scholarly journals Assignment of Focus Position with Convolutional Neural Networks in Adaptive Lens Based Axial Scanning for Confocal Microscopy

2022 ◽  
Vol 12 (2) ◽  
pp. 661
Author(s):  
Katharina Schmidt ◽  
Nektarios Koukourakis ◽  
Jürgen W. Czarske
Keyword(s):  
Neural Network ◽  
Applied Voltage ◽  
Laser Scanning ◽  
Degrees Of Freedom ◽  
Single Shot ◽  
Training Procedure ◽  
Focus Position ◽  
The Neural Network ◽  
Complex Adaptive ◽  
Axial Scanning

Adaptive lenses offer axial scanning without mechanical translation and thus are promising to replace mechanical-movement-based axial scanning in microscopy. The scan is accomplished by sweeping the applied voltage. However, the relation between the applied voltage and the resulting axial focus position is not unambiguous. Adaptive lenses suffer from hysteresis effects, and their behaviour depends on environmental conditions. This is especially a hurdle when complex adaptive lenses are used that offer additional functionalities and are controlled with more degrees of freedom. In such case, a common approach is to iterate the voltage and monitor the adaptive lens. Here, we introduce an alternative approach which provides a single shot estimation of the current axial focus position by a convolutional neural network. We use the experimental data of our custom confocal microscope for training and validation. This leads to fast scanning without photo bleaching of the sample and opens the door to automatized and aberration-free smart microscopy. Applications in different types of laser-scanning microscopes are possible. However, maybe the training procedure of the neural network must be adapted for some use cases.

Solution to the Forward and Inverse Kinematic Equations Using Multilayer Back-Propagation Neural Network

1994 ◽  
Author(s):  
Phani K. Nagarjuna ◽  
Athamaram H. Soni
Keyword(s):  
Neural Network ◽  
Degrees Of Freedom ◽  
Robot Manipulator ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Inverse Kinematic ◽  
Robot Arm ◽  
Kinematic Equation ◽  
Neural Network Approach ◽  
The Neural Network

Abstract The problem of inverse kinematics in Robotics, is a nonlinear mapping from a given cartesian coordinates to the desirable joint coordinates of the robot arm. It is found that an appropriately designed neural network can be trained to learn the non-linearity of the Inverse Kinematic Equation (IKE). We present an approach for solving the Forward Kinematic Equation (FKE) and the IKE by means of a Multi Layer Back-Propagation Neural Network (Rumelhart et al., 1986). The neural network approach is applied to a Two Degrees-of-Freedom (DOF) robot manipulator and the results are compared with those obtained using the analytical solution. The results obtained from the simulation of the neural network indicate a fairly accurate learning of the FKE and IKE by the Multi Layer Back-Propagation Neural Network.

Traveltime-table compression using artificial neural networks for Kirchhoff-migration processing of microseismic data

Geophysics ◽  
2020 ◽  
Vol 85 (5) ◽  
pp. U121-U128
Author(s):  
Serafim I. Grubas ◽  
Georgy N. Loginov ◽  
Anton A. Duchkov
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Network Architecture ◽  
Training Procedure ◽  
Velocity Models ◽  
Kirchhoff Migration ◽  
Microseismic Data ◽  
The Neural Network ◽  
Artificial Neural

Massive computation of seismic traveltimes is widely used in seismic processing, for example, for the Kirchhoff migration of seismic and microseismic data. Implementation of the Kirchhoff migration operators uses large precomputed traveltime tables (for all sources, receivers, and densely sampled imaging points). We have tested the idea of using artificial neural networks for approximating these traveltime tables. The neural network has to be trained for each velocity model, but then the whole traveltime table can be compressed by several orders of magnitude (up to six orders) to the size of less than 1 MB. This makes it convenient to store, share, and use such approximations for processing large data volumes. We evaluate some aspects of choosing neural-network architecture, training procedure, and optimal hyperparameters. On synthetic tests, we find a reasonably accurate approximation of traveltimes by neural networks for various velocity models. A final synthetic test shows that using the neural-network traveltime approximation results in good accuracy of microseismic event localization (within the grid step) in the 3D case.

Fuzzy Control of a Novel Type of Translational Meshing Motors

2010 ◽  
Vol 121-122 ◽  
pp. 1038-1043
Author(s):  
Wei Wang ◽  
Xin Jian Shan ◽  
Shi Min Wei
Keyword(s):  
Neural Network ◽  
Control System ◽  
Fuzzy Control ◽  
Fuzzy Controller ◽  
Torque Control ◽  
Training Procedure ◽  
Test Results ◽  
Model Reference Control ◽  
The Neural Network ◽  
Control Scheme

Owing to the nonlinear characteristic of a novel type of translational meshing motor with model uncertainties, a model reference control system which consists of a neural network and a fuzzy controller is used. The torque model is identified based on BP neural network, and then Fuzzy controller works as the controller. The description of the control system and training procedure of the neural network are given. The test results obtained for a torque control scheme suitable for the control of the motor are also presented to verify the effectiveness of the proposed nonlinear control scheme. It has been found that the fuzzy control system is able to work reliably.

scholarly journals Adaptive Neural Network Motion Control of Manipulators with Experimental Evaluations

2014 ◽  
Vol 2014 ◽  
pp. 1-13 ◽  
Author(s):  
S. Puga-Guzmán ◽  
J. Moreno-Valenzuela ◽  
V. Santibáñez
Keyword(s):  
Neural Network ◽  
Degrees Of Freedom ◽  
Gravitational Force ◽  
Tracking Accuracy ◽  
Adaptive Neural Network ◽  
Tracking Errors ◽  
The Neural Network ◽  
Proportional Derivative Controller ◽  
Velocity Tracking ◽  
Uniformly Bounded

A nonlinear proportional-derivative controller plus adaptive neuronal network compensation is proposed. With the aim of estimating the desired torque, a two-layer neural network is used. Then, adaptation laws for the neural network weights are derived. Asymptotic convergence of the position and velocity tracking errors is proven, while the neural network weights are shown to be uniformly bounded. The proposed scheme has been experimentally validated in real time. These experimental evaluations were carried in two different mechanical systems: a horizontal two degrees-of-freedom robot and a vertical one degree-of-freedom arm which is affected by the gravitational force. In each one of the two experimental set-ups, the proposed scheme was implemented without and with adaptive neural network compensation. Experimental results confirmed the tracking accuracy of the proposed adaptive neural network-based controller.

scholarly journals Decoupling Control for Three Degrees of Freedom Servo System Based on Neural Network

2013 ◽  
Vol 765-767 ◽  
pp. 1840-1843
Author(s):  
Zhen Bi Li ◽  
Bai Ting Zhao ◽  
Yuan Yuan Jiang
Keyword(s):  
Neural Network ◽  
Degrees Of Freedom ◽  
Inertial Navigation System ◽  
Servo System ◽  
Good Control ◽  
Decoupling Control ◽  
Simulating Experiment ◽  
Control Precision ◽  
The Neural Network ◽  
Three Degrees Of Freedom

Three degrees of freedom servo system (TDFSS) is one of the key equipments of inertial testing, such as evaluation of inertial navigation system and test of inertial components. It is a kind of servo system with some non-linearity and uncertainty. This thesis takes advantage of the characteristic of Neural network in approaching non-linear function, applies the Neural network on the three-axis simulator, provides a method for the TDFSS. Simulating experiment has been used to verify the advantage of the scheme and achieved completely decoupling control. The scheme gives good control precision, and it is simply structured and easily implemented.Introduction

scholarly journals MODELING QUANTUM MECHANICAL OBSERVERS VIA NEURAL-GLIAL NETWORKS

2012 ◽  
Vol 26 (09) ◽  
pp. 1250060 ◽  
Author(s):  
EIJI KONISHI
Keyword(s):  
Neural Network ◽  
Degrees Of Freedom ◽  
Quantum Mechanical ◽  
Classical Information ◽  
The Neural Network ◽  
Large N ◽  
Quantum Hamiltonian ◽  
Synaptic Junctions ◽  
The Brain ◽  
Novel Model

We investigate the theory of observers in the quantum mechanical world by using a novel model of the human brain which incorporates the glial network into the Hopfield model of the neural network. Our model is based on a microscopic construction of a quantum Hamiltonian of the synaptic junctions. Using the Eguchi–Kawai large N reduction, we show that, when the number of neurons and astrocytes is exponentially large, the degrees of freedom (d.o.f) of the dynamics of the neural and glial networks can be completely removed and, consequently, that the retention time of the superposition of the wavefunctions in the brain is as long as that of the microscopic quantum system of pre-synaptics sites. Based on this model, the classical information entropy of the neural-glial network is introduced. Using this quantity, we propose a criterion for the brain to be a quantum mechanical observer.

Neuroadaptive control of elastic-joint robots using robust performance enhancement

Robotica ◽  
2001 ◽  
Vol 19 (6) ◽  
pp. 619-629 ◽  
Author(s):  
C.J.B. Macnab ◽  
G.M.T. D'Eleuterio
Keyword(s):  
Neural Network ◽  
Degrees Of Freedom ◽  
Performance Enhancement ◽  
Control Law ◽  
Network Stability ◽  
Elastic Joint ◽  
Lyapunov Techniques ◽  
The Neural Network ◽  
Control Scheme ◽  
Neuroadaptive Control

A neuroadaptive control scheme for elastic-joint robots is proposed that uses a relatively small neural network. Stability is achieved through standard Lyapunov techniques. For added performance, robust modifications are made to both the control law and the weight update law to compensate for only approximate learning of the dynamics. The estimate of the modeling error used in the robust terms is taken directly from the error of the network in modeling the dynamics at the currant state. The neural network used is the CMAC-RBF Associative Memory (CRAM), which is a modification of Albus's CMAC network and can be used for robots with elastic degrees of freedom. This results in a scheme that is computationally practical and results in good performance.

Hybrid modelling of anaerobic wastewater treatment processes

2001 ◽  
Vol 43 (1) ◽  
pp. 43-50 ◽  
Author(s):  
A. Kamara ◽  
O. Bernard ◽  
A. Genovesi ◽  
D. Dochain ◽  
A. Benhammou ◽  
...  
Keyword(s):  
Neural Network ◽  
A Priori ◽  
Hybrid Approach ◽  
Training Procedure ◽  
Dynamic Representation ◽  
Knowledge Based ◽  
Treatment Processes ◽  
Kinetic Growth ◽  
The Neural Network ◽  
Component Training

This paper presents a hybrid approach for the modelling of an anaerobic digestion process. The hybrid model combines a feedforward network, describing the bacterial kinetics, and the a priori knowledge based on the mass balances of the process components. We have considered an architecture which incorporates the neural network as a static model of unmeasured process parameters (kinetic growth rate) and an integrator for the dynamic representation of the process using a set of dynamic differential equations. The paper contains a description of the neural network component training procedure. The performance of this approach is illustrated with experimental data.

scholarly journals Neural Networks Trained via Reinforcement Learning Stabilize Walking of a Three-Dimensional Biped Model With Exoskeleton Applications

2021 ◽  
Vol 8 ◽  
Author(s):  
Chujun Liu ◽  
Musa L. Audu ◽  
Ronald J. Triolo ◽  
Roger D. Quinn
Keyword(s):  
Neural Network ◽  
Degrees Of Freedom ◽  
Neural Control ◽  
Spinal Cord Injuries ◽  
Sagittal Plane ◽  
Three Dimensional ◽  
Lower Limb Exoskeleton ◽  
Neural Network Controller ◽  
The Neural Network ◽  
Improved Stability

Our group is developing a cyber-physical walking system (CPWS) for people paralyzed by spinal cord injuries (SCI). The current CPWS consists of a functional neuromuscular stimulation (FNS) system and a powered lower-limb exoskeleton for walking with leg movements in the sagittal plane. We are developing neural control systems that learn to assist the user of this CPWS to walk with stability. In a previous publication (Liu et al., Biomimetics, 2019, 4, 28), we showed a neural controller that stabilized a simulated biped in the sagittal plane. We are considering adding degrees of freedom to the CPWS to allow more natural walking movements and improved stability. Thus, in this paper, we present a new neural network enhanced control system that stabilizes a three-dimensional simulated biped model of a human wearing an exoskeleton. Results show that it stabilizes human/exoskeleton models and is robust to impact disturbances. The simulated biped walks at a steady pace in a range of typical human ambulatory speeds from 0.7 to 1.3 m/s, follows waypoints at a precision of 0.3 m, remains stable, and continues walking forward despite impact disturbances and adapts its speed to compensate for persistent external disturbances. Furthermore, the neural network controller stabilizes human models of different statures from 1.4 to 2.2 m tall without any changes to the control parameters. Please see videos at the following link: 3D biped walking control.

HARDWARE PROTOTYPES OF A BOOLEAN NEURAL NETWORK AND THE SIMULATED ANNEALING OPTIMIZATION METHOD

1996 ◽  
Vol 07 (01) ◽  
pp. 45-52
Author(s):  
JARKKO NIITTYLAHTI
Keyword(s):  
Neural Network ◽  
Simulated Annealing ◽  
Logic Gates ◽  
Optimization Method ◽  
General Purpose ◽  
Boolean Logic ◽  
Training Procedure ◽  
The Neural Network ◽  
Neural Network Hardware ◽  
Dynamically Configurable

Boolean Neural Network is a neural network that operates with binary weight values of “1” and “0”. Otherwise it is formally analogous to the Multilayer Perceptron (MLP).1 Simulated Annealing is a stochastic optimization method that is suitable for performing nonlinear multivariable optimization tasks. Training a Boolean Neural Network is a well-suited problem to this algorithm. However, the Simulated Annealing method is computationally heavy, which makes the training procedure slow. The training speed can be improved by using custom designed hardware for the whole system including the optimization method and the neural network. Hardware prototypes of a Boolean Neural Network and the Simulated Annealing optimization method have been designed using discrete components. The Boolean Neural Network implementation is basically a dynamically configurable feedforward network of Boolean logic gates of two inputs. The Simulated Annealing implementation is a general purpose hardware tool for multivariable optimization tasks. Here it is applied to do supervised training of the Boolean Neural Network hardware.

Sign in / Sign up

Export Citation Format

Share Document