Developing an embedded application for the IoT
summary
Built a wearable that detects a fall with an accelerometer, filters spikes with a templated moving‑average and hardware debouncing, and sends an MQTT message over a CC3000 Wi‑Fi link (adapter wraps Arduino client) to a Mosquitto broker on a Raspberry Pi which forwards commands over serial to an Arduino that drives a servo/H‑bridge. Demo is flaky due to captive‑portal Wi‑Fi, C++ adds a small code‑size overhead but improves safety and reuse. Arduino + accelerometer MQTT via Mosquitto Raspberry Pi broker and camera
AI-written summary · about
chapters (23)
- 0:00Intro
- 1:20The dark secret
- 2:21Using a button
- 2:51Switch - interrupts
- 5:38The sensor: ADXL 335
- 7:15Moving average filter
- 11:34Main code
- 15:36Security?
- 16:35The cloud service
- 18:24The remote executor
- 18:57Remote executor HW
- 24:39Serial port internals
- 26:23Arduino Motor Controller
- 27:50Rotary encoder
- 28:05H-Bridge
- 29:49Serial communication
- 31:20Arduino protocol
- 31:53Motor Controller: Old
- 32:58Simple buffer
- 33:39A generic parser
- 34:26A message handler
- 37:07The full picture
- 41:29Conclussions
related talks
28mSensors, Servos, and Signals with Rust
33mBuild your own (Rust-y) robot!
30mEmbed it yourself
29mRust: Embedded, Async, All the Way
30mKeeping humans safe around robots with Rust
21mChips don't bite! An introduction to µController programming for Rusteaceans
23mCreative coding or serious engineering: why not both?
05mmqtt_cpp: Boost.Asio based mqtt communication library