Closed-loop control — quadcopter

Draft

This tutorial is scaffolded. It narrates examples/vehicles/quadcopter.py, which is complete and runnable today.

Build a quadcopter, synthesize an LQR regulator about a hover trim, and fly it to a position setpoint with the estimate from an EKF in the loop.

To cover

• Declaring the quadcopter (four thrusters, IMU, position sensor).
• Choosing the operating point (x_ref / u_ref) and the controllable subspace (regulate=) — a single-axis-thrust craft regulates position through attitude.
• Picking Q / R cost weights.
• The closed loop: ekf.state_dict() → lqr.control(...) → sim.step(...).
• Retargeting the setpoint mid-flight.

Run it

python -m examples.vehicles.quadcopter

Source material

• Code: examples/vehicles/quadcopter.py
• Concepts: Control synthesis (LQR)