Seeing an Arduino used to power an automated irrigation system is nothing new, but seeing one used to power a learning irrigation system in what its creator calls an "artificial neuron" implementation is a new one on us. Alexander Troshkov's latest Instructable, though, is worth reading even if you're not interested in automatic plant watering systems for its great description of a simple artificial intelligence.
Based on what Alexander laughingly describes as the result of having "visited a whole class of [a] neurology course," the Arduino-powered "artificial neuron" works in the same way as its biological equivalent: decisions are made based on a signal from a "synapse" and an excitation threshold to provide weighting. A moisture sensor, modified with longer-life probes, provides feedback on the plant's status, and can be back-propagated with the plant's "comfortable" moisture level through initial manual watering - after which the system will keep the moisture level around the plant's comfort zone. "Now you know almost everything a certified neurologist knows about your brains," Alexander jokes.
Using low-cost moisture sensors, a water level sensor, water pump, relay, buzzer, and an RGB LED for feedback, Alexander's project is both practical in that it keeps his plants nicely watered without drowning them and also a great way to learn basic AI concepts - not to mention how the human brain works. "It was one neuron with two synapses," Alexander explains of his creation. "Imagine our brain with 86 billion neurons each having 1000 to 10,000 synapses!"
Artificial Intelligence of my Arduino consists so far of a single soft (programmed) neuron, which I can teach and leave my pot flowers' irrigation in the trust of. In this instructable we are going to implement an artificial neuron into Arduino. I visited a whole class of neurology course, so I can explain its functioning to you in detail.