Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Conference or Workshop Paper |
| Scope | Discipline-based scholarship |
| Published in Proceedings | Yes |
| Title | Deep Drone Acrobatics |
| Organization Unit | |
| Authors |
|
| Presentation Type | paper |
| Item Subtype | Original Work |
| Refereed | Yes |
| Status | Published in final form |
| Language |
|
| Event Title | Robotics: Science and Systems |
| Event Type | conference |
| Event Location | Online |
| Event Start Date | July 12 - 2020 |
| Event End Date | July 16 - 2020 |
| Place of Publication | Online |
| Publisher | Science and Systems |
| Abstract Text | Performing acrobatic maneuvers with quadrotorsis extremely challenging. Acrobatic flight requires high thrustand extreme angular accelerations that push the platform to itsphysical limits. Professional drone pilots often measure their levelof mastery by flying such maneuvers in competitions. In thispaper, we propose to learn a sensorimotor policy that enablesan autonomous quadrotor to fly extreme acrobatic maneuverswith only onboard sensing and computation. We train the policyentirely in simulation by leveraging demonstrations from anoptimal controller that has access to privileged information. Weuse appropriate abstractions of the visual input to enable transferto a real quadrotor. We show that the resulting policy can bedirectly deployed in the physical world without any fine-tuningon real data. Our methodology has several favorable properties:it does not require a human expert to provide demonstrations,it cannot harm the physical system during training, and it canbe used to learn maneuvers that are challenging even for thebest human pilots. Our approach enables a physical quadrotorto fly maneuvers such as the Power Loop, the Barrel Roll, andthe Matty Flip, during which it incurs accelerations of up to 3g. |
| Other Identification Number | merlin-id:20316 |
| PDF File |
Download from ZORA
|
| Export |
BibTeX
EP3 XML (ZORA)
|