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

Sensors ◽  
2021 ◽  
Vol 21 (18) ◽  
pp. 6302
Author(s):  
Xupei Zhang ◽  
Zhanzhuang He ◽  
Zhong Ma ◽  
Peng Jun ◽  
Kun Yang
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Feature Fusion ◽  
Estimation Error ◽  
Estimation Method ◽  
Signal Interference ◽  
Estimation Accuracy ◽  
Measurement Unit ◽  
End To End ◽  
Altitude Estimation

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.

scholarly journals End-Effector Pose Estimation in Complex Environments Using Complementary Enhancement and Adaptive Fusion of Multisensor

2021 ◽  
Vol 2021 ◽  
pp. 1-18
Author(s):  
Mingrui Luo ◽  
En Li ◽  
Rui Guo ◽  
Jiaxin Liu ◽  
Zize Liang
Keyword(s):  
Kalman Filter ◽  
Pose Estimation ◽  
Unscented Kalman Filter ◽  
Rbf Neural Network ◽  
Estimation Error ◽  
Estimation Method ◽  
Original Data ◽  
Measurement Unit ◽  
Complex Environments ◽  
End Effector

Redundant manipulators are suitable for working in narrow and complex environments due to their flexibility. However, a large number of joints and long slender links make it hard to obtain the accurate end-effector pose of the redundant manipulator directly through the encoders. In this paper, a pose estimation method is proposed with the fusion of vision sensors, inertial sensors, and encoders. Firstly, according to the complementary characteristics of each measurement unit in the sensors, the original data is corrected and enhanced. Furthermore, an improved Kalman filter (KF) algorithm is adopted for data fusion by establishing the nonlinear motion prediction of the end-effector and the synchronization update model of the multirate sensors. Finally, the radial basis function (RBF) neural network is used to adaptively adjust the fusion parameters. It is verified in experiments that the proposed method achieves better performances on estimation error and update frequency than the original extended Kalman filter (EKF) and unscented Kalman filter (UKF) algorithm, especially in complex environments.

scholarly journals A model-free hull deformation measurement method with time delay compensation

2018 ◽  
Vol 14 (11) ◽  
pp. 155014771881069 ◽  
Author(s):  
Ying He ◽  
Xiafu Peng ◽  
Xiaoli Zhang ◽  
Xiaoqiang Hu
Keyword(s):  
Neural Network ◽  
Time Delay ◽  
Unscented Kalman Filter ◽  
Estimation Method ◽  
Deformation Measurement ◽  
Estimation Accuracy ◽  
Deformation Model ◽  
Delay Compensation ◽  
The Neural Network ◽  
Time Delay Compensation

Estimation and compensation for hull deformation is an indispensable step for the ship to establish a unified space attitude. The existing hull deformation measurement methods are dependent on the pre-established deformation model, and an inaccurate deformation model will reduce the deformation estimation accuracy. To solve this problem, a hull deformation estimation method without deformation model is proposed in this article, which utilizes the neural network to fit the hull deformation. To train the neural network online, connection weights of the neural network are regarded as system state variables which can be estimated by the Unscented Kalman Filter. Simultaneously, considering the time delay problem of inertial data, a time delay compensation method based on the quaternion attitude matrix is proposed. The simulation results show that the proposed method can obtain high estimation accuracy without any deformation model even when the inertial data are asynchronous.

scholarly journals Data-to-Text Generation with Content Selection and Planning

2019 ◽  
Vol 33 ◽  
pp. 6908-6915 ◽  
Author(s):  
Ratish Puduppully ◽  
Li Dong ◽  
Mirella Lapata
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Large Scale ◽  
Network Models ◽  
Text Generation ◽  
Neural Network Models ◽  
Generation Task ◽  
Content Selection ◽  
End To End ◽  
Two Stages

Recent advances in data-to-text generation have led to the use of large-scale datasets and neural network models which are trained end-to-end, without explicitly modeling what to say and in what order. In this work, we present a neural network architecture which incorporates content selection and planning without sacrificing end-to-end training. We decompose the generation task into two stages. Given a corpus of data records (paired with descriptive documents), we first generate a content plan highlighting which information should be mentioned and in which order and then generate the document while taking the content plan into account. Automatic and human-based evaluation experiments show that our model1 outperforms strong baselines improving the state-of-the-art on the recently released RotoWIRE dataset.

scholarly journals Tension Estimation Method for Cable With Damper Using Natural Frequencies

2021 ◽  
Vol 7 ◽  
Author(s):  
Aiko Furukawa ◽  
Katsuya Hirose ◽  
Ryosuke Kobayashi
Keyword(s):  
Natural Frequencies ◽  
Numerical Models ◽  
Estimation Error ◽  
Estimation Method ◽  
Higher Order ◽  
Bending Stiffness ◽  
Estimation Accuracy ◽  
Test Cases ◽  
Numerical Verification ◽  
Vibration Method

In the maintenance of cable structures, such as cable-stayed bridges and extra-dosed bridges, it is necessary to estimate the tension acting on the cables. The safety of a cable is confirmed by checking whether the tension acting on the cable is within the allowable value. In current Japanese practice, the tension of a cable is estimated using the vibration method or the higher-order vibration method, which considers the natural frequencies of the cable. However, in recent years, the aerodynamic vibration of cables caused by wind has become a problem owing to the recent increase in the cable length and low damping performance of the cable itself. To suppress the aerodynamic vibration of cables, dampers are installed onto the cables. Because the damper changes the cable’s natural frequencies, the vibration method and higher-order vibration method are inappropriate for measuring the tension of a cable with a damper. In this paper, a new tension estimation method for a cable with a damper is proposed. To model a cable with a tensioned Bernoulli-Euler beam, theoretical equations for estimating the natural frequencies were derived. The proposed method inversely estimates the tension and bending stiffness of the cable and damper parameters, simultaneously, from the natural frequencies. The validity of the proposed method was confirmed by conducting numerical simulations and experiments. In the numerical verification, the performance of the proposed method was investigated using 80 numerical models. In the experimental verification, the estimation accuracy of the proposed method was investigated by considering 16 test cases. Thus, it was confirmed that the tension estimation accuracy was high, whereas the bending stiffness and damper parameter estimation accuracy was unsatisfactory. The tension estimation error was within 10% in all experimental cases, and within 5% if two test cases are excluded. The results obtained by the numerical and experimental verifications confirmed the effectiveness of the proposed method in tension estimation.

scholarly journals Online Capacity Estimation for Lithium-Ion Batteries Based on Semi-Supervised Convolutional Neural Network

2021 ◽  
Vol 12 (4) ◽  
pp. 256
Author(s):  
Yi Wu ◽  
Wei Li
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Convolutional Neural Network ◽  
Lithium Ion Batteries ◽  
Estimation Method ◽  
Lithium Ion ◽  
Estimation Methods ◽  
Estimation Accuracy ◽  
Capacity Estimation ◽  
Practical Applications

Accurate capacity estimation can ensure the safe and reliable operation of lithium-ion batteries in practical applications. Recently, deep learning-based capacity estimation methods have demonstrated impressive advances. However, such methods suffer from limited labeled data for training, i.e., the capacity ground-truth of lithium-ion batteries. A capacity estimation method is proposed based on a semi-supervised convolutional neural network (SS-CNN). This method can automatically extract features from battery partial-charge information for capacity estimation. Furthermore, a semi-supervised training strategy is developed to take advantage of the extra unlabeled sample, which can improve the generalization of the model and the accuracy of capacity estimation even in the presence of limited labeled data. Compared with artificial neural networks and convolutional neural networks, the proposed method is demonstrated to improve capacity estimation accuracy.

scholarly journals Artificial Neural Network Approach to Mobile Location Estimation in GSM Network

2017 ◽  
Vol 63 (1) ◽  
pp. 39-44 ◽  
Author(s):  
Longinus S. Ezema ◽  
Cosmas I. Ani
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Performance Metrics ◽  
Federal Communication Commission ◽  
Location Estimation ◽  
Location Based Services ◽  
Estimation Accuracy ◽  
Neural Network Approach ◽  
Mobile Location ◽  
Gsm Network

AbstractThe increase in utilisation of mobile location-based services for commercial, safety and security purposes among others are the key drivers for improving location estimation accuracy to better serve those purposes. This paper proposes the application of Levenberg Marquardt training algorithm on new robust multilayered perceptron neural network architecture for mobile positioning fitting for the urban area in the considered GSM network using received signal strength (RSS). The key performance metrics such as accuracy, cost, reliability and coverage are the major points considered in this paper. The technique was evaluated using real data from field measurement and the results obtained proved the proposed model provides a practical positioning that meet Federal Communication Commission (FCC) accuracy requirement.

scholarly journals Consensus-Based Distributed Target Tracking with False Data Injection Attacks over Radar Network

2021 ◽  
Vol 11 (10) ◽  
pp. 4564
Author(s):  
Yongtao Shui ◽  
Yu Wang ◽  
Yu Li ◽  
Yongzhi Shan ◽  
Naigang Cui ◽  
...  
Keyword(s):  
Parameter Estimation ◽  
State Estimation ◽  
Target Tracking ◽  
Estimation Error ◽  
Estimation Method ◽  
Estimation Accuracy ◽  
Consensus Method ◽  
False Data Injection ◽  
Radar Network ◽  
Global Consistency

For target tracking in radar network, any anomaly in a part of the system can quickly spread over the network and lead to tracking failures. False data injection (FDI) attacks can damage the state estimation mechanism by modifying the radar measurements with unknown and time-varying attack variables, therefore making traditional filters inapplicable. To tackle this problem, we propose a novel consensus-based distributed state estimation (DSE) method for target tracking with FDI attacks, which is effective even when all radars are under FDI attacks. First, a real-time residual-based detector is introduced to the DSE framework, which can effectively detect FDI attacks by analyzing the statistical properties of the residual. Secondly, a simple yet effective attack parameter estimation method is proposed to provide attack parameter estimation based on a pseudo-measurement equation, which has the advantage of decoupled estimation of state and attack parameters compared with augmented state filters. Finally, for timely attack mitigation and global consistency achievement, a novel hybrid consensus method is proposed which can compensate for the estimation error caused by FDI attacks and provide estimation accuracy improvement. The simulation results show that the proposed solution is effective and superior to the traditional DSE method for target tracking in the presence of FDI attacks.

scholarly journals End-to-end neural network architecture for fraud scoring in card payments

2018 ◽  
Vol 105 ◽  
pp. 175-181 ◽  
Author(s):  
Jon Ander Gómez ◽  
Juan Arévalo ◽  
Roberto Paredes ◽  
Jordi Nin
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Neural Network Architecture ◽  
End To End

scholarly journals A Hybrid End-to-End Approach Integrating Conditional Random Fields into CNNs for Prostate Cancer Detection on MRI

2020 ◽  
Vol 10 (1) ◽  
pp. 338 ◽  
Author(s):  
Paulo Lapa ◽  
Mauro Castelli ◽  
Ivo Gonçalves ◽  
Evis Sala ◽  
Leonardo Rundo
Keyword(s):  
Neural Network ◽  
Prostate Cancer ◽  
Random Fields ◽  
Network Architecture ◽  
Graphical Model ◽  
Conditional Random Fields ◽  
High Variability ◽  
Detection Methods ◽  
Peak Performance ◽  
End To End

Prostate Cancer (PCa) is the most common oncological disease in Western men. Even though a growing effort has been carried out by the scientific community in recent years, accurate and reliable automated PCa detection methods on multiparametric Magnetic Resonance Imaging (mpMRI) are still a compelling issue. In this work, a Deep Neural Network architecture is developed for the task of classifying clinically significant PCa on non-contrast-enhanced MR images. In particular, we propose the use of Conditional Random Fields as a Recurrent Neural Network (CRF-RNN) to enhance the classification performance of XmasNet, a Convolutional Neural Network (CNN) architecture specifically tailored to the PROSTATEx17 Challenge. The devised approach builds a hybrid end-to-end trainable network, CRF-XmasNet, composed of an initial CNN component performing feature extraction and a CRF-based probabilistic graphical model component for structured prediction, without the need for two separate training procedures. Experimental results show the suitability of this method in terms of classification accuracy and training time, even though the high-variability of the observed results must be reduced before transferring the resulting architecture to a clinical environment. Interestingly, the use of CRFs as a separate postprocessing method achieves significantly lower performance with respect to the proposed hybrid end-to-end approach. The proposed hybrid end-to-end CRF-RNN approach yields excellent peak performance for all the CNN architectures taken into account, but it shows a high-variability, thus requiring future investigation on the integration of CRFs into a CNN.

Sign in / Sign up

Export Citation Format

Share Document