This project extends the classic inverted pendulum control problem by placing the pendulum atop a quadrotor aerial vehicle. Researchers developed control strategies to stabilize the pendulum during static equilibrium and dynamic trajectories, such as circular motion. Utilizing a novel “Virtual Body Frame” for yaw-independent quadrotor dynamics, the study achieves real-time stabilization. Experiments conducted in ETH Zurich’s Flying Machine Arena validate the system’s performance and highlight its potential for adaptive learning techniques to mitigate systematic errors.

Experiments with an inverted pendulum on a quadrocopter.

This video demonstrates an iterative learning algorithm that allows accurate trajectory tracking for quadrocopters executing periodic maneuvers.

This video shows two quadrocopters capable of not only balancing an inverted pendulum, but also of launching it off the vehicle and catching it again.

This video shows a quadrotor performing a looping trajectory while balancing an inverted pendulum at the same time.