Improving Aircraft Localization: Experiences and Lessons Learned from an Open Competition

08/06/2022
by   Martin Strohmeier, et al.
0

Knowledge about the exact positioning of aircraft is crucial in many settings. Consequently, the opportunistic and independent localization of aircraft based on their communication has been a longstanding problem and subject of much research. Originating from military settings, the capability to conduct aircraft localization has moved first towards the institutional civil aviation domain and can now be undertaken by anyone who has access to multiple cheap software-defined radios. Based on these technological developments, many crowdsourced sensor networks have sprung up, which collect air traffic control data in order to localize aircraft and visualize the airspace. Due to their unplanned and uncontrolled deployment and heterogeneous receiver technology traditional solutions to the Aircraft Localization Problem (ALP) can either not be applied or do not perform in a satisfactory manner. In order to deal with this issue and to find novel approaches to the ALP itself, we have designed and executed a multi-stage open competition, conducted both offline and online. In this paper, we discuss the setup, experiences, and lessons learned from this Aircraft Localization Competition. We report from a diverse set of technical approaches, comprising 72 participating teams over three stages. The participants reached a localization accuracy of up to 25 meters in a setting with fully GPS-synchronized receivers and 78 meters in a largely unsynchronized receiver setting. These results constitute a significant improvement over the previous baseline used in the OpenSky research network. We compare the results of the study, discuss the current state of the art, and highlight the areas that, based on our experience from organizing a competition, need to be improved for optimal adoption of the competitive approach for other scenarios.

READ FULL TEXT
research
05/03/2018

The Algorithm Selection Competition Series 2015-17

The algorithm selection problem is to choose the most suitable algorithm...
research
11/30/2020

LocaRDS: A Localization Reference Data Set

The use of wireless signals for purposes of localization enables a host ...
research
10/15/2018

Traffic Signs in the Wild: Highlights from the IEEE Video and Image Processing Cup 2017 Student Competition [SP Competitions]

Robust and reliable traffic sign detection is necessary to bring autonom...
research
04/14/2022

Retrospective on the 2021 BASALT Competition on Learning from Human Feedback

We held the first-ever MineRL Benchmark for Agents that Solve Almost-Lif...
research
08/06/2021

Impressions of the GDMC AI Settlement Generation Challenge in Minecraft

The GDMC AI settlement generation challenge is a PCG competition about p...
research
05/20/2023

NTN-based 6G Localization: Vision, Role of LEOs, and Open Problems

Since the introduction of 5G Release 18, non-terrestrial networks (NTNs)...

Please sign up or login with your details

Forgot password? Click here to reset