MicroPython is a tiny, open source Python programming language interpreter than runs on embedded boards. It was originally developed in 2013 and released in 2014 by Dr. Damian P. George of Cambridge University. It implements Python 3 and is designed specifically to run on devices with limited resources.
As with many things in the IoT world, the use of a particular language is greatly influenced by the tools, libraries, and packages one has available. MicroPython comes with an interactive read-evaluate-print-loop (REPL) for quick prototyping and also fully supports loading MicroPython scripts onto a device to run as well. It supports an extensive Python based library and for lower level operations can be extended with C/C++ functions as needed.
Many of the tasks that beginning IoT folks start off with, such as blinking lights, reading switches, driving servos, etc. are all easy to do in MicroPython. Further, it is often incredibly easy to read as well as it is all Python based. For example, if one has MicroPython installed on a NodeMCU ESP8266 board, a popular and inexpensive board to start working in IoT, with a basic LED light connected to GPIO Pin 5Â it is pretty simple to get the light to turn on (high state) and off (low state).
import machine led = machine.Pin(5, machine.Pin.OUT) # Turn on LED led.high() # Turn off LED led.low()
The machineÂ module is included with MicroPython and handles many of the hardware connections from software. Â It allows access to things like GPIO pins, CPU frequency, I2C busses, etc. If it is low-level on the hardware, theÂ machine module is the way to go!
As one can imagine, crammingÂ everything from Python 3 into a micro-controller isn't possible, but core data types and modules which make sense for embedded systems are included. Another limitation is that since Python is a high level programming language it isn't as fast or lean as C or C++. While many operations aren't impacted by this, if your application requires tight timing or performance is of utmost importance, using MicroPython can be an issue. However, as stated earlier, one can extend MicroPython to use C/C++ in these circumstances.
Since MicroPython is open source, check it out on GitHub, one can further extend or adapt it specific use case scenarios. For example, if you need something specific to run at boot time you can include it in the package that is flashed to the device. Pretty handy and convenient.
Hopefully I have at least sparked your interest enough in using MicroPython for IoT to take a further look at it. The development site actually has a live, interactive device where you can try out some code and see it run on a real world device. Give it a try and leave a comment below to share with others what you've done and built!