Math, Juggling, Hidden Markov Models, and embedded Python
summary
Builds a formal model of juggling as a siteswap state diagram, proves nonobvious patterns like 4-4-1 are reachable, and attempts to turn throws into live signals using a hardware prototype: an STM microcontroller, accelerometer/gyro, radio, RGB LEDs, battery and a 3D-printed shell. Firmware runs on embedded MicroPython but sampling hits only 60–70 Hz so freefall detection and throw classification fail for narrow throws; students produce working boards and lights but the hard shell and low sample rate prevent reliable real-time siteswap identification.
AI-written summary · about
chapters (8)
- 0:00Start
- 1:57Juggling Patterns
- 6:25Axioms
- 16:54Fundamental Statements about Juggling
- 18:14Juggling Pattern 3
- 20:19Three Ball Patterns
- 37:23Origami
- 38:06Crease Pattern
related talks
49mJuggling Patterns and Programs
33mMaths Whimsy with Python
42mJavaScript in All the Creative Places
37mRobots Love JavaScript (too)
01h 40m"Student showcase"
21mUsing comedy as an excuse to play with python-programmed microcontrollers
28mStrike a Pose - Gesture recognition in JavaScript w/ ML & Arduino
22mStrike a Pose: Gesture Recognition in JS with Machine Learning