Current status project/os

Hi @roryaronson
it went a bit quiet the last weeks and I guess many users would be interested in what you are actually working on at the moment. The last blog entry is nearly one month old…
Which bug fixes/ improvements can we expect in the next version and when will that be approx.?

Thanks best
Klim

1 Like

@Klimbim, thanks for asking :smile:

Over the last few weeks, Gabriel and I have been mostly working on the next iteration of hardware, v1.3, that incorporates feedback from the community and makes general improvements to the design. We’ve also been working with our suppliers and manufacturers to begin our second production run of another 250 full kits + partial kits (just motors/cables, just plates, etc) + individual components for DIY builders. Some highlights of v1.3 hardware are:

  • An IP67 rated waterproof power supply
  • Moved the vacuum pump to the z-axis so the suction is more powerful, and we put it under a plastic housing
  • No more “Sockit Box” because of the above two changes
  • Slightly larger cable carriers which makes it easier to feed all the cables and tubing through. This change required many of the supports and other plates/brackets to become a little larger as well
  • A new open-source electronics board named Farmduino that combines the functionality of RAMPS and the Arduino into one board, and has a much better layout and connectors that make it more suited for FarmBot. (RAMPS is for 3D printing, which works, but it is far from ideal)
  • A new electronics box that is better sealed from the elements and fits the new board
  • A bunch of the new parts will be injection molded rather than machined from aluminum, which is helping us to bring the cost of the device down. These parts are still 3D printable and machinable for anyone wanting to build their own FarmBot

On the software side of things we’ve been working on the following:

  • Adding a FIND HOME command to the sequence builder
  • The firmware respecting the axis lengths as determined by calibration or via the LENGTH input boxes in the web app
  • WiFi stability improvements in the OS
  • Toggle switches to invert the jog buttons in the Controls widget so that you can match the buttons to your perspective
  • We’ve had to do quite a bit of refactoring of how points are created and stored in the database. This is the groundwork needed for move absolute commands to be able to import coordinates from the plants in the farm designer. This refactor has required code changes in virtually every part of the backend, frontend, and OS, which means it will need a good amount of QA before we release it (and the above improvements) to everyone. We’re expecting to have it pushed to the production servers at the end of next week.
8 Likes

4 posts were split to a new topic: Farmduino vs RAMPS

2 posts were split to a new topic: Recommended Power Supplies

Hi @roryaronson any news / updates on what you are working?
Targets which you set to reach e.g. by the end of this year?
Anything that you could share with us concerning the status of the project / company at the moment?

Sure,

Hardware:

  • A few weeks ago we released the v1.3 hardware documentation and CAD models on Onshape.
  • Nearly 200 v1.3 kits have been shipped to customers and are starting to come online now. We’ll probably be sold out of these kits in December.
  • The parts for our online shop are on their way to our warehouse, and we plan to put together some upgrade packages for v1.2 to v1.3. For example: repositioning the vacuum pump, switching to Farmduino, etc.
  • We’re beginning work on v1.4 hardware, which will include Farmduino v2, as well as improvements to our manufacturing/supply chain so we can ship faster.
  • We’ve applied/are applying to several SBIR (small business innovation research) grants that will help us boost R&D efforts for different FarmBot shapes and sizes and abilities. We’ll know if we win any of the grants in the next few months.
  • We’re planning a big hardware announcement in the next few months.

Software:

  • We’re very close to switching the Message Broker over to RabbitMQ
    • This will provide performance improvements for all bots
    • Allow us to fix some long-standing bugs with the bot/browser appearing falsely connected/disconnected and syncing being unreliable
    • All logs will be able to be sent to the API in real-time for storage (rather than batch uploaded) which is what sometimes causes missing messages in the ticker after a browser refresh
    • This will allow the API to send messages, enabling auto-sync (which will come shortly after the RabbitMQ release) as well as broadcast messages to all devices to announce system maintenance, etc.
  • A very large FarmBotOS refactor has been in the works for a while which will improve the configurator, initialization, and the virtual machine that executes sequences
  • After the new message broker and OS are released, we will likely shift focus to having variables in sequences. This will be a complex feature that will touch all areas of the software, and likely take a while to be implemented. I don’t have a target for the release, but it is the next large feature that we’ll be tackling.

Data:

  • Our graphic designer has now produced over 200 crop icons which are now hosted on OpenFarm.
  • We’ve entered basic crop information (descriptions, photos, spread, seed depth, etc) into OpenFarm for over 300 of the most common garden crops, and adding more information every week.
  • We’re about to import 96,000 more crop names into OpenFarm.
4 Likes

Hi Rory, thanks for the update!

What is that about?

Is there any roadmap that you could share with the community in order to see which topics, developments and directions are on the roadmap for the next year?

I do understand that you want and must to sell devices in order to generate income but I guess the software part of it should be prioritized. I guess at the moment no one is using the FB different than using it as a watering tool which is triggered manually by the user (or periodically but for sure not “intelligent”). If I am wrong I am more than happy to be corrected.
This is really not the only function it was foreseen and developed for, so I would be glad to receive some general roadmap ideas and “your” list of things to be developed, tested and implemented…

FarmBot Genesis XL? :astonished:

That’s great. I’ve personally always felt that FarmBot was more for programmers than gardeners (no offense), but the web apps tries to be very dummy proof and therefore doesn’t let you do many complex fancy programmer thingies. Having variables is a great step forward! :confetti_ball:

No, I tend to agree with Klimbim on this one. FarmBot is still a long way off from being reliable in an unattended setting. I want to feel safe about letting the bot do its thing 24/7 without me supervising it. Right now, the software isn’t stable enough to give me this security. While I agree that hardware is just as much a priority (to address issues such as vacuum power), I believe it’s just a matter of growing the FarmBot team to be able to tackle both fronts at the same time.

Did I mention that you should poke me when you’re hiring programmers? :slight_smile:

2 Likes

We’re actively working on hardware, software, and data improvements to the FarmBot ecosystem because it will take all three to achieve our vision of FarmBot as a “home appliance”. ie: an affordable and reliable hardware device that is easy to set up and maintenance, a software system that is fun and easy to use while not compromising on customizability, and a robust database that serves as the backbone for smart and efficient food production - especially for those who are not experienced farmers/gardeners.

Without good hardware kits, FarmBot would only be suited for makers and hardcore DIYers. I published open-source hardware plans for 3 years and saw less than 10 independent replications. Once we made kits available, hundreds of people got involved. What will happen when FarmBot is more affordable and can be set up in a single afternoon by one lay-person? My guess is that we will see thousands of people getting involved.

Without good software, FarmBot is only suited for programers and hackers. I understand that right now, the software takes a decent amount of time to learn, and many things are tedious to accomplish. For people who love to play with software and hack around the limitations, this is less of an issue. But for the many people out there who don’t want to spend a lot of time on the computer trying to get it to work, we’ll need many more powerful features (such as variables, looping, etc) and layers of abstraction that allow anyone to understand and use the system quickly and efficiently.

And no amount of software will substantially help novice growers grow anything without good data driving it all. In fact, perhaps the most valuable aspect of the FarmBot ecosystem will one day be the open data behind everything - data that any device (FarmBot or not), person (using a device or just their hands), and/or software system (device control interface, mobile app, voice assistant, etc) can tap into.

So while I understand that because you already have a hardware system, you are now most interested in software updates. And software is where we’re spending the vast majority of our time and developmental resources. But we aren’t necessarily prioritizing software over the continuous improvement of hardware or adding more data, because its going to take all three.

In regards to a roadmap, I’m hesitant to lay anything out (especially with dates) because it is very difficult to predict how long things will take to develop and I don’t want to set incorrect expectations. I will say though that we are on the road towards our vision of FarmBot as a home appliance, which means we’ll continue advancing the hardware, software, and data all at the same time. Not to mention the business and community side of things as well.

The immediate focuses though are opening the online shop, our big hardware announcement, variables, and adding more data to OpenFarm.

6 Likes

Hey @roryaronson,
almost half a year gone since your last bigger picture outlook. Yes we are receiving news (changelogs) in every system update now and then but else it went very quiet again. As the spring period in Europe is coming closer, could you please give everyone a bigger picture outlook about the current status (teammembers etc.) and what your plans are for this year?

Thanks Klim!

Sure thing :slight_smile:

Since the last update in this thread we’ve accomplished quite a bit:

In December we debuted FarmBot Genesis XL and began accepting pre-orders for v1.4 hardware. We’re still wrapping up some aspects of the v1.4 hardware design, but about 80% of the parts are now in production or have been ordered. We’ve decided to manufacture 500 v1.4 devices (200 XL and 300 standard size) based on the pre-order rate and our available cash. We’re pretty excited to deliver these to customers as there are some really nice improvements that make FarmBot easier to assemble and more functional. (Details will not be shared until the documentation is released when the new bots start shipping)

We launched our full online shop which includes partial kits and all individual parts in addition to the full FarmBot kits. This has opened up our business to the DIYers out there who want to build/source some of the FarmBot themselves but want to order the specialty parts from us (IP67 power supply, Farmduino, electronics box, camera, motors with encoders, etc). We’ve also seen a number of orders from people upgrading their bots, which is pretty cool.

As mentioned in the Software category updates, we’ve been regularly adding new features to the codebases and laying the groundwork for larger improvements such as the long-heralded variables feature. Something that didn’t make it into the March 15 update post but I wanted to share now is just how cool it is to see photos rotated, scaled, and positioned in the garden map, which is all @Gabriel’s math and code at work. Here is what it looks like when you have your camera calibrated and do a scanning sequence of the whole garden bed:

Moving forward we’re continuing to invest in all aspects of FarmBot:

Hardware
We’re researching other form factors of FarmBot, as well as other complexities/price points to suit different people’s needs. We’ll also continue to improve the electronics, look into partially assembled kits, improve the manufacturing/supply chain to support ever larger orders, and continue improving the open-source CAD resources and documentation.

Software
We’ll continue investing the bulk of our human resources into software, particularly in making the configuration of FarmBot faster, especially for high numbers of plants and new users. Think: variables, looping, pre-built sequences and regimens, etc. Some of these features will require extensive development time, testing, and groundwork. Luckily, we’ve been planning for them for a long time, so much of our software so far has been built with the long term vision in mind.

Data
We’re still adding more icons and information to the OpenFarm database. Unfortunately software development over there has been slow recently, but we plan to hire a dedicated person for improving and maintaining the OpenFarm codebase later this year. On the FarmBot side of things, we’re working towards being able to store and view historical sensor data. Think: saving hourly data from a temperature sensor and viewing it on a graph, or mapping soil moisture content of the whole garden and visualizing it in the farm designer.

Community
We recently passed 100k followers on our Facebook page, more and more people are joining the forum and asking questions/making contributions on GitHub, and thousands of new people every day are learning about FarmBot and getting excited!

Business
In July we’re moving our currently separate office and warehouse spaces into a combined 4,000 square foot facility (about 4x more space than we currently rent). Our team remains small, as we’ve had to use our saved up cash to always be funding the next round of manufacturing. We’re currently looking into investment opportunities that would allow us to expand the team while keeping cash reserves in place for ever larger manufacturing runs. If anyone is an accredited investor or representative of a VC firm, feel free to get in touch with me: rory@farmbot.io.

Cheers :peace_symbol:

3 Likes

Hello @roryaronson and all at Farmbot HQ
I think the Farmbot is amazing!
I do have a couple of questions that I think fit in this topic and thread - feel free to move/remove if I am wrong.
Question #1: I see that the Genisis XL is still in the “Early adopter” phase. While I would love to be a part of that, I think I would prefer to spend that considerable amount of money on a more polished and refined product. Having said that, how long before Farmbot is released as a finished (at that revision) item?
Question #2: Do you have the data showing how many and what versions of Farmbots are already out in the world? I would like to know this to help me understand how overall adoption is going. No disrespect to Farmbot, but if I am going to pay for the amazing bot, I want to know that the community is growing and vibrant and that your company is on the rise.
Thanks for taking the time to read. I really think this is a great item and I hope to hear back that everything is positive and you can’t keep up with demand!
:smile:
~Tony

Hey @Intelbotfarmer,

FarmBot will always be a work-in-progress that gets better with each version in the same way that iPhones and any other piece of technology get better every year. There will never be a point at which we’re “finished” developing FarmBot because there will always be ways to further improve it, make it less expensive, add functionality, etc.

That being said, what you’re probably looking for is a response like “Q2 2019” or “v1.6”. Unfortunately I don’t know what your idea of “polished”, “refined”, or “finished” means, so I can’t give you an answer like that. What I can say though is that each version will get better and at some point you’ll look at our offering of hardware, software, data, and documentation and consider it developed enough to be worth your time and money. Hopefully we get there sooner than later!

In regards to adoption, we don’t have a complete picture of the total number of bots out there or actual usage due to several reasons:

  • We’ve shipped over 600 FarmBots to customers between v1.2 and v1.3. However, not everyone has finished setting them up while others haven’t even unpacked them from the boxes yet. Sometimes that’s because the bot is being reserved for a school project starting in the Fall, or sometimes because people are moving houses and waiting to get settled before installing it, etc.
  • We do not know how many DIY builds are out there. You can see some of them on social media, and some here on the forum, and sometimes people email us with photos, but it’s impossible to know how many are really out there. My guess is there are 100-300 DIY builds in progress out there, but only a fraction of those completed, functioning, and/or documented online.
  • We have some analytics into how many people are downloading FarmBotOS and using our web app, though there are many people who host the web app themselves and people frequently download the OS multiple times if they’re tinkering, so that throws off any stats we have.

Looking forward, our community and company are certainly growing:

  • The v1.4 kits include our most impressive hardware yet that resolves some deficiencies with previous kits. Full documentation will be released when the first kits get delivered so you’ll be able to see exactly what I’m talking about soon!
  • We’re seeing more orders each month than ever before and anticipate further growth once the v1.4 kits are available for immediate shipping and the documentation is released.
  • We’re starting to see more 3rd party development action and open-source contributions. We’re working on a new developer documentation hub that should help stoke that fire.
  • We’re planning to expand our team significantly over the next year through increased sales and other capital infusions.
2 Likes

Rory,
Thanks for the very detailed reply!
You provided me with more than enough information about how FB is doing in the world, and it sounds good (great)! From the data and metrics you mention, it seems that this is going to be a great thing moving forward.

To better define my comment “a more polished and refined product”, I think I mean I just want to see a wider adoption and a more seasoned manufacturing machine. Having said that, I think with your release of the XL and your upcoming move to bigger and better digs, FB is at a point that I consider it a VERY viable and ready product. I did not mean any slight or inference that FB is not consumer-ready - it most defiantly is! My big roadblock really is convincing my wife that it is as awesome as I know it is and scraping together enough $$$ to get the XL - go big or go home, right? Also, I have two kids (12 & 16) and I cannot think of a better way to expose them to so much pertinent science, ethics, skills and knowledge. Just the fact that this sheds light on robotics, open-source, collaboration, coding, mech engineering, physics, nature, horticulture and so many other areas should bee selling points enough to sway the wife, but the savings account hold much sway in our household. Some silly crap about “we don’t want to live in debt” or “we would need to eat Top Ramen every night” or stuff like that… :wink: She is very frugal, but we have credit scores well above 800 because she is.

Also, I feel a bit bad because you did explaine and talk about this quite a bit in the top of this thread - apologies and thanks for being patient and tolerant :slight_smile: )
I do understand that FB should never be a “finished” product - if it was it would a) become obsolete and b) never grow as it should always do!

I will stay VERY tuned to you, as I think this is a wave I don’t want to miss.
Looking to the future of FB with so much excitement!
~Tony

1 Like