Today marks the release of our cloud platform environment. This includes the option of storing your data in the cloud and gaining access to our dashboard. We’ve noticed an increasing interest in these services so we’ve been working hard to supply it for you all. Myself and my team would be happy to answer questions and work with the community to progress together.
plantOS Welcomes Community Users - Code & Schematics on Github. Connect to our cloud platform optimized for precision agriculture for free!
We need help in several areas:
- Completing our docker image. We are having issues with it
- Mockups and ideas for our dashboard design. We use Stencil for mockups
- Machine Learning and Computer Vision
- Open Hardware certification
We are also on Slack at https://plantos.slack.com. Alternatively we would be happy to answer questions here and work with all of you to make us better. We are currently exploring options to manufacture our hardware at scale and lower costs significantly to democratize digital agriculture.
My name is Mick and I’m also with plantOS. We’ve been working very hard to get this version ready for you guys so I’m hoping you will like it. We’re looking forward to working with you guys to improve your overall experience!
If there are any questions, then feel free to either ask them here :)!
Hello plantOS team,
I am also building a cloud backend using the Google cloud platform. My focus is on the range of OpenAg food computers. My code will be all open source, so I’ll let you know when it is ready for review.
In the meantime, I will look at what you have done. Thank you for making it open.
Thanks. I was under the impression OpenAG was going with Amazon Web Services (AWS)
We can look into additional sensors to support to fully support OpenAg food computers however our users are vertical farm operators and plantations.
Would be good if we can collaborate. Sure will check out openag_cloud then
I did review AWS, Azure and doing it all ourselves. And I’m still testing parts of GCP and getting some help from Google with it.
I had started down the road of using their pub sub MQTT IoT protocol, but it is changing daily (beta v1) and won’t stabilize until after v2 is out. So, at their suggestion I’m looking at the HTTP implementation.
That openag_cloud repo is nothing but a bookshelf tutorial I was messing with. Nothing interesting in there now. But that will be the repo where I will put the OpenAg cloud code.
I look forward to collaboration, or at least sharing code / ideas / data formats.
My other focus is to standardize on how we organize data on the backend, so it is open and usable by other projects.
Do you have a date for v2? Assuming it is stable at that time, is this your architectural direction (my vote is yes).
As long as we don’t kill them, there are several free mqtt hosts available for play till OpenAg gets one of its own.
With GCP, it is easy to use HTTP and later switch to MQTT as the underlying
protocol for pub-sub. The differences are in performance and
configuration. Our Google developer support contact said that is the
best way to go for now.
I assume the HTTP header is the message and the body is the content. Put a facade around the core code and there shouldn’t be a significant difference in implementation.
Do you have a server and test code to play with for the pub/sup?
Working on it now. Next week I’ll put it on github for review.
@rbaynes Yes we do use the HTTP implementation on GCP and can vouch for that. Glad to know we can collaborate.
@webbhm We welcome anyone using their PFC actively. If GCP can handle Pokemon Go and Spotify I am sure it can handle a few more PFCs connected to plantOS.
Hi @Everyone, we would like to drive discussions improve our dashboard to support OpenAg PFCs? Ideas are great but mockups are better. We recommend using Pencil which is Open Source.
- What else are we missing to support a PFC? (ex: sensors, actuators)
- We are looking to expand our KPI feature to incorporate digital plant recipes. We would feedback on designing this. Have seen OpenAg examples of it being a JSON file with upper and lower limits for a few parameters. However we are looking at implementing ML models in Tensorflow(Python) for crop types we support and perform predictive analytics based on our dataset.
We look forward to this exciting phase of understanding how users are using these systems and measure optimization techniques achieved on plantOS platform. Currently we have drafts for Chinese Cabbage and building one for Oil Palm but we hope to work with the community to increase the number of crops plantOS platform can support.
Could anyone here help share code for PFC sensors please? Would help for myself and @mick to write a connector for PFC to connect directly into plantOS platform with community access. Makes it easier for any PFC user to use plantOS platform as @Sichen shared.
@devtar, the link you provided was to @webbhm’s personal instance of the MVP. Keep in mind he has added extra sensors that no one else has (yet). Here is the standard MVP stack’s Github: https://github.com/webbhm/OpenAg-MVP. We’ve just got a SI7021 sensor on the standard MVP. @webbhm has daded MHZ16 for Co2, as well as DSB1820 for water temp. We are also working on an aeroponics module, Co2 actuation, and peristaltic dosing. Are you interested in just sensors, or also actuation? I know you’ve mentioned that some of your customers are open-field so I wasn’t sure what your intent was.
Here’s the sensor code on the V2 inside of openag_brain GitHub:
Can you please elaborate on how you plan to implement Tensorflow? You may be interested in looking at the openag_sim GitHub repo as well:
Hello, I’m Andrew from Ghana, I am working on a vertical farming project in my home country. How do I get access to the plantOS slack group?
I want to know what is your teams recent development? My teams prepare to apply Tensorflow from the PFC image datasets. CNN is my teams starting points.
Through this ML, my development purpose is to detect precision “Feature” represent s Plants’s growing or disease. Second is to identify or detect plants species.
I applied “OpenCV” function to my PFC images. It is good result but preprocessing is not automated in various environment. I familiar with “Color Quantization”, “morphological image transformations”, “Findng Contours and Convex hull”.
Any sharing knowledge and experience from me and your is very helpful.
From KgwangHee Han.