A few updates for the AGI-ESP8266 project:
I ordered a pair of Wemos D1 Mini ESP8266 microcontrollers:
The Wemos D1 Mini's are better than the Adafruit Huzzah ESP8266 microcontrollers because they have a microUSB connector, which makes development easier. What this means I won't need to connect a UART serial connection to the microcontroller for programming, as it's capable of forming a serial connection over USB. This makes it so much easier.
Additionally, I have begun extending an open source project, Homie-esp8266 to include the various plant related sensors:
The reason I chose Homie-ESP8266 relates to software. The Homie-ESP8266 convention allows me to abstract my code for just the sensors to include build environments other than the Adafruit Huzzah and D1 Mini environment that I'm writing them for. Additionally, because the Homie-ESP8266 convention has native, built-in support for Over The Air (OTA) updates via MQTT I can update the software on the microcontroller over a wireless network. It also allows me to focus just on writing software, I won't need to configure each individual device using the Arduino IDE and having that software work only for that device.
I have ordered several mini 5VDC submersible water pumps.
I will be using these to add a self-watering and water circulation system to the planter system. The reason I need to circulate water is because the plants themselves actually need temperature regulation when set on a window sill. One of the core problems with my current setup is that the lettuce gets far too hot during the day. The lettuce begins wilting and reaches temperatures of 30 to 35 degree's celsius. This is bad from the plant's perspective because lettuce ideally grows in temperatures ranging from 16 to 21 degree's celsius. The final system will have the opposite problem up north, namely, it'll be too cold near the window. Now, I've elected for an approach that involves heating/cooling the soil with a passive system. Circulating water held at room temperature will bring the temperature up or down without any complicated, inefficient and potentially dangerous heating/cooling systems. It's also stupid simple.
All of this brings me to what I consider to be a major design change:
The system as it exists relies upon laying commercially available planters horizontally along a window sill. I figured originally that it would be the most efficient use of space. However, the horizontal design has presented several problems. If I am using aluminum aircraft cable to mount the planting trays horizontally, there are obvious forces involved that make this difficult. Mounting cable vertically along four corners of a window and then attaching the planters is not only difficult but it increases the forces on those four cables with each planter added. The new design that I am working on involves hanging a cable horizontally and hanging multiple individual planter pots/cells along it. I'm currently looking at possible designs, but I suspect the end product will radically alter the setup and make it far simpler to install, maintain and utilize. Especially from the perspective of attaching a self-watering and water circulated temperature regulation system.
So, that in a nutshell is what I have planned for the next few weeks.