Nonlinear quadrocopter attitude control

A platform for aerial robotics research and demonstration: The Flying Machine Arena

A flying inverted pendulum

Stability and control of a quadrocopter despite the complete loss of one, two, or three propellers

Adaptive fast open-loop maneuvers for quadrocopters

A computationally efﬁcient algorithm for state-to-state quadrocopter trajectory generation and feasibility veriﬁcation

Cooperative quadrocopter ball throwing and catching

Admittance control for physical human-quadrocopter interaction

Raffaello D'Andrea: Raffaello D'Andrea: Kekuatan atletis quadcopters yang mengagumkan

Raffaello D'Andrea: Kekuatan atletis quadcopters yang mengagumkan

Di dalam laboratorium robot di TEDGlobal, Raffaelo D'Andrea mendemonstrasikan quadcopters terbangnya: robot yang berpikir seperti atlet, memecahkan masalah fisik dengan algoritma yang membantu mereka belajar. Dalam serangkaian demonstrasi yang menarik, D'Andrea menunjukan robot bermain tangkap, menyeimbangkan dan mengambil keputusan bersama-sama -- dan perhatikan demo "Aku ingin ini" quads yang dikendalikan oleh Kinect.

Quadcopters can now perform feats once reserved for athletes, like balancing poles, catching balls, and even flying with damaged propellers. Their agility comes from algorithms that use physics and control theory to learn and adapt, allowing these machines to make split-second decisions and recover from unexpected setbacks. In one demonstration, a quadcopter adjusted its moves every 20 milliseconds to hit a flying ball, showing how machines can master athletic skills through rapid calculation and practice. By combining machine athleticism with human gestures, new forms of collaboration and interaction are emerging.