Learning fast in autonomous drone racing - Nature.com

In the AlphaPilot Challenge, teams compete to fly autonomous drones through an obstacle course as fast as possible. The 2019 winning team MAVLab reflects on the challenge of beating human pilots.

A unique property of the AIRR competitions compared to previous events is the level playing field for all teams as they have to work with identical flying robots, making the teams’ software the only differentiating factor. The 2019 season began with a qualification round in which teams had to race virtually. Of the 424 teams who participated, only nine teams were selected for the main competition with real drones, which took place over three races and a championship, in different US cities. The overall goal of AlphaPilot was to create a sandbox for the rapid development of AI systems for drones. By providing a hardware-in-the-loop simulator for testing, and the robot hardware during races, the teams could focus entirely on developing artificial intelligence.

Access options

Access through your institution

Change institution

Buy or subscribe

Subscribe to Journal

Get full journal access for 1 year

99,00 €

only 8,25 € per issue

Subscribe

Tax calculation will be finalised during checkout.

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

Rent or Buy

All prices are NET prices.

Additional access options:

• Log in
• Learn about institutional subscriptions

References

1. 1.

   Moon, H. et al. Intell. Serv. Robot. 12, 137–148 (2019).

   Article  Google Scholar 

2. 2.

   De Wagter, C., Parades-Valles, F., Sheth, N. & de Croon, G. Preprint at ArXiv http://arXiv.org/abs/2109.14985 (2021).

3. 3.

   Foehn, P., Romero, A. & Scaramuzza, D. Sci. Robot. 6, 1–9 (2021).

   Article  Google Scholar 

Download references

Author information

Affiliations

1. Micro Air Vehicle Lab, Faculty of Aerospace Engineering, Delft University of Technology, Delft, Netherlands

   C. De Wagter, F. Paredes-Vallés, N. Sheth & G. de Croon

Authors

1. C. De Wagter
   View author publications

   You can also search for this author in PubMed Google Scholar

2. F. Paredes-Vallés
   View author publications

   You can also search for this author in PubMed Google Scholar

3. N. Sheth
   View author publications

   You can also search for this author in PubMed Google Scholar

4. G. de Croon
   View author publications

   You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to C. De Wagter or G. de Croon.

Ethics declarations

Competing interests

The authors declare no competing interests.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

De Wagter, C., Paredes-Vallés, F., Sheth, N. et al. Learning fast in autonomous drone racing. Nat Mach Intell (2021). https://ift.tt/3DA8iWH

Download citation

• Published: 11 October 2021

• DOI: https://doi.org/10.1038/s42256-021-00405-z

Adblock test (Why?)

Learning fast in autonomous drone racing - Nature.com
Read More