Can it bee?
Honeybees fly with a remarkable agility that would be the envy of stunt pilots, yet their navigation ‘software’ is packed into a brain the size of a sesame seed.
Now their techniques are being studied carefully at the All-Weather Bee-flight Facility at the Australian National University (ANU) in Canberra. There are plans to use them in miniature flying robotic spies and unmanned planetary exploration probes.1
How bees navigate
Bees have airspeed gauges; gyroscopes; a ‘compass’ that detects the polarization of sunlight; UV sensors to track the horizon to measure tilt; and two compound eyes, each with 7,000 hexagonal (six-sided) facets.
These facets are windows to sub-eyes called ommatidia, which are tiny tubes containing their own lens and light-detecting cells. Each tube points in a different direction, enabling vision over a wide area.
The hexagonal shape is ideal. It uses as little edge-cell material as possible (which is why the honeycomb is also hexagonal), has the least sharp corners needing less reinforcement, and is the most symmetrical structure. Such eyes are superb for detecting motion, since a small shift means different facets detect the image.
Now the ANU researchers have shown that bees use motion detection for navigation. Imagine travelling fast in a car or train. The posts on a nearby fence seem to be whizzing backwards, while objects further away seem to move backwards more slowly, and clouds seem almost to travel with you. The movement of images is called optic flow, and the faster anything seems to move, the closer it is.
By making the bees fly in tunnels where patterns on walls could be artificially moved, the researchers proved that the bees used optic flow. When the pattern was stationary, the bees flew dead centre in the tunnel, because only then would the image flow rate on both sides be identical. If the pattern on one side moved in the same direction as the bees flew, i.e. slowing down relative to the bees, they detected the slower apparent motion and calculated that the wall was now further away, and veered towards it.
The researchers found that bees are programmed to fly such that the image speed stays constant. E.g. when patterns on both sides of the walls moved with the bees, so the bees thought they were flying slower, they flew faster.
This is vital so that bees will fly fast in open spaces but slow down in more cluttered spaces, or veer away if images suddenly start to move very fast on one side, signalling that an obstacle is very close. It also helps bees land, because they can slow down automatically to keep the optic flow constant as they descend closer to the ground at a constant angle. This way they don’t need to know their airspeed or height.
Researchers proved this by making foraging bees fly through narrow tunnels which generated higher optic flow, which they calculated as flying further.
Then these bees communicated this misinformation to their fellow bees, which then started searching for the food at greater distances.
Optic flow does require contrast in the surroundings, so that images change enough to be detectable. This works very well in nature, and fails only in artificial environments such as glass windows and painted walls—explaining why bees sometimes become disoriented and keep bouncing off these surfaces.
Optic flow may solve problems that are unavoidable with conventional guidance systems. The Global Positioning System (GPS) relies on satellite mapping, but an enemy can jam satellite signals. Also, it only works on premapped objects and won’t stop a spy craft from crashing into a rubbish bin, for example. For a space probe like the Mars Pathfinder mission, it was even worse.
Signals between Earth and Mars took 11 minutes to travel the distance of 190 million km, so the robot had to crawl at a snail’s pace—52 m in 30 days. Any faster and the rover might have fallen into a crevasse before mission control even knew that the rover was in danger, let alone send the signal to change course.
However, optic flow would allow a robot to be self-steering. A prototype 1.5-m-long, 7-kg helicopter can use optic flow to hover in one spot, a major achievement that outclasses remote controlled machines. However, there is some way to go before it could navigate winding canyons, and the computer program (algorithm) is not yet perfected. It also currently needs power-hungry Pentium chips to operate.
However, a ‘specially designed chip that better mimics the bee’s energy-efficient design’ may enable a ‘100-fold reduction in power consumption, and a 10-fold reduction in weight.’1
Also, engineers have a long way to go to make a bee-sized flying robot, about 100th the length and 10,000th the weight of their prototype. For one thing, ordinary gears and pulleys don’t work properly when miniaturized.
They plan to mimic insect aerodynamics; insects flap their wings by vibrating their outer covering (exoskeleton).
Also, insect wings flap with a very complex motion, rotating and changing the tilt to achieve the required lift—the algorithm for this motion has previously been programmed into robot simulations of insect wing flapping.4,5
Bees: designed for flight
Some evolutionists have claimed that the compound eye is a bad design that no good designer would use, so it must have evolved. However, it is actually an excellent design for small creatures, enabling bees to navigate by the highly efficient optic flow method.
Also, an assertion about what a designer wouldn’t do is actually a pseudo-theological argument, not a scientific argument, that mutations and natural selection could produce this structure. It also invokes the idea that since creation or evolution are the only alternatives, evidence against one is evidence for the other.
Strangely, evolutionists protest loudly when creationists use this approach! When it comes down to hard evidence for evolution, there are huge problems.
Recent molecular evidence counts strongly against the idea that compound eyes all evolved from a common ancestor, and instead points to multiple independent origins, consistent with separate creations by a single designer.6
At the time of writing, debate rages in Ohio, USA, about whether there is ‘design’ in nature, or if the teaching of intelligent design is even ‘science’. However, as shown, the best robotics engineers have yet to design a navigation program as good as a bee’s and run it on a chip with the energy efficiency of a bee’s brain. So it’s reasonable to believe that the bee was designed by a Master Programmer whose intelligence is beyond our own.
References and notes
- Fox, D., Electric Eye, New Scientist 171(2305):38–42, 25 August 2001. Return to text.
- Esch, H., Zhang, S., Srinivasan, M.V. and Tautz, J., Honeybee dances communicate distances measured by optic flow, Nature 411(6837):581–583, 31 May 2001. Return to text.
- Doolan, R., Dancing bees, Creation 17(4):46–48, 1995. Return to text.
- On a wing and a vortex, New Scientist 156(2103):24–27, 11 October 1997. Return to text.
- Insects: Defying the laws of aerodynamics? Creation 20(2):31, 1998. Return to text.
- Oakley, T.H. and Cunningham, C.W., Molecular phylogenetic evidence for the independent evolutionary origin of an arthropod compound eye, PNAS USA 99(3):1426–1430, 5 February 2002. Their abstract says, ‘These results illustrate exactly why arthropod compound eye evolution has remained controversial, because one of two seemingly very unlikely evolutionary histories must be true. Either compound eyes with detailed similarities evolved multiple times in different arthropod groups or compound eyes have been lost in a seemingly inordinate number of arthropod lineages.’ Return to text.