbulence that might complicate signal processing, he notes.
The innovations being developed by Müller, Nowotny and Windmill’s teams should find their way into the technology of the future. So if a decade from now you run across a bat-eared, toxin-detecting drone with a keen sense of hearing, watch out...