Two workloads are about to pass:
- One time-consuming job is done, money will take time to arrive. So I had to postpone the BeagleBone order again. Update: 2014-01-30: Money still not arrived. Update: 2015-01-10: Money still not arrived. Was told by customer's insurance that they won't pay me until the customers didn't pay their insurance bill. That's a complication. :/
- The most time-consuming other task is finally getting to an end (teacher's project) - after approximately one and a half years of headaches and bad dreams because of that crazy simplify-teachers' lives project for the didactics department ... (still hoping they decide to make it an open source project).
Then I only have one last exam and my thesis left for a giant rush as time runs out. I have few months - don't know if that will be enough for getting at least a "pass" in the thesis. Update: 2014-01-30: My thesis now is the Teacher's project. This bundles my efforts.
The hydro plant
has advanced significantly and development has to speed up even more as frontiers harden here in the farm bankruptcy problem. Giant biogas energy farmers are knocking us out, we're loosing our animals' food for producing energy (don't know where they take the money for paying these horrific amounts of field leasing money). Update: 2014-01-30: The rather small scale hydro plant alone will not deliver enough energy. Looking for exotic alternatives. What about tying magnets to trees? (=
AMOR comes back to focus again as our sheep shift plan seems to turn perspectiveless:
- The preliminary vacuum system is done.
And I came up with a solution for our 3D milling, printer, ...
combination: There one often has a workpiece-fixing problem and I think we can solve that easily now using strong vacuum suckers. What do you think? This will fix the part no matter where you place it. I plan using individually controllable suckers where only those starting at the origin that is calibrated on the fly to LinuxEMC are active. The rest is inactive not to create too much vacuum leakage but perhaps this is not even necessary. Has to be determined.
<-- This no longer is up to date. After some discussion with a farmer, I now believe vacuum is problematic if the milling object has a non-flat shape. A plane might be sucked to the ground properly by the vacuum suckers. But is this true for a ball too? I think no. (consider how the vacuum will leak while prototyping details of the head of the Statue of Rhodes
If I'm lucky I can finally get around preparing RODOS for us, too. --> R & D Topic RODOS
I will divide my efforts threefold in the next time (priority descending):
- Teachers'/University project. <-- now this is the Thesis' topic, but they had crazy special wishes, this will consume my next two months all day long (it has already started two weeks ago, but now it's official).
- [strike]Thesis: RODOS UDPserver & FlightGear interfacing (i.e. using their OpenCascade interfacing).[/strike] <-- Update: Topic now is Teacher's project.
Also I was asked for a thread scheduling visualization tool. There exist plenty (e.g. from Uni Karlsruhe) but somehow they seem not to like 'Open source' and better all roll their own. <-- Update: I have cancelled this task because I don't wish to reinvent the wheel (there are already proven open source tools). Instead I forwarded YARTISS (open source visualization tool for thread and for testing scheduling algorithms, attention: it needs Java 1.7 or newer) and this awful good state-of-the art master thesis.
- Hydro plant schematics for the 700V DC batteries and the finalization of the AC micro-grid. A later option will be to add solar panels or fuel cells, that's why 700V DC bus is chosen (just like we chose it for buildtheenterprise.com where I invite our fellows to join our efforts). Warning: 700V DC can kill and will if one handles it wrongly! Warnung: Ein Stromschlag von 700V Gleichstrom ist tödlich!
- Think about using DiVER (DC) grid for AMOR. Probably at first we now go the easiest route and simply take grid power using a long extension cord (preferably three phase). All else will prove troublesome if enough power for cooling milk has to be available. Using piping for leading milk to the existant stationary milk tanks are an option but will limit the mobility of the 'mobile' milking robot significantly. If we have no other choice, we will go that route.
Input and participation in development is very welcome. I understand people doubt the usefulness of an automatic milking system. I wonder whether they still think the same once they have to make milk, yoghurt-/cheese (or other milk products) for a bigger ecosystem let's say an open source village or a tribe in a developing country. Cheesemaking et al. already requires vast amounts of time and now image you have to milk the 100 cows for supplying enough milk for the population just using a single bucket milker. Bearing hundreds of liters of milk in the bucket. And it will take half your day - twice a day.
Perhaps you will think different of my craziness about 'autonomous' milking once you did bucket milking for y e a r s. And an automatic milking system is not much more difficult than milking buckets. The pulsators are the most critical part anyway as vacuum fluctuations easily can destroy your animals' udders, so the complification of using a manipulator will not add too much. Okay, I see, it's hard to convince you. lol