An Integrated Navigation Method for UAV Autonomous Landing Based on Inertial and Vision Sensors

An image‐based runway detection method for fixed‐wing aircraft based on deep neural network

“…Yin et al proposed an improved faster region-based convolutional neural network (RCNN) detection method for airport detection in large scale remote sensing images, and adopted the modified multitask loss function to improve the accuracy [10]. Shang et al designed an end-to-end runway detection neural network RunwayNet to generate runway segmentation, and an algorithm of edge line extraction and pose estimation based on the segmented area of runway [11].…”

“…Shang et al. designed an end‐to‐end runway detection neural network RunwayNet to generate runway segmentation, and an algorithm of edge line extraction and pose estimation based on the segmented area of runway [11].…”

“…FIGURE11 Value in the bar chart is the rate of correctly detected samples for each channel (AimP, CTL, LEDG, REDG, THR), presented by TP+TN (true positive and true negative) rate in 10 epochs for training and testing, value is on the top of each bar.…”

A Real-Time Simulation Environment for Aircraft Vision-Based Autonomous Landing System Design

“…To that matter, the use of different optical sensors and/or additional sensors have been investigated. In particular, Multisensor Data Fusion (MDF) algorithms hybridizing stereo camera information with GPS/GNSS, inertial measurements (IMU), or thermal sensors have been recently proposed in [7], [14]- [16]. The development of such vision-based data fusion approaches motivates the need for reliable and powerful computing units at a design stage.…”