Meshtastic to build environmental networks

I’m building sensors to monitor the environment for properties related to climate change and pollution.

The key is to make the sensors/nodes

  1. very very low cost - so the very poorest communities in the rural 3rd world get access
  2. reliable
  3. autonomous

The aim is to give people control of their own environment and help them contribute to tackling climate change. It’s also to get the cost of sensor nodes down low enough to monitor complex environments.

As an example I’ve designed a ground-water depth monitoring system to predict flood risk and soil carbon-loss based on an ESP32 and a couple of sensor components which is accurate, robust and costs <10$. My aim is to get that cost down a lot further through further simplification and component sourcing in bulk.

So how do I get data from the sensor node to some server? LoRa looks like a good option but range is an issue, so a mesh of intercommunicating nodes, forming both a sensor network and a communications network looks attractive. The data volumes are tiny so bandwidth isn’t an issue but while a network across open country in flat land looks entirely feasible, connections through hilly jungle is going to be a challenge for any system.

I’d welcome comments from this community about how I can use Meshtastic to create networks of microprocessor+LoRa sensor nodes plus the necessary power supplies at the very lowest cost imaginable.

In particular what microprocessor+LoRa setup should I use to get to $<5 per node and what battery+PV-solar might I use to give enough power and keep cost right down?

3 Likes

Hi DavidJ, welcome to the meshtastic community I hope your stay here is pleasant :grin:

I love your idea of sensors as you can measure pollution and at the same time be a means of communication in areas of no interest to the phone companies

I think a plugin could make that possible :thinking:

But the hardware would not be that cheap, you will exceed 10$ :sweat_smile: but they could raise some money among the community.

I would recommend you a heltec board, a well tuned antenna for your frequency as vendors don’t give you the best… (check here the frequency for your area Frequency Plans by Country | The Things Network)

And here some sensor tests and their cost :slight_smile:

I hope the best in your project and keep updated :wink:

1 Like

Hi Anothercarolina,

Thank you very much for your thoughts. I’m very glad you like the idea.

I’m not clear what kind of plugin you have in mind. Can you explain? I have a lot to learn!

The cost issue is crucial - and if you don’t mind I’ll take your view that I’ll exceed $10 (for the microcontroller and LoRa) as a friendly challenge!

I was going to write a much longer explanation as to why I’m motivated to beat the price barrier and why I think it’s feasible - but I think it might be impolite to write a long essay about it on the forum of a group I’ve only just joined. Suffice to say that I’m a soil physicist with some idea of the necessary tech (but not the LoRa or the microcontroller tech beyond the basic Arduino-management-type stuff) and an absolute commitment to help marginalised communities take control of their own environmental issues.

So thanks again. I’m looking forward to engaging here.

1 Like

I love the idea of helping people living in rural areas :slight_smile:

Make a plugin with the API
https://github.com/meshtastic/Meshtastic-device/blob/master/docs/software/plugin-api.md

Regarding the hardware, you could use an esp32 board, a lora radio and the sensor… (they sell some without display) I was thinking about an oled display but this one won’t be necessary ^^ ah and also a solar panel :thinking:.

If you want you could post the text in a pastebin? Or right here :slight_smile:

You could also check this post

And maybe @luandro could help you :slight_smile:

Good luck :hugs:

2 Likes

@crossan007 has an environmental plug-in in the works. An early version was merged in a few nights back. What he has is really slick. Maybe you and he can collaborate on it?

3 Likes

Thank you.

Intriguing.

I clearly need to do some reading before I can start to understand how to use this. Still, what use is a day without a challenge?

The key point about cost is that most people, throughout the world, are being impacted by climate change, environmental pollution and over-exploitation yet have little or no power to influence their own futures because crucial information is either missing or unavailable to them.

They don’t want to be part of the problem yet what choice do most of them have?

Here’s one example: to the north of Kuala Selangor, in Malaysia (around 3.6N 101.2E) is an area of about 60km^2 of peat swamp. I’ve been working there over the last few years, looking at changes to the soil. Around the fringes are about a further 20km^2 of former swamp now in cultivation, much of it for oil palm. In order to grow the palm trees the area has been drained and, as a result, the peat is being destroyed - much of it ending up back in the atmosphere to feed global warming.

Some of these palm plantations are the property of small family farmers who have as a result, for the first time, enough money to live comfortable lives and see their children educated. They are very far from comfortable by US or European standards but, nonetheless, for them oil palm has been revolutionary.

Yet unless water levels in the peat are allowed to rise towards the surface the peat loss is going to accelerate. The farmers have no choice at present but to keep water levels low because they have only limited means to control water flow. What they need is some way to monitor ground water levels at regular intervals so they can manage the drainage and raise water levels without harming the trees on which they depend.

And that’s where we come in. The farmers can’t afford expensive monitoring systems so such systems must be very cheap indeed or they simply won’t be installed - and nothing will change. Yes, we can probably raise money through donations to help start the installations but, wonderful as this is, it still leaves the farmers without control. It’s not that they want to see their environments damaged - and their kids are pressing them to become more engaged with the environment - but farmers all over the world feel strongly about their relation to their land and decades of fieldwork with farming communities have taught me to respect that and work with it rather than against it, or in spite of it.

One day we’ll price carbon accurately and saving the peat lands will become an economic no-brainer. Until then we have to get those sensor nodes as close to $20 each as we can or they simply aren’t going to happen.

There are many other examples in many other contexts, close to me in the UK and all over the world. My own town is at risk of flooding from a combination of changing rainfall patterns and sea-level rise so getting a network of ground-water and river sensors installed is a big priority for us - so why not combine our efforts and develop a common range of low cost, open-source sensors for all our communities. There’s a role for community engagement and education here too, as well as building community connections right round the world.

I look forward to learning a lot more about how Meshtastic might help. I guess it’s just going to take me a while to get my head round ProtobufPlugins and Nodeinfos. :woozy_face:

2 Likes

I like what @luandro is doing. Thanks for pointing it out.

Great to see other connections to make.

3 Likes

Sounds good - though you guys are working at a technical IT level I can’t actually conceive of let alone contribute towards.

Excuse a dumb question but how do I copy him into a response like this?

1 Like

You can type @ which will bring up options of users, and as you type for example @lua my name will probably show up You’ve done that already :wink:

I was already preparing a reply to this thread, find it facinating and have thought a lot about using Meshtastic with sensors. The use cases I’ve come across with the communities I work with are mostly for detecting wild fires and deforestation, but I’m sure there are many other cases I haven’t thought of.

I’m preparing another project using Meshtastic to connect some remote villages in the Amazon, will publish about it soon. @DavidJ I’m available if you’d like to chat, either on this thread or in private.

4 Likes

Message him by typing the @ sign followed by his name.

Thank you. Have a great day.

D

Thanks @luandro

Sensing wildfires and deforestation is absoultely essential so it’s great that you’re already thinking in this direction. Me too - and it’s certainly relevant to my work in Malaysia.

I have primatologist colleagues (e.g. https://sergewich.com/) working with drones to map forest primate communities in Borneo and I’ve been thinking with them about detecting smoke at low concentrations using adapted domestic smoke detectors and then launching autonomous drones to spot fires by IR camera.

My feeling is that it’s all becoming possible and, though I’m really not an IT guy I do know enough about the physics of the environment to understand the phenomena we’re trying to detect and thus help tech people design low cost sensors which work.

I’m very happy to chat but prefer to do so on the open forum so others can step in.

2 Likes

Hello @crossan007

I’d value your comments on building low-cost networks of environmental sensors.

When talking about costs there are a number of ways to save money.

Starting with ‘off the shelf’ assemblies from the usual markets is going to cost a lot more than what the sub components direct from the manufactures will.

If you have more time than money and enjoy soldering tiny components, and are willing to fit out a workshop there is a lot you could do. But you would have to make a lot of devices just to justify the tooling costs.

A hardware platform like RAK Wireless has would be ideal.

Maybe RAK offers volume discounts, or we could seek sponsorship for certain projects.

2 Likes

@DavidJ donno if you’re aware, but there’s this amazing tool, Global Forest Watch for monitoring wild fire and deforestation throuhg sattelite imagery. It’s quite powerful. Could help, but def not the same as having sensors in place.

1 Like

Instead of buying an antenna, you can also make them.

Although if you want to tune them you should get a nanovna.

Don’t buy this tool from ebay or ali as they are clones ^^

1 Like

Hi,

That’s helpful. Thanks.

Since the aim is costs as close to single digits as possible I was certainly thinking that assembling our own hardware from components direct from the manufacturer might be a way to go. At $4 a pop, however, the Pi Pico would be hard to beat. Do you think that it, plus the cheapest LoRa board we could find would be a way forward? The cheapest ESP32/LoRa from Heltec (without the screen) I can see is about $12.

Lots to think about!

Cheers,

David

Hi,

I wasn’t aware. That’s very cool. I’ll have a look later when my work’s done.

Open source satellite imagery is fantastically useful. I can stare at Google Earth for days on end, just as I used to eat up maps when I was a kid.

I’m just starting work with the Amnesty International Citizens Evidence Lab which uses imagery including satellite to search for evidence of other kinds of crime.

The best answer, of course, is to have both - and use the installation of the sensor networks to leverage better data/access with the satellite image providers.

The way things are going price/tech-wise we’ll be able to launch our own satellites before too long!

1 Like

Great resources.

I’m 60 years old and I feel like there are more, better, more exciting things to learn than ever before.

That’s a fantastic gift to be given.

And we get to play with Maxwell’s equations!

Thanks.

2 Likes