TEDTalks: Robotic Machinations Season 1 Episode 15 Raffaello D'Andrea: The Astounding Athletic Power of Quadcopters

Raffaello D'Andrea is a professor of dynamic systems and control at ETH Zurich in Switzerland, as well as a co-founder of Kiva Systems, a robotics company that is now part of Amazon. In this TEDTalk, D'Andrea shows off some of the latest developments in quadcopters, including autonomous flying, real-time obstacle avoidance, and an impressive ability to perform complex acrobatics.

D'Andrea begins by presenting some of the early work he did on self-assembling robots. These were small, simple machines that could snap together to form larger structures. He then moves on to the current advances in quadcopters, starting with a demonstration of the agility and precision of these machines when flying in formation.

One of the key features of quadcopters is the ability to control them in real-time using feedback from sensors and cameras. D'Andrea demonstrates this by showing a quadcopter that can fly through a window and navigate a maze autonomously. He then goes on to demonstrate the quadcopters' ability to avoid obstacles, showing footage of a quadcopter flying through a forest without crashing into any trees.

The most impressive part of the talk is when D'Andrea shows off the quadcopters' acrobatic abilities. He starts by showing a quadcopter that can balance a pole on its end, and then moves on to more complex routines involving multiple quadcopters. These include throwing and catching a ball, spinning the ball on a stick, and even juggling multiple objects.

Throughout the talk, D'Andrea emphasizes the importance of both control theory and machine learning in developing these advanced quadcopters. He points out that even simple systems, like the self-assembling robots he started with, can be incredibly powerful when given the right control algorithms.

Overall, this TEDTalk is an awe-inspiring display of what quadcopters are capable of when programmed by experts like D'Andrea. From autonomous flight to acrobatic performances, these machines are pushing the boundaries of what we thought was possible for aerial robots.