My team is assembled.
I have the following amazing people ready to pull the project from possibility to implementation:
Peter Steadman - project guru, mentor
Myself - chief programmer and integrator
Alana Gaudet - green expert and gardener extraordinaire
Jason Niu - Aquaponics expert
Keith Thomson - Electronics wizard
Stephanie Amann - kid-wrangler and all round support
What we may measure and control.
Water - One of the principle things people get wrong in gardening. We will use soil humidity sensors and motorized water valves to regulate watering for optimal growth, with specific table driven watering patterns for various crops.
Lighting (both natural and artificial) is crucial for optimal growth. We will measure and control that. Canada's winter may have too little - I am considering LCD lighting as the most efficient artificial provider.
Temperature is important - most plants have an optimal range for healthy growth, and we need to measure and control that.
Humidity control is desirable for optimal growth and minimal disease of plants. We will measure and control that.
Soil Quality. This is more tricky - pH, organic nitrogen, other elements are important. We can consider measuring concentration of salts (to avoid over/under-fertilization).
Air - plants (when photosynthesizing) consume CO2 and emit O2. We can measure these and chart them. We will control air-flow with vents/fans, which will also control temperature.
There may be other factors I haven't yet thought of. The team will decide.
Broadbrush design.
I like the idea of inter-connectable modular greenhouse units. These modules could be individually covered, or open (to use in a larger greenhouse or indoors). They could have individual lighting or not.
Each unit would be self contained, communicating to the central computer system, and grow one crop. They would be daisy chain-able to others. We could also utilize the same sensors for conventional (outdoor bed) gardening, and measure/control irrigation.
I am seeing use of the Beaglebone as the central computer, aggregating information from the microcontroller satellites, publishing data to the cloud.
Each modular unit would have a CC3200 based controller with WiFi for communication with the Beaglebone.
Modules may (additionally) have MSP430 based sensor and/or actuator modules, possibly using CC2500 for wireless data transmission especially where wires would be inconvenient or impractical.
