Uploaded February 2025 | Updated September 2026, 2 hours ago

---

This was a slow week at times... Lots of waiting for data collection and optimizers to run. Some progress has been made towards finding a more optimal/correct robot geometry, but there's still some work to be done.

Filmed on February 14th, 2025

=========== Links Specific to This Stream ===========

🔗 Stewart Platform Error Identification/Compensation Paper:
researchgate.net/publication/342499104_An_Error_Identification_and_Compensation_Method_of_a_6-DoF_Parallel_Kinematic_Machine

🔗 Calibration pose generator:
github.com/Project-DeepBlue-Juggling/Jugglebot/blob/main/ros_ws/src/jugglebot/jugglebot/calibrate_platform_pattern_generator.py

🔗 Geometry optimizer (`optimize_robot_geometry.py`):
github.com/Project-DeepBlue-Juggling/Jugglebot/tree/main/experimenting/platform_calibration

================= General Links =================

🔗 Here's a link to the project's Github page:
github.com/Project-DeepBlue-Juggling/Jugglebot

🔗 Link to the main Onshape Assembly:
cad.onshape.com/documents/4d6545620c55797d3bb36f68/w/42ced3a9fe5d1bf7afc6efff/e/bf84687f92e1fd7b5e47b388

🔗 And to the project's Zulip site:
pdj.zulipchat.com

🔗 If you want to help support this project, the best way to do so is through my Patreon, here:
patreon.com/HarrisonLow

---
00:00 - Intro Countdown
00:18 - Stream Starts
01:46 - Cutting Down on Slop
07:53 - A Hidden Source of Error?
22:53 - How Many Unknowns Do We Need to Optimize?
30:00 - How to Fit ANY Model to the Collected Data?
37:50 - What Data Are We Working With?
45:07 - What Optimization Techniques am I Trying?
59:54 - Next Steps
1:08:00 - Adjusting Expectations
1:12:00 - More Detail on the Recorded Data