RoboGroMo

joined 2 years ago
MODERATOR OF
 

The new gui can be found here - https://github.com/Pragmatismo/pigrow_windows_remote/blob/master/test_guiWin_008.zip

This includes picam3 support with rpicam, they're great cameras and make focusing much easer - really good for close up work where depth of field makes focal distance important.

Also Venv support which is required in Bookworm, the newest version of raspberry pi OS which is required for the picam3, pi5 and a few other new things. If using bookworm then set up the venv before running the install wizard or installing any python modules.

Note - using bookworm there may be other changes they've made, for example the location of the /boot/config.txt file has changed so the boot config tools in system_pnl wil not work, i'll fix them but for now make any required changes manually (e.g. i2c baudrate, 1wire config)

The timelapse and localfiles panel have a few small updates and fixes, and there are a couple of other minor improvements.

I'll keep testing with bookworm and fix anything that needs it but hopefully will be able to focus on moving over the graphs tab so we can finally reorder everything. I've got a few projects up and running that will gathering timelapse while i'm away, i'll also be using some of the time i'm away to work on finishing some designs so hopefully i'll have a load of new stuff to add soon.

 

A simple mix of two cheap salts, Sodium Sulfate and Table Salt (Sodium Chloride) dissolved in boiling water can create a really useful Phase Change Material that has a melting point of 18c (65f) which allows it to be recharged back into it's cooling state simply by putting it in a basement and can then be used as a cooling blanket, back rest, neck pillow, etc to help keep you cool in hot weather. Longer lasting and less energy intensive to charge than an ice-pack.

In the video he talks about the potential for using a similar higher temp PCM behind solar panels to reduce efficiency loss or damage from over heating. It could also be a really interesting thing to use for transporting heat from where it's unwanted to where it's needed.

[–] RoboGroMo 3 points 8 months ago

This is really cool, I've been mapping my area and thinking about making a game that does a similar thing, nothing special just thought it'd be fun to make something that tries to generate levels using map data. I've got plenty of other stuff to get on with and mapping is addictive enough on it's own so i might just try this instead.

[–] RoboGroMo 8 points 9 months ago (2 children)

this is sadly one of the big problems with farmbot, keeping the trackways clear in an actual use situation requires as much work as weeding. My hope is that ambulatory platforms (spider robots) which are able to nimbly move through a garden space while carrying the required tool heads will be able to use these methods to work in much larger areas and without so much visually disagreeable infrastructure.

[–] RoboGroMo 12 points 9 months ago (2 children)

i feel weird about self promoting but I make a project called the Pigrow which is an open source garden automation project designed to be as cheap and flexible as possible by using a raspberry pi and basic components from ebay or wherever. It's got tools to help you monitor and control the grow environment including various watering tools. It's not as polished as farmbot but under constant development and having new features added all the time, if you want to set up a cheap irrigation automation system then it might be a good option and if not then i'd love to hear what features it'd need for you to consider using it so i can try and add them. More info at !pigrow@slrpnk.net

[–] RoboGroMo 3 points 9 months ago

I love FarmBot, not really super practical at the moment and wildly expensive but it's been making great progress and is unquestionably heading in a great and very useful direction. I think we're very close to ai controlled robotics being able to replace the clunky CNC system at which point the project will really come into it's own. Really glad they're still working on the automated weeding and stuff, that'll all be so useful when more reliable and easier hardware platforms are available.

[–] RoboGroMo 3 points 9 months ago

that's a great graph, really looking forward to seeing those times where renewables exceed demand get more common then all flatten out as systems designed to use excess power are added - pumped storage for example and desalination can be scaled to match the excess.

Basically you have a desalination plant near your solar or wind and any excess power gets diverted there, fresh water is then pumped into lakes and reservoirs which can either be converted back into power via hydro or used in agriculture, industry and homes as needed.

 

I used the camera on one of my pigrows to take pictures out my window every five min then used the daygrid tool to put them all into a single image - this is a slightly updated version with more readable labels and a red line at sun up and sunset, i'll update it in the repo as soon as i've tied it up.

 

The 1MW project, which is expected to deploy in late 2025, will use Inyanga’s innovative HydroWing tidal stream technology.

The HydroWing tidal stream turbines will be connected to the electrical network of Capul, an off-grid island currently relying on a 750 kW diesel power plant. The first stage of the project consists of a 1MW tidal power plant, to be connected into a microgrid network coupled with Solar PV and energy storage, delivering a reliable, sustainable, and cost-competitive alternative to fossil-based power generation.

Here's the diesel power plant that it's replacing https://maps.app.goo.gl/bfUhRTxcTLGRJ21d9

There's been a huge push for large scale tidal projects recently with proposals for several in big rivers in the UK but I think these smaller projects are really interesting too, replacing the islands diesel generator means they won't need regular diesel deliveries and so not only will have far more reliable power but also one less dirty cargo ship polluting the area - hopefully the limited amount of vehicles on the island can be replaced with electric especially ebikes and electric outboards so that they never need any fuel delivered - and when all the islands start doing it the boat delivering fossil fuels will no long be required and the port it sails from can clean up a bit without all the oil based products, maybe even remove the pipeline feeding it entirely if the demand for boat fuel has decreased enough due to cheap electrical prices.

It's a relatively small project but filling an important niche, will be really interesting to see how it works out.

[–] RoboGroMo 43 points 9 months ago (3 children)

tarmacking is a horrible job especially at night, personally I'd rather reduce the cost of infrastructure maintenance using automation and then pay people a living wage to do nicer jobs.

 

Driverless vehicle that uses sensors to measure road surface quality and repair small cracks to stop them turning into potholes and hopefully decreasing the cost of road maintenance while improving average surface quality.

 

I'm trying to get things to a good position where i can run a competition to create the best modules for the timelapse tab, i'm hoping to use it to attract more attention to the project while also inspiring people to create useful timelapse tools we can all use.

The modules i'm demonstrating here are the analyse set tools which scan through every image in a timelapse set, process them or analyse them then gives a single image output - currently they're all very basic and written pretty much just to demonstrate how the tools work, they are very useful though so i'll be upgrading the code when i get the chance.

Currently this tool is only available in the timelapse tab of the new gui but it's been designed in a way that allows them to be incorporated into datawall displays when i expand that feature. Also i'm hoping that the timelapse analysis tools can be a great way of testing for useful metrics which we can use to help monitor and maintain grows that are in progress, once we've established that an analysis tool is accurate with pre-recorded timelapse we can write a version designed to run every time a new image is captured.

[–] RoboGroMo 9 points 11 months ago

This is really positive news especially as most the efficiency savings come from things that are only at the start of their roll-out, a lot of the infrastructure development for solar and wind is already in place with construction already in progress for huge amounts of generation. It likely also that the lower demand for electricity comes in part due to more efficient devices gaining market share; better water heaters, heat-pumps, LED lighting, etc combined with better insulation and more focus on efficiency - plus of course home solar or similar, an increasing amount of people are at least partly off-grid and use home generated power which reduces demand on the power grid.

We also have some really useful new tech starting to reach market like tidal generation, tandem solar cells, Perovskite (which we've been hearing about for ages but they're actually starting to build factories), e-fuels (again long heralded but actually starting to move into commercial production), and various new electric planes, boats, charging technologies, energy storage mediums, and etc all of which will help increase the rate of adoption and help decrease carbon emissions.

 

I'll have a video coming soon showing how to use the new timelapse tools and write modules to stylise, analyse, select frames, etc.

https://github.com/Pragmatismo/pigrow_windows_remote/blob/master/test_guiWin_007.zip

You can copy your gui settings file and frompigrow folder to carry everything over. To make and play timelapse you'll need MPV installed - there's a script that does it on their website https://mpv.io/installation/

It doesn't currently have the overlay log feature as that will be added when the Graphing is (because it uses the same log loading mechanism).

As it's still too cold to paint or pour resin I still can't finish my growbox which means i'll probably have the user logs and graphs tabs ported into the new gui soon - also i'll be making a video explaining how to make tools for the timelapse tool.

Here's an example of one of the new analyse tools; https://imgur.com/a/UtOPgz4

it makes it very easy to see missing images, light level changes, etc. Depending on your computer it may take a long time with bigger image sets.

new graph too, https://imgur.com/a/uPmVQrT

it's showing seconds since the first image was taken, a smooth ramp means even spacing and any sudden steps suggest a missing period.

I'll be adding a few more upgrades and new tools, any suggestions or ideas i'd love to hear them.

 

I've almost finished porting the timelapse pnl over so there will be a new GUI release very soon, most of it is pretty much the same with improvements and a few sections are now modular so it's easy to drop in new scripts when required. This is the stylize timelapse tool which feeds the images through the selected python script to create a new image set, the stylize tools included to start with are mostly just to demonstrate how it works and look cool but i'd like to create some useful and informative ones also so anyone with any ideas for anything i'd love to hear.

 

Still fixing my water tank with layers of resin and stuff so don't have any space to work on the turbidity, pH sensor or anything but I'll get back to that as soon as my desk is cleared - hopefully be able to get the timelapse panel ported over while i wait for the resin and paint to set then there's only two more tabs and we can delete the old gui :)

 
[–] RoboGroMo 2 points 1 year ago

oh that's an interesting one, maybe take some soil from both places, dry it out and weigh a set amount then add a measured quantity of water to calibrate them? i don't know if it would work but it's probably where i'd start experimenting.

[–] RoboGroMo 3 points 1 year ago

Here's a video of it https://www.youtube.com/watch?v=jHNsRzzsbZ4

The video talks about controlling the fans but the heater is essentially the same, if you just want to log data and see the graphs then all you need is a Pi Zero (or any pi model) connected to a BME280 (a few other temp/humid sensors are supported but the bme is cheap and reliable)

I use the Pigrow software i wrote (free and open source) which can do most the complex stuff for you and make a whole range of graphs to break it down in useful ways. If you want to try it then i'd be more than happy to help you through setting it up, and there are guides on the github wiki and a few videos with more coming soon on the youtube channel.

If you do put in a Pi then also consider getting a camera (picam or webcam), pigrow can help you capture and assemble timelapse video too, it's a great way of seeing your plants grow and react to events especially when paired with the graphs (you can overlay them so they sync up) - there's a whole load of other sensors supported too and options for setting up watering schedules and stuff but yeah a pi and a bme280 is all you need to measure temp and humidy.

[–] RoboGroMo 5 points 1 year ago

Yeah that makes a lot of sense, personally I think a big part of the solution will have to be some form of community focused solution to end the culture of conspicuous consumerism and economic based social value - i don't think it's wrong to want good things or beautiful things but valuing something more because it's got a certain logo is absurd, even more so when you're valuing people simply based on which logo's they can afford. I fear when worker run cooperatives compete it'll still create that imperative to advertise and gain market edge which has caused so much of our twenty-first century woes.

My solution would have to involve a strong attack on not just copyright, patents and monopolies but on the very structure of our industrial economy. We need school and universities to be actively participating in community science and design projects to create verified and tested open source designs which can be fabricated locally anywhere in the world - it's a pretty radical idea really because it involves changing pretty how we do and think about pretty much everything but it's got a lot of positives.

Firstly it's basically how the PhD system was intended, you put all that effort into doing a bit of science and when you've done it they check it's ok and say 'yep, you're a real scientist now' and that science gets added to the public storehouse of knowledge for the benefit of all - we could extend that so the education system teaches and guides participation in community benefiting projects like citizen science and collaborative design -- for kids things like data gathering, group experiments, etc while university students are doing design work, materials testing, creating documentation, user guides, or other related media depending on specialisation. Projects will be worked on by community members in various ways, either as part of official efforts, community projects or individual work - basically the same model as social media, sometimes a random person goes viral for making something cool and sometimes a big company uses their budget to make good content.

The thing i always think about is washing machines because they're so painfully simple and yet when you look at the choices available in stores there's a crazy amount of totally meaningless choice - we ended up having to pick between one with 'sport' mode and one with 'sanitary' mode - presumably actually essentially the same thing but my why do they have these weird settings? because then they can have one with limited choices as the cheap one, then the next level up one that can do most the things you'll probably want and expensive ones that can do it all and have an app.. it's all just software settings, it doesn't cost them anything to have a mode that spins the drum for X seconds and runs the heater for X seconds - we could have a really simple design for a washing machine that's easy to fabricate and repair, an easy to flash microprocessor connected to controls so you can easily choose the modes you're likely to want and change your mind if the situation changes (for example you take up sports and require a longer soak and wash cycle or a new cleaning agent is developed which works better when used differently)

school kids could do supervised and documented tests to determine ideal washing conditions, it'd be a fun way to learn about science and how it relates to real-life plus they'd have more of a connection to the world they're part of, the washing machine wouldn't be a weird alien device from on high it'd be something they actually helped create - a wonderful feeling.

design students could participate in various design related challenges and projects such as creating custom displays and dial configurations, art students on making various options for making them look cool and beautiful - all passing work (i.e. work that meets the required criteria) is added to the general database of designs and options which people anywhere around the world can access when ordering an open source washing machine fabricated from their local small industrial firm, community run fablab, or to create with their own tools.

Again not saying this is the final or ideal solution but everyone should be used to having access to the very best, most efficient, and well designed things - if someone wants to show off then they should show off their good taste not their ability to outspend people without generational wealth. A community working together to design and create is always going to be better than a community battling itself.

[–] RoboGroMo 7 points 1 year ago (3 children)

I think about this a lot it's an interesting one, the current system of money is kinda crazy but the principle of having a token that can be used in trades is great.

Like @keepthepace said I can see it's significance diminish but not entirely vanish, if I want you to come and do the colour scheme for my living room then it's taking time from your life which i'd like to repay you but you might not need anything i know how to do -however personC might want something i can do and be able to do something you need - rather than having to work out every trade and find the people to agree we just use a token, that token is money.

The problem comes when you need that token to live and they're all in the hands of a greedy group of crazy people obsessed with having the most tokens - maybe we actually need more types of money not less, like maybe we should get land tokens that allow us to trade land but everyone gets a set amount and you can't just buy a thousand acres because your great-grandfather sold opium... Maybe even two types of money to buy food, a basic ration that affords for a complete and healthy diet of your choice plus a surplus coin which is earned by supplying the economy with foods or materials required (e.g. if you grow apples and supply them to the community pool you get 1 surplus token per kg but strawberries you get 1.3 st per kg due to local demand) these tokens can then only be spent on luxuries, rare items, and non-essential services.

I'm certainly not saying that's the system i propose or support simply that there's a lot more options and possibilities than we normally consider - maybe one land token gets you a small beach-front property or a huge bit of old farmland to restore, that gives everyone personal choice and helps manage demand with all sorts of interesting challenges - if you move onto a ruined plot of land and make it beautiful then you deserve more tokens than it cost you to get there but that opens of the possibility of someone purposely getting a rough bit of land, paying others to work on it using their excess surplus tokens then claiming the extra land tokens for themselves... and is that a bad thing or a good thing?

Thinking about things in obscure ways can really help to crystallise the interesting and important parts of something we're so used to thinking of in everyday terms, like what really is money and what is money supposed to be.

[–] RoboGroMo 1 points 1 year ago

unaffordable to poor people, annoying to the middle classes, utterly inconsequential to the rich. economically slavery was pretty good too, there's a lot more to making a good society than economics.

trying to price people out of living only affects those that already can't afford it, we need to be creating actual solutions at price points where they can gain widespread adoption.

view more: next ›