VIAE-Net: An End-to-End Altitude Estimation through Monocular Vision and Inertial Feature Fusion Neural Networks for UAV Autonomous Landing

Abstract

Altitude estimation is one of the fundamental tasks of unmanned aerial vehicle (UAV) automatic navigation, where it aims to accurately and robustly estimate the relative altitude between the UAV and specific areas. However, most methods rely on auxiliary signal reception or expensive equipment, which are not always available, or applicable owing to signal interference, cost or power-consuming limitations in real application scenarios. In addition, fixed-wing UAVs have more complex kinematic models than vertical take-off and landing UAVs. Therefore, an altitude estimation method which can be robustly applied in a GPS denied environment for fixed-wing UAVs must be considered. In this paper, we present a method for high-precision altitude estimation that combines the vision information from a monocular camera and poses information from the inertial measurement unit (IMU) through a novel end-to-end deep neural network architecture. Our method has numerous advantages over existing approaches. First, we utilize the visual-inertial information and physics-based reasoning to build an ideal altitude model that provides general applicability and data efficiency for neural network learning. A further advantage is that we have designed a novel feature fusion module to simplify the tedious manual calibration and synchronization of the camera and IMU, which are required for the standard visual or visual-inertial methods to obtain the data association for altitude estimation modeling. Finally, the proposed method was evaluated, and validated using real flight data obtained during a fixed-wing UAV landing phase. The results show the average estimation error of our method is less than 3% of the actual altitude, which vastly improves the altitude estimation accuracy compared to other visual and visual-inertial based methods.

High Speed Safe Autonomous Landing Marker Tracking of Fixed Wing Drone Based on Deep Learning

“…(1) Each frame uses target detection network to locate the landing markers such as Safadinho [12] and Nepal [15]; (2) The target tracking method is used to obtain the landing markers' position in the subsequent frames. For example, Kyristsis [5] and Polvara [6] used the Extended Kalman Filter (EKF) algorithm, In Yang [8], YOLOv3 was used in conjunction with the data association method to obtain the target tracking; Liu [18] used the TLD-KCF method for tracking; In ref.…”

“…We will elaborate on these related technologies in the "Related work" section. In the relevant state-of-the-art literature that we have collected, most of them [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17] studied the landing technology of rotorcraft UAVs, and only [1][2] investigated the technology related to the fixed-wing UAV landing. However, Kong [1] stated that his method relies on the ground-based stereo guidance system, and Zhang [2] focused on a method for high-precision altitude estimation.…”

“…In the relevant state-of-the-art literature that we have collected, most of them [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17] studied the landing technology of rotorcraft UAVs, and only [1][2] investigated the technology related to the fixed-wing UAV landing. However, Kong [1] stated that his method relies on the ground-based stereo guidance system, and Zhang [2] focused on a method for high-precision altitude estimation. As far as we know, there are barely no studies on the airborne visual navigation landing algorithms for fixed wing UAVs without ground equipment.…”

Towards mmWave Altimetry for UAS: Exploring the Potential of 77 GHz Automotive Radars

“…Accurate altitude estimation is paramount for ensuring the safety of unmanned aerial systems (UASs), particularly during the crucial phase of landing [1]. While various technologies contribute to this essential task, each method possesses unique limitations.…”