Part 01 / Software & machine learning
01 / 09Write your own quadrotor sim
Make every force, frame and timestep explicit.
Build the physics you will later ask a network to understand. Validate simple cases before introducing feedback, then compare a short trajectory with an independent simulator.
The goal for this build
What you’ll be able to do.
- Derive translational and rotational motion from forces and torques.
- Represent orientation without accumulating quaternion norm drift.
- Separate rigid-body physics, motor dynamics and the firmware rate loop.
- Produce reproducible validation evidence before tuning a controller.
What to have ready
- Python arrays and functions
- Vectors, cross products and derivatives
- Download and unpack the course starter kit
Free account
Read the lesson, run the lab, pass the gate.
The whole course is free. Create an account to unlock the lesson, the build lab and the gate command for this module, and to download the starter kit.
03 / Prove it
Keep the evidence.
Then run the gate.
What you ship
- A pure sim.step function and a complete Params definition
- Firmware rate-loop adapter with a pinned upstream revision
- Hover, free-fall and reference-trajectory artifacts
- Module 01 evidence and an honest PROGRESS.md entry
Module 01 gateEvidence required
- Hover thrust differs from mg by no more than 1%.
- Free-fall positions agree with the analytic trajectory within the provisional 1 mm, at g = 9.81 m/s².
- Maximum position difference from the matched reference run is at most 2 cm, and attitude difference at most 2°, over the 0.5 s course excitation.
- Record conventions, parameter values and source revisions with the result.
With a free account: the gate command and how to record your evidence.
The story behind the build
Episode companions.
Written companions are available now. Recordings and customer access are forthcoming.
Made it through the lesson?
Save your reading progress here. Record your gate result in your project.