UAV Pose Estimation in GNSS-Denied Environment Assisted by Satellite Imagery Deep Learning Features

Hou, Huitai; Xu, Qing; Lan, Chaozhen; Lu, Wanjie; Zhang, Yongxian; Cui, Zhixiang; Qin, Jianqi

doi:10.1109/access.2020.3048342

Cited by 7 publications

(4 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Learned features may provide more robust observations for localization. An example of a recent deep learning-based feature detector and descriptor applied to UAV localization is by Hou et al [8], who demonstrate reduction of odometry error of a UAV in short trajectories (750 m) in presence of seasonal appearance change. The approach is based on minimizing reprojection errors of a combination of D2-Net [9] features for map matching and ORB [10] features for visual odometry.…”

Section: Related Workmentioning

confidence: 99%

Season-Invariant GNSS-Denied Visual Localization for UAVs

Kinnari¹,

Verdoja

Kyrki

2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Localization without Global Navigation Satellite Systems (GNSS) is a critical functionality in autonomous operations of unmanned aerial vehicles (UAVs). Vision-based localization on a known map can be an effective solution, but it is burdened by two main problems: places have different appearance depending on weather and season, and the perspective discrepancy between the UAV camera image and the map make matching hard. In this work, we propose a localization solution relying on matching of UAV camera images to georeferenced orthophotos with a trained convolutional neural network model that is invariant to significant seasonal appearance difference (wintersummer) between the camera image and map. We compare the convergence speed and localization accuracy of our solution to six reference methods. The results show major improvements with respect to reference methods, especially under high seasonal variation. We finally demonstrate the ability of the method to successfully localize a real UAV, showing that the proposed method is robust to perspective changes.

show abstract

Section: Related Workmentioning

confidence: 99%

Season-Invariant GNSS-Denied Visual Localization for UAVs

Kinnari¹,

Verdoja

Kyrki

2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…Methods for camera-to-map matching include detection and description of individual feature points in both UAV images and a map. Examples of feature-based solutions include manually engineered feature descriptor methods such as [4], [5], as well as more recent deep learning-based feature detection and description methods [6]. Among featurebased UAV localization works, [6] demonstrates successful localization of an UAV in short trajectories (750 m) in the winter using a map acquired in the summer, when the UAV is flying at a high altitude (500 m) with a large field of view of the camera and from known starting pose.…”

Section: Related Workmentioning

confidence: 99%

Season-invariant GNSS-denied visual localization for UAVs

Kinnari,

Verdoja,

Kyrki

2021

Preprint

View full text Add to dashboard Cite

“…Deep learning model-based classification has been proved efficient in many real applications [1][2][3][4][5]. Generally, the performance of a deep learning model depends on the captured features [6][7][8]. To capture more features for higher accuracy, the structure of models becomes bigger while the accuracy is limited by many factors like the vanishing gradient problem [9][10][11].…”

Section: Introductionmentioning

confidence: 99%

Environment Adaptive Deep Learning Classification System Based on One-shot Guidance

Jin¹,

Pei²,

Zhao³

et al. 2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

When utilizing the deep learning models in some real applications, the distribution of the labels in the environment can be used to increase the accuracy. Generally, to compute this distribution, there should be the validation set that is labeled by the ground truths. On the other side, the dependency of ground truths limits the utilization of the distribution in various environments. In this paper, we carried out a novel system for the deep learning-based classification to solve this problem. Firstly, our system only uses one validation set with ground truths to compute some hyper parameters, which is named as one-shot guidance. Secondly, in an environment, our system builds the validation set and labels this by the prediction results, which does not need any guidance by the ground truths. Thirdly, the computed distribution of labels by the validation set selectively cooperates with the probability of labels by the output of models, which is to increase the accuracy of predict results on testing samples. We selected six popular deep learning models on three real datasets for the evaluation. The experimental results show that our system can achieve higher accuracy than state-of-art methods while reducing the dependency of labeled validation set.

show abstract

UAV Pose Estimation in GNSS-Denied Environment Assisted by Satellite Imagery Deep Learning Features

Cited by 7 publications

References 37 publications

Season-Invariant GNSS-Denied Visual Localization for UAVs

Season-Invariant GNSS-Denied Visual Localization for UAVs

Season-invariant GNSS-denied visual localization for UAVs

Environment Adaptive Deep Learning Classification System Based on One-shot Guidance

Contact Info

Product

Resources

About