Computer regulated enclosures

[stanislas]

Recentlly I automated my cockroach enclosures. 
An automated day/night heatingcycle. Gradual light dimming with red lights at night. Automated fogging to add moisture. This all combined with cameras that register and quantify movement. 
That gives me quite a good insight on the activity of these critters. 
I don't think that it's in any way needed for their well being. But it's fun to do... for me at least.  

I use heating mats (for the Polyphaga roaches) and a heating bulb (for the Therea). 
Raspberry pi's to control the heaters (PID controllers implemented in Python. Light: Milight wifi lights controlled by a raspberry pi with radio. 
Each enclosure has a raspberry pi with a camera to monitor the activity. All data is send to a raspberry pi with Emoncms systen to collect data and make graphs. 

So far it's (as expected) clear that these roaches are noctural  

Any one else with a similar approach for his/her roaches? 

[NeverLift]

WOW this is very cool. Computer controlled roach bins! I would love to hear more about your setups, maybe a youtube video?

[CritterChick]

I find it interesting that the P. saussurei almost seem crepuscular.  The majority of their activity is within 2 hours of sunrise and sunset, but nothing in the middle of the night.  I'd also be curious to see if activity in the P. obscura would change if the temperature didn't rise in sync with the light source.  While this is obviously how it would occur in nature, my curiosity is whether it's the heat or the light that gets them hiding.

[stanislas]

1 hour ago, CritterChick said:

I find it interesting that the P. saussurei almost seem crepuscular.  The majority of their activity is within 2 hours of sunrise and sunset, but nothing in the middle of the night.  I'd also be curious to see if activity in the P. obscura would change if the temperature didn't rise in sync with the light source.  While this is obviously how it would occur in nature, my curiosity is whether it's the heat or the light that gets them hiding.

The P. saussurei indeed a crepuscular activity pattern. They are often inactive in the middle of the night. The animals in the enclosure are adults. 
The P. obscura roaches are much younger nymphs. And they are active the whole night. 
The Therea roaches (also nymphs) show a much longer activity period. Only at the middle of the day they becomes less active.
Here a graph of the last 3 days: 

I do have some data from the Polyphaga saussurei bin when there wasn't a light bulb installed. They did receive some light during the day (no direct daylight), although very little. 
It looks to me that they where mostly active when the temperature dropped: 

It's an interesting suggestion that you make! Putting the light and temperature out of sync for a week or so. Good idea! 

 

[stanislas]

3 hours ago, NeverLift said:

WOW this is very cool. Computer controlled roach bins! I would love to hear more about your setups, maybe a youtube video?

Maybe I should document it all on a blog or something. That would allow me to share code and designs (which a gladly would like to share). 
It's fun to see their activity in real-time. I have to admit that I tend to 'check in' on my roach bins while I'm at work (and the boss isn't looking ). Downside is that one can get worried if the roaches haven't moved for a few days (which the P. saussurei's tend to to once in while).  
 

My Polyphaga obscura enclosure:

 

[stanislas]

For those interested: I've made a time lapse video of the Polyphaga saussurei cockroaches. The activity detection algorithm draws circles in this video whenever something moves. 

video of activity detection Polyphaga saussurei

So far the algorithm is fairly robust and does not get thrown off when the light levels change during the day. 

[stanislas]

Youtube took my video down? Are cockroaches inappropriate???? 

Update: youtube reinstated my cockroach video. No idea why they found in inappropriate at first. 

Edited April 8, 2017 by stanislas
Update

[Psydeus565]

It's still up, I'm guessing they disliked your audio? 

[varnon]

Very cool. I am moving in a few months, and I have been thinking about implementing something like that with a Propeller microcontroller. I will be setting up a bug behavior lab with roaches and bees. Stanislas, have you considered publishing this set up? I know a few labs that are becoming more and more interested in roach behavior. I think there would be an interest in something like you described.

[stanislas]

On 4/9/2017 at 5:03 AM, varnon said:

Very cool. I am moving in a few months, and I have been thinking about implementing something like that with a Propeller microcontroller. I will be setting up a bug behavior lab with roaches and bees. Stanislas, have you considered publishing this set up? I know a few labs that are becoming more and more interested in roach behavior. I think there would be an interest in something like you described.

Perhaps it would indeed be a good idea to put it all on my website.It's relative low cost, as it's mostly off-the-shelf material. I'll try to organize and document it all. 

What took most of the time was building the software to do the motion detection, create decent PID temperature controllers + software, and the light control and moisture system. But once it's all up and running, its fairly easy to add more of them. 

And indeed, it's very interesting to have such a control system. I have a PhD in molecular genetics. Never worked with insect though, but now I have my setup I see quite some opportunities to learn more about these animals.