Hi! Some time ago, together with my friends, I started a side project and I’d like to share with you some thoughts about it Day-to-day we’re working as programmers building and deploying software for different companies. Once we came up with an idea of designing and building an intelligence system for gardens watering. Our purpose is to create an economic and environmentally friendly solution. The plan is to reduce the usage of water which resources are running out and related costs, especially in developing countries. We put most logic of the system into a cloud-based web application that communicates with the controller device. This one we based on Onion Omega2S. The system works with 12 solenoid valves and we planning to implement two modes. Automatic with build-in AI, which applies watering schedule (time and duration) basing on current weather conditions (temperature, humidity, rainfall, wind), weather forecast, history of watering, conditions specific for particular climate zone and type of plants. This mode doesn’t require any specialistic knowledge to be run and used successfully. So almost none configuration, just plug it in and it will do the job. The second one will be manual, so users will be able to configure all the parameters on their own. All configuration and management will be done via the mobile application for both platforms (we are thinking also about integration with Apple HomeKit). The controller device is compatible with most of the solenoid valves and uses a 2-inch display to show some base information. It connects to the network via WiFi, so switching to this solution from already existing systems should be easy also. Overall we estimate that our solution could cost half of the price of comparable devices on the market. Currently, we have a prototype with a dedicated operating system and part of the controller software. In the video below, you can see how a sample of the controller device interface looks like.
What do you think about this project? Do you think this idea has any chance to succeed? Will anyone would like to use this? Would you like to have such a solution in your garden?
Thank you very much for all the comments and have a great day!

Video - booting

Hi,

Without going into details. I don't know if you know, but the hardware watchdog never worked in Omega2S+/2+. This was a disqualifying problem for me.

I am enclosing a patch that fixes this problem.

Have a nice day

--- a/drivers/mtd/spi-nor/spi-nor.c
+++ b/drivers/mtd/spi-nor/spi-nor.c
@@ -1086,7 +1086,7 @@
 	{ "mx25u6435f",  INFO(0xc22537, 0, 64 * 1024, 128, SECT_4K) },
 	{ "mx25l12805d", INFO(0xc22018, 0, 64 * 1024, 256, 0) },
 	{ "mx25l12855e", INFO(0xc22618, 0, 64 * 1024, 256, 0) },
-	{ "mx25l25635e", INFO(0xc22019, 0, 64 * 1024, 512, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ) },
+	{ "mx25l25635e", INFO(0xc22019, 0, 64 * 1024, 512, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ | SPI_NOR_4B_OPCODES) },
 	{ "mx25u25635f", INFO(0xc22539, 0, 64 * 1024, 512, SECT_4K | SPI_NOR_4B_OPCODES) },
 	{ "mx25l25655e", INFO(0xc22619, 0, 64 * 1024, 512, 0) },
 	{ "mx66l51235l", INFO(0xc2201a, 0, 64 * 1024, 1024, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ | SPI_NOR_4B_OPCODES) },

@crispyoz Thanks a lot for your opinion. Ps. beautiful garden.

@user5456

• By default both the Omega2+ CPU and Flash memory boots from a 3 Byte address mode
• During boot loader the CPU upgrades both Flash memory and itself to a 4 Byte address mode in order to access higher memory addresses
• During an unexpected reset (brownout) the CPU resets but the memory remains in the 4 Byte address mode.
• The mismatch in the address mode prevent the CPU from booting normally.

We have added a remote firmware update.
Below is a photo of an early prototype.

Firmware update: https://env.pl/files/irrig07.mp4

@crispyoz Thanks for your support. I will certainly let you know when we take the next big step.

A working prototype! [video]

It took a while, but the work is progressing. This is not one of those projects that is dying in the early stages We have a prototype and first implementation. Hope the owner doesn't love his garden very much

In some time I will write about further work progress.
Have a nice day!

@Modest-Polykarpovich Thanks you for your comments and views. We strive to make the user experience as positive as possible. The next stage of work on the controller is coming to an end. We will publish the results soon.

@crispyoz Thank you. This is a temporary casing.

We have added a remote firmware update.
Below is a photo of an early prototype.

Firmware update: https://env.pl/files/irrig07.mp4

@crispyoz Thanks for your support. I will certainly let you know when we take the next big step.

@crispyoz The irrigation controller software is complete. Work is currently underway on the management application for iOS and Android.

We will probably provide software for Omega for free and a mobile application in the basic version (schedule without AI) along with documentation on how to build such a controller based on Omega.

Unfortunately, the work slowed down because our team got smaller. We have no external financing and we have to divide our time between professional work and this project.

I am convinced that the project will be completed. We still need some time.

@user5456

• By default both the Omega2+ CPU and Flash memory boots from a 3 Byte address mode
• During boot loader the CPU upgrades both Flash memory and itself to a 4 Byte address mode in order to access higher memory addresses
• During an unexpected reset (brownout) the CPU resets but the memory remains in the 4 Byte address mode.
• The mismatch in the address mode prevent the CPU from booting normally.

Hi,

Without going into details. I don't know if you know, but the hardware watchdog never worked in Omega2S+/2+. This was a disqualifying problem for me.

I am enclosing a patch that fixes this problem.

Have a nice day

--- a/drivers/mtd/spi-nor/spi-nor.c
+++ b/drivers/mtd/spi-nor/spi-nor.c
@@ -1086,7 +1086,7 @@
 	{ "mx25u6435f",  INFO(0xc22537, 0, 64 * 1024, 128, SECT_4K) },
 	{ "mx25l12805d", INFO(0xc22018, 0, 64 * 1024, 256, 0) },
 	{ "mx25l12855e", INFO(0xc22618, 0, 64 * 1024, 256, 0) },
-	{ "mx25l25635e", INFO(0xc22019, 0, 64 * 1024, 512, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ) },
+	{ "mx25l25635e", INFO(0xc22019, 0, 64 * 1024, 512, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ | SPI_NOR_4B_OPCODES) },
 	{ "mx25u25635f", INFO(0xc22539, 0, 64 * 1024, 512, SECT_4K | SPI_NOR_4B_OPCODES) },
 	{ "mx25l25655e", INFO(0xc22619, 0, 64 * 1024, 512, 0) },
 	{ "mx66l51235l", INFO(0xc2201a, 0, 64 * 1024, 1024, SPI_NOR_DUAL_READ | SPI_NOR_QUAD_READ | SPI_NOR_4B_OPCODES) },

@Carey-Capaldi Is communication over MQTT encrypted in your solution? Does 'umqtt' support TLS / TLS-PSK encryption at all?

@crispyoz It's nice that you like it. Thanks.

@crispyoz Thank you. Check if you can play the video now.