Register now After registration you will be able to apply for this opportunity online.
Learning to fly soft drones
Advances in aerial robotics and autonomy have led to remarkable achievements, primarily through the use of drones with rigid frames. However, nature offers a different paradigm: animals’ wings are significantly softer than these rigid structures, yet they achieve highly efficient and agile flight. This project reimagines aerial drone design by focusing on soft, deformable frames and leveraging machine learning to control their unique dynamics.
This project explores the intersection of soft robotics and machine learning to enable autonomous flight for soft hovering drones. The aim is to develop a machine-learning-based control policy for autonomous flight of soft hovering drones. Using a recently developed simulator for soft aerial robots, the research will investigate how to effectively handle and potentially leverage the complex, high-order dynamics introduced by the softness of the drone's body.
This project explores the intersection of soft robotics and machine learning to enable autonomous flight for soft hovering drones. The aim is to develop a machine-learning-based control policy for autonomous flight of soft hovering drones. Using a recently developed simulator for soft aerial robots, the research will investigate how to effectively handle and potentially leverage the complex, high-order dynamics introduced by the softness of the drone's body.
Identify and implement in simulation relevant soft embodiment. Develop a robust reinforcement learning framework to train control policies that manage and optimize flight dynamics of soft aerial robots. If feasible, deploy such a model in a real-world environment.
Identify and implement in simulation relevant soft embodiment. Develop a robust reinforcement learning framework to train control policies that manage and optimize flight dynamics of soft aerial robots. If feasible, deploy such a model in a real-world environment.
Please send a motivation statement, and a copy of your CV and transcript to Stefano Mintchev (stefano.mintchev@usys.ethz.ch)
Please send a motivation statement, and a copy of your CV and transcript to Stefano Mintchev (stefano.mintchev@usys.ethz.ch)
