detecting houseflies

dwbowen

Hey Guys,

Thanks for the input my last piece ended up working out well. If you would like to see it go to this link.... http://www.dwbowen.com/flies.html

I have started working on a new piece that will navigate a space based on the movements of a swarm (roughly250) houseflies. The flies will be housed in a 10" spherical chamber and if I figure out how to pull it off their collective movements will determine the movements of the device. I am having trouble finding the right sensor to detect them and was wondering if any one has any suggestions. I tried using ultrasonic but the ping does not seem to be sensitive enough and I found that if more than one is used their signals will interfere with one another. An IR transmitter/receiver combo does not give me enough range. I've also been playing around with the TSL230R with an LED focused on it to detect the shadows of the flies but it seems like that will only work in a completely dark space as it picks up too much ambient light. Photo-resistors are out for the same reason. Can the TSL230R be tuned to detect a only certain type of light? Is there an infrared receiver or way to use Stock#: 350-00014 to give me a greater number of variables like a photo-resistor or TSL230R? Any ideas would be greatly appreciated.

cheers!

david bowen

Tracy Allen

How about a microphone to listen for their buzzing? Tuned to their vibes?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com

Phil Pilgrim (PhiPi)

If you use modulated light, the ambient light won't be such a big problem. The trick is to take two readings: one with your light source on, then another with it off. Subtract the second reading from the first to get the actual response. To do this quickly enough, a light-to-voltage sensor, like the TSL13S, would be a better bet. You will need an A/D converter to convert the voltage to a number you can use.

You will still have to make sure your modulated light source is relatively bright w.r.t. the ambient light and that niether saturates the sensors.

-Phil

Dgswaner

what about a video camera, make the container holding the flies white then measure the black dots. multiple camera angle could provide further data.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
A complex design is the sign of an inferior designer. - Jamie Hyneman, Myth Buster

uxorious

New meaning to the term "fly paper"............. very nice.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
~~ dRu ~~

LoopyByteloose

I suspect that combining two sensor technologies will eliminate false positives. Maybe audio detection with some choice of optimal motion detection.

Ultrasonic motion detection is the most appealing to me, but you may have to develop your own receiver to handle identifying such small critters as the detection spikes might not register on a normal setup. I have doubts that optical devices will focus on the right region unless there is a scanning auto focus setup.

It is much easier to look at flies in a tank than in the real world. Understanding their behavior is relevant. For example, flies do know how to hide when a threat is imposed. I once worked in a construction trailer where every morning we would kill hundreds of flies in residence because they move slower at lower temperatures and reside overnight on the ceiling. It becomes quite obvious that they prefer to land and remain motionless on black bodies when under attack. So, the fight zone will only be occupied if they feel safe.

The thermal corelation may be useful too. All insects slow down as the temperature drops. Ants and termites in Taiwan are quite hyper in the summer months as it is 85 degrees 24/7, but they do slow and disappear in the winters.

Good luck.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:

Do you want a quickly operational black box solution or the knowledge included therein?······

···················· Tropically,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan

Post Edited (Kramer) : 10/25/2007 9:04:48 AM GMT

Bill Chennault

dwbowen--

Gee. This is fascinating!

(Why does the chamber have to be round?) Why not mount some photo-receptors in a circle on one side of the container and shine a light on'em from the other side. As the flies buzz around, the photo-receptors will produce different voltages. Figure out your direction based on the greatest voltage or voltages.

If all voltages are the same, it might be time to re-fly.

--Bill

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
You are what you write.

Bill Chennault

David--

I watched the movie. That might have been the neatest thing I have ever seen. Have you performed any analysis on the drawings? Have you considered using creatures with a hive mentality? Bees? Ants? K-State grads?

--Bill

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
You are what you write.