Exploiting AirSim as a Cross-dataset Benchmark for Safe UAV Landing and Monocular Depth Estimation Models

Jon Iñiguez De Gordoa, Javier Barandiaran, Marcos Nieto

2022

Abstract

As there is a lack of publicly available datasets with depth and surface normal information from a drone’s view, in this paper, we introduce the synthetic and photorealistic AirSimNC dataset. This dataset is used as a benchmark to test the zero-shot cross-dataset performance of monocular depth and safe drone landing area estimation models. We analysed state-of-the-art Deep Learning networks and trained them on the SafeUAV dataset. While the depth models achieved very satisfactory results in the SafeUAV dataset, they showed a scaling error in the AirSimNC benchmark. We also compared the performance of networks trained on the KITTI and NYUv2 datasets, in order to test how training the networks on a bird’s eye view affects in the performance on our benchmark. Regarding the safe landing estimation models, they surprisingly showed barely any zero-shot cross-dataset penalty when it comes to the precision of horizontal surfaces.

Paper Citation

in Harvard Style

Iñiguez De Gordoa J., Barandiaran J. and Nieto M. (2022). Exploiting AirSim as a Cross-dataset Benchmark for Safe UAV Landing and Monocular Depth Estimation Models. In Proceedings of the 14th International Joint Conference on Computational Intelligence - Volume 1: ROBOVIS; ISBN 978-989-758-611-8, SciTePress, pages 454-462. DOI: 10.5220/0011562200003332

in Bibtex Style

@conference{robovis22,
author={Jon Iñiguez De Gordoa and Javier Barandiaran and Marcos Nieto},
title={Exploiting AirSim as a Cross-dataset Benchmark for Safe UAV Landing and Monocular Depth Estimation Models},
booktitle={Proceedings of the 14th International Joint Conference on Computational Intelligence - Volume 1: ROBOVIS},
year={2022},
pages={454-462},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011562200003332},
isbn={978-989-758-611-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Joint Conference on Computational Intelligence - Volume 1: ROBOVIS
TI - Exploiting AirSim as a Cross-dataset Benchmark for Safe UAV Landing and Monocular Depth Estimation Models
SN - 978-989-758-611-8
AU - Iñiguez De Gordoa J.
AU - Barandiaran J.
AU - Nieto M.
PY - 2022
SP - 454
EP - 462
DO - 10.5220/0011562200003332
PB - SciTePress