Neural Network Based Approaches for Detecting Signals With Unknown Parameters

2007 ◽  
Author(s):  
David de la Mata-Moya ◽  
Pilar Jarabo-Amores ◽  
Manuel Rosa-Zurera ◽  
Raul Vicen-Bueno ◽  
Jose Carlos Nieto-Borge
Keyword(s):  
Neural Network ◽  
Unknown Parameters

scholarly journals Deep Residual Network with Sparse Feedback for Image Restoration

2018 ◽  
Vol 8 (12) ◽  
pp. 2417 ◽  
Author(s):  
Zhenyu Guo ◽  
Yujuan Sun ◽  
Muwei Jian ◽  
Xiaofeng Zhang
Keyword(s):  
Neural Network ◽  
Image Restoration ◽  
Deep Neural Network ◽  
Convergence Speed ◽  
Fast Convergence ◽  
Experimental Results ◽  
Residual Network ◽  
Unknown Parameters ◽  
Noisy Images ◽  
Blurred Images

A deep neural network is difficult to train due to a large number of unknown parameters. To increase trainable performance, we present a moderate depth residual network for the restoration of motion blurring and noisy images. The proposed network has only 10 layers, and the sparse feedbacks are added in the middle and the last layers, which are called FbResNet. FbResNet has fast convergence speed and effective denoising performance. In addition, it can also reduce the artificial Mosaic trace at the seam of patches, and visually pleasant output results can be produced from the blurred images or noisy images. Experimental results show the effectiveness of our designed model and method.

Applying Neural Networks to the Solution of the Inverse Heat Conduction Problem in a Gun Barrel

2008 ◽  
Vol 130 (3) ◽  
Author(s):  
Y. Hwang ◽  
S. Deng
Keyword(s):  
Neural Network ◽  
Heat Flux ◽  
Heat Conduction ◽  
Heat Conduction Problem ◽  
Nonlinear Function ◽  
Inverse Heat Conduction ◽  
Unknown Parameters ◽  
Acceptable Error ◽  
Gun Barrel ◽  
Inverse Heat Conduction Problems

The primary cause of gun barrel erosion is the heat generated by the shell as its travels along the barrel. Therefore, calculating the heat flux input to the gun bore is very important when investigating wear problems in the gun barrel and examining its thermomechanical properties. This paper employs the continuous-time analog Hopfield neural network (CHNN) to compute the temperature distribution in various forward heat conduction problems. An efficient technique is then proposed for the solution of inverse heat conduction problems using a three-layered backpropagation neural network (BPN). The weak generalization capacity of BPN networks when applied to the solution of nonlinear function approximations is improved by employing the Bayesian regularization algorithm. The CHNN scheme is used to calculate the temperature in a 155mm gun barrel and the trained BPN is then used to estimate the heat flux of the inner surface of the barrel. The results show that the proposed neural network analysis method successfully solves forward heat conduction problems and is capable of predicting the unknown parameters in inverse problems with an acceptable error.

A Reinforcement Learning Neural Network for Robotic Manipulator Control

2018 ◽  
Vol 30 (7) ◽  
pp. 1983-2004 ◽  
Author(s):  
Yazhou Hu ◽  
Bailu Si
Keyword(s):  
Neural Network ◽  
Reinforcement Learning ◽  
Performance Index ◽  
Robotic Manipulator ◽  
The State ◽  
Unknown Parameters ◽  
Proposed Model ◽  
Action Network ◽  
Action Policy ◽  
The Stability

We propose a neural network model for reinforcement learning to control a robotic manipulator with unknown parameters and dead zones. The model is composed of three networks. The state of the robotic manipulator is predicted by the state network of the model, the action policy is learned by the action network, and the performance index of the action policy is estimated by a critic network. The three networks work together to optimize the performance index based on the reinforcement learning control scheme. The convergence of the learning methods is analyzed. Application of the proposed model on a simulated two-link robotic manipulator demonstrates the effectiveness and the stability of the model.

A Neural Network Method for Miniature Unmanned Helicopter Heading Control

2012 ◽  
Vol 468-471 ◽  
pp. 93-96
Author(s):  
Meng Bai ◽  
Min Hua Li
Keyword(s):  
Neural Network ◽  
Control Method ◽  
Network Control ◽  
Neural Network Control ◽  
Unmanned Helicopter ◽  
Unknown Parameters ◽  
Neural Network Controller ◽  
The Neural Network ◽  
Heading Control ◽  
Helicopter Aerodynamics

A neural network control method for heading control of miniature unmanned helicopter is proposed. For the complexity of miniature helicopter aerodynamics, it is difficult to identify the unknown parameters of yaw dynamics model. To design heading controller of miniature helicopter without modelling yaw dynamics, BP neural network is designed as heading controller, which is trained by collected flight data. By training, the neural network controller can learn the artificial operation strategy and realize the heading control of miniature unmanned helicopter. Simulation results demonstrate the validity of the proposed neural network control method.

scholarly journals Neural-network fusion processing and inverse mapping to combine multi-sensor satellite data and to analyze significant features

2021 ◽  
Author(s):  
Gunjan Joshi ◽  
Ryo Natsuaki ◽  
Akira Hirose
Keyword(s):  
Neural Network ◽  
Satellite Data ◽  
Optical Sensors ◽  
Remote Sensing Data ◽  
Activation Function ◽  
Significant Feature ◽  
Unknown Parameters ◽  
Inverse Mapping ◽  
Observation Area ◽  
Synthetic Aperture Radars

<div>In the last decade, the increase in the number of active and passive earth observation satellites has provided us with more remote sensing data. This fact has led to increased interests in the field of fusion of the different satellite data since some of the satellites have properties complementary to one another. Fusion techniques can improve the estimation in areas of interest (AOIs) by using complementary information and inferring unknown parameters. However, when the observation area is large, extensive human labor and domain expertise are required for processing and analysis. Thus, we propose a neural network which combines and analyzes the data obtained from synthetic aperture radars (SAR) and optical sensors. The neural network employs a modified logarithmic activation function, unlike conventional networks, to realize inverse mapping for significant feature analysis based on dynamics consistent with its forward processing. In this paper, we focus on earthquake damage detection by dealing with the data of the 2018 Sulawesi earthquake in Indonesia. The fusion-based results show increased classification accuracy compared to the results of independent sensors. We further attempt to understand which input features are the significant contributors for which classification outputs by inverse-mapping in the data fusion neural network. We observe that inverse mapping shows reasonable explanations in a consistent manner. It also indicates contributions of features different from straightforward counterparts, namely pre- and post-seismic features, in the detection of particular classes.</div>

scholarly journals Variational Inference Formulation for a Model-Free Simulation of a Dynamical System with Unknown Parameters by a Recurrent Neural Network

2021 ◽  
Vol 43 (2) ◽  
pp. A1305-A1335
Author(s):  
Kyongmin Yeo ◽  
Dylan E. C. Grullon ◽  
Fan-Keng Sun ◽  
Duane S. Boning ◽  
Jayant R. Kalagnanam
Keyword(s):  
Neural Network ◽  
Dynamical System ◽  
Recurrent Neural Network ◽  
Variational Inference ◽  
Unknown Parameters ◽  
Model Free

scholarly journals Implementation of Neural Predictive Control To Distillation Column

REAKTOR ◽  
2017 ◽  
Vol 10 (1) ◽  
pp. 24
Author(s):  
S. Anwari
Keyword(s):  
Neural Network ◽  
Predictive Control ◽  
Distillation Column ◽  
Controller Design ◽  
Fundamental Equation ◽  
Small Range ◽  
Unknown Parameters ◽  
Predictive Controller ◽  
Multivariable Models ◽  
Reliable Solution

This paper presents a neural predictive controller that is applied to distillation column. Distillation columns represent complex multivariable system, with fast and slow dynamics, significant interactions and directionality. A phenomenological model (i.e. a model derived from fundamental equation like mass and energy balance) of a distillation column is very complicated. For this reason, classical linear controller, such as PID (Proportional, Integral and Derivative) controller, will provide robustness only over relatively small range operation because of complexity and operation without lack of robustness. In this work, a neural network is developed for modeling and controlling a distillation column based on measured input-outputdata pairs. In distillation column, a neural network is trained on the unknown parameters of the system. The resulting implementationof the neural predictive controller is able to eliminate the most significant obstacles encountered in conventional predictive control application by facilitating  the development of complex multivariable models and providing a rapid, reliable solution to the control algorithm. Controller design and implementation are illustrated for a plant frequently referred to in the literature. Result are given for simulation experiments, which demonstrate the advantage of the neural based predictive controller both at the transient region and at the steady state region to overcome any overshoots.Keywords : neural predictive controller, distillation column, complex multivariable models

scholarly journals Adaptive Neural Backstepping Sliding Mode Heading Control for Underactuated Ships with Drift Angle and Ship-Bank Interaction

2020 ◽  
Vol 2020 ◽  
pp. 1-18
Author(s):  
Xue Han
Keyword(s):  
Neural Network ◽  
Sliding Mode Control ◽  
Control Algorithm ◽  
Sliding Mode ◽  
Adaptive Backstepping ◽  
Unknown Parameters ◽  
Sea State ◽  
Environmental Disturbances ◽  
Mode Control ◽  
Backstepping Sliding Mode Control

In order to track the desired path under unknown parameters and environmental disturbances, an adaptive backstepping sliding mode control algorithm with a neural estimator is proposed for underactuated ships considering both ship-bank interaction effect and shift angle. Using the features of radial basis function neural network, which can approximate arbitrary function, the unknown parameters of the ship model and environmental disturbances are estimated. The trajectory tracking errors include stabilizing sway and surge velocities errors. Based on the Lyapunov stability theory, the tracking error will converge to zero and the system is asymptotically stable. The controlled trajectory is contractive and asymptotically tends to the desired position and attitude. The results show that compared with the basic sliding mode control algorithm, the overshoot of the adaptive backstepping sliding mode control with neural estimator is smaller and the regulation time of the system is shorter. The ship can adjust itself and quickly reach its desired position under disturbances. This shows that the designed RBF neural network observer can track both the mild level 3 sea state and the bad level 5 sea state, although the wave disturbance has relatively fast time-varying disturbance. The algorithm has good tracking performance and can realize the accurate estimation of wave disturbance, especially in bad sea conditions.

Sign in / Sign up

Export Citation Format

Share Document