Assisted Research and Optimization of the Proper Neural Network Solving the Inverse Kinematics Problem

2012 ◽  
Vol 463-464 ◽  
pp. 827-832
Author(s):  
Adrian Olaru ◽  
Serban Olaru ◽  
Dan Paune ◽  
Oprean Aurel
Keyword(s):  
Neural Network ◽  
Time Delay ◽  
Inverse Kinematics ◽  
Hyperbolic Tangent ◽  
Variable Parameters ◽  
The Neural Network ◽  
Variable Step ◽  
Inverse Kinematics Problem ◽  
Sensitive Function ◽  
Target Data

Finding the better solution of the neural network design to solve the inverse kinematics problem with the minimum of the trajectory errors is very difficult, because there are many variable parameters and many redundant solutions. The presented paper show the assisted research of the influences of some more important parameters to the final end-effector trajectory errors of the proposed neural network model solving the inverse kinematics problem. We were been studied the number of neurons in each layers, the sensitive function for the first and second layer, the magnifier coefficient of the trajectory error, the variable step of the time delay and the position of this block, the different cases of target data and the case when the hidden target data were adjusted. All obtained results were been verified by applying the proper direct kinematics virtual LabVIEW instrumentation. Finally we were obtained one optimal Sigmoid Bipolar Hyperbolic Tangent Neural Network with Time Delay and Recurrent Links (SBHTNN(TDRL)) type, what can be used to solve the inverse kinematics problem with maximum 4% of trajectory errors.

Controlling of the 3D Space Trajectory of the Multi Robots Applications by Using the Proper Iterative Pseudo Inverse Jacobian Neural Network Matrix Method

2016 ◽  
Vol 841 ◽  
pp. 227-233
Author(s):  
Adrian Olaru ◽  
Serban Olaru ◽  
Niculae Mihai ◽  
Doru Bardac
Keyword(s):  
Neural Network ◽  
Matrix Method ◽  
Inverse Kinematics ◽  
Parallel Robot ◽  
Coupled Method ◽  
Space Curves ◽  
3D Space ◽  
The Neural Network ◽  
Inverse Kinematics Problem ◽  
Pseudo Inverse

In many applications we used the multi robots with the central coordination of the 3D space trajectory. In the controlling of the space movement of the end effecter of the all robots from this type of applications and the robot’s joints one of the most important problem is to solve the forward and inverse kinematics, that is different from the single robot application. It is important to know with the extreme precision the joints relative displacements of all robots. One of the most precise method to solve the inverse kinematics problem in the robots with redundant chain is the complex coupled method of the neural network with Iterative Jacobian Pseudo Inverse method. In this paper was proposed and used the proper coupled method Iterative Pseudo Inverse Jacobian Matrix Method (IPIJMM) with Sigmoid Bipolar Hyperbolic Tangent Neural Network with Time Delay and Recurrent Links (SBHTNN-TDRL). The paper contents the mathematical matrix model of the forward kinematics of multiple robots applications, mathematical model of the proper iterative algorithm and all proper virtual LabVIEW instrumentation, to obtain the space conventional and unconventional curves in different Euller planes for one case study of three simultaneously robots movement with extreme precision of the end-effecter less than 0.001mm. The paper shown how can be changed the multi robots application in to one application with parallel robot structure with three independent robots. The presented method and the virtual instrumentations (VI) are generally and they can be used in all other robots application and for all other conventional and unconventional space curves.

Assisted Optimization of the Servo Driving by Using Virtual Labview Instrumentation and the Elementary Transfer Functions and Neural Network Methods

2013 ◽  
Vol 325-326 ◽  
pp. 970-983 ◽  
Author(s):  
Adrian Olaru ◽  
Serban Olaru ◽  
Aurel Oprean
Keyword(s):  
Neural Network ◽  
Time Delay ◽  
Dynamic Behavior ◽  
Transfer Functions ◽  
Error Function ◽  
Closed Loops ◽  
The Neural Network ◽  
Convergence Process ◽  
On Line ◽  
Sensitive Function

In the optimisation stage of the systems one of the more important step is the optimisation of the dynamic behavior of all elements with priority the elements what have the slow frequency, like motors. The paper try to show how will be possible to optimise very easily the dynamic behavior of elements and systems, using LabVIEW propre instrumentation and the application of the transfer functions and neural tnetwork theory. By appling the virtual LabVIEW instrumentation is possible to choose on-line the optimal values for each constructive and functional parameters of the elements and the systems to obtain one good dynamic answer: maximal acceleration without vibration, minimum answer time and maximal precision. The paper presents some of the more important used transfer functions in the assisted analyse of the elements and systems and some practical results of the assisted optimisation by using the neural network method. In the research were been used some different way to optimize the convergence process, for example: using one time- delay of the first and second output from the neural layers; using the recursive link and time- delay; using the bipolar sigmoid hyperbolic tangent sensitive function replacing the sigmoid simple sensitive function. By on-line simulation of the neural network was possible to know what will be the influences of all network parameters like the input data, weight, biases matrix, sensitive functions, closed loops and time- delay, to the gradient errors, in a convergence process. In the optimization research we used the minimization of the gradient error function between the output and the target.

Optimize the Satellite Orientation by Using the Inertial Pulse Method, Intelligent Damper, Dynamics, Kinematics and Proper Neural Network

2012 ◽  
Vol 232 ◽  
pp. 665-673
Author(s):  
Adrian Olaru
Keyword(s):  
Neural Network ◽  
Inverse Kinematics ◽  
Inverse Dynamics ◽  
Pulse Method ◽  
Good Precision ◽  
Small Satellite ◽  
The Neural Network ◽  
Control Orientation ◽  
Inverse Kinematics Problem ◽  
Electrical Motors

In the paper are shown one assisted method of the control orientation inertial mechanism of the small satellite. It is presented some of the general components of the small satellite and the possibility to orientation them between three axes using the inertial pulse of the electrical motors. The angular space position was determines by command of the final point and using the proper neural network to solve the inverse kinematics problem with the minimum of the errors. The precision of the space orientation position was increased by using in the proper controlling schema a direct kinematics module, direct and inverse dynamics, intelligent damper and one multiple inertial pulse method what was described in the paper. Finally was obtained one very good precision, before 2%.The neural network, the inertial pulse method, the results and the virtual LabVIEW instrumentation could be used in many other researches on the field.

Assisted Research of the Dynamic Neural Networks with Time-Delays and Recurrent Links

2012 ◽  
Vol 463-464 ◽  
pp. 1094-1097 ◽  
Author(s):  
Adrian Olaru ◽  
Serban Olaru ◽  
Dan Paune
Keyword(s):  
Neural Network ◽  
Time Delay ◽  
Error Function ◽  
Virtual Instrumentation ◽  
Dynamic Neural Networks ◽  
Network Parameters ◽  
The Neural Network ◽  
Convergence Process ◽  
On Line ◽  
Sensitive Function

The paper showed the assisted research of one new model of digital dynamic neural network by using the LabVIEW proper virtual instrumentation and proper mathematical model. In the research were used some different way to optimize the convergence process, for example: using one time- delay of the first and second output from the neural layers; using the recursive link and time- delay; using the bipolar sigmoid hyperbolic tangent sensitive function replacing the sigmoid simple sensitive function. By on-line simulation of the neural network it is possible to know what will be the influences of all network parameters like the input data, weight, biases matrix, sensitive functions, closed loops and time- delay, to the gradient errors, in a convergence process. By on-line using the proper virtual LabVIEW instrumentation, were established some influences of the network parameters: number of input vector data, number of neurons in each layers, to the number of iterations before canceled the mean square error to the target. In the optimization research we used the minimization of the gradient error function between the output and the target.

Proper Assisted Research Method Solving of the Robots Inverse Kinematics Problem

2014 ◽  
Vol 555 ◽  
pp. 135-146 ◽  
Author(s):  
Adrian Olaru ◽  
Serban Olaru ◽  
Niculae Mihai
Keyword(s):  
Inverse Kinematics ◽  
Descent Method ◽  
Forward Kinematics ◽  
Hyperbolic Tangent ◽  
End Effector ◽  
Variable Parameters ◽  
Space Curves ◽  
Complete Study ◽  
Inverse Kinematics Problem ◽  
Study Case

Finding the better solution of the inverse kinematics problem, with the minimum of the trajectory errors, is very difficult because there are many variable parameters and many redundant solutions. The presented paper show the assisted solving of the inverse kinematics with the goal to minimize the final end-effector trajectory errors, by optimizing the distance between the and-effector final position and the target. All obtained results were been verified by applying the proper forward kinematics virtual LabVIEW instrumentation. The paper tries to answer at the inverse kinematics problem for one known mathematical trajectory and identifying the cinematic errors after the establishing and applying the proper assisted solving method using the Cycle Coordinate Descent Method coupled to the proper Neural Network Sigmoid Bipolar Hyperbolic Tangent (CCDM-SBHTNN). We were shown one complete study case to obtain one circle space trajectory using one arm type robot fixed on the ceiling. The presented method is general and can be used in all other robots types and in all other conventional and unconventional space curves.

scholarly journals Cerebellum-inspired neural network solution of the inverse kinematics problem

2015 ◽  
Vol 109 (6) ◽  
pp. 561-574 ◽  
Author(s):  
Mitra Asadi-Eydivand ◽  
Mohammad Mehdi Ebadzadeh ◽  
Mehran Solati-Hashjin ◽  
Christian Darlot ◽  
Noor Azuan Abu Osman
Keyword(s):  
Neural Network ◽  
Inverse Kinematics ◽  
Network Solution ◽  
Inverse Kinematics Problem

The Research on Recognition of the Irregular Objects

2012 ◽  
Vol 241-244 ◽  
pp. 1602-1607
Author(s):  
Guang Hai Han ◽  
Xin Jun Ma
Keyword(s):  
Neural Network ◽  
Bp Neural Network ◽  
Convergence Speed ◽  
Know How ◽  
The Neural Network ◽  
Variable Step ◽  
Adaptive Momentum ◽  
The Way

It usually need different ways to process different objects in the manufacturing, Therefore, firstly we need to distinguish the categories of objects to be processed, then the machine will know how to deal with the objects. In order to automatically recognize the category of the irregular object, this paper extracted the improved Hu's moments of each object as the feature by the way of processing images of the working platform that the irregular objects are putting on. This paper adopts the variable step BP neural network with adaptive momentum factor as the classifier. The experiment shows that this method can effectively distinguish different irregular objects, and during the training of the neural network, it has faster convergence speed and better approximation compared with the traditional BP neural network

Gray-Box Approach for Fault Detection of Dynamical Systems

2003 ◽  
Vol 125 (3) ◽  
pp. 451-454 ◽  
Author(s):  
Han G. Park ◽  
Michail Zak
Keyword(s):  
Neural Network ◽  
Time Delay ◽  
Fault Detection ◽  
Stochastic Model ◽  
Deterministic Model ◽  
Feed Forward Neural Network ◽  
The Neural Network ◽  
Simulated System ◽  
Auto Regressive ◽  
Nonlinear Components

We present a fault detection method called the gray-box. The term “gray-box” refers to the approach wherein a deterministic model of system, i.e., “white box,” is used to filter the data and generate a residual, while a stochastic model, i.e., “black-box” is used to describe the residual. The residual is described by a three-tier stochastic model. An auto-regressive process, and a time-delay feed-forward neural network describe the linear and nonlinear components of the residual, respectively. The last component, the noise, is characterized by its moments. Faults are detected by monitoring the parameters of the auto-regressive model, the weights of the neural network, and the moments of noise. This method is demonstrated on a simulated system of a gas turbine with time delay feedback actuator.

Dynamics of Neural Networks as Nonlinear Systems with Several Equilibria

2011 ◽  
pp. 331-357 ◽  
Author(s):  
Daniela Danciu
Keyword(s):  
Neural Network ◽  
Dynamical System ◽  
Neural Networks ◽  
Time Delay ◽  
Dynamical Systems ◽  
Qualitative Properties ◽  
The Neural Network ◽  
The Neural Networks ◽  
Global Asymptotics ◽  
Dynamical Properties

Neural networks—both natural and artificial, are characterized by two kinds of dynamics. The first one is concerned with what we would call “learning dynamics”. The second one is the intrinsic dynamics of the neural network viewed as a dynamical system after the weights have been established via learning. The chapter deals with the second kind of dynamics. More precisely, since the emergent computational capabilities of a recurrent neural network can be achieved provided it has suitable dynamical properties when viewed as a system with several equilibria, the chapter deals with those qualitative properties connected to the achievement of such dynamical properties as global asymptotics and gradient-like behavior. In the case of the neural networks with delays, these aspects are reformulated in accordance with the state of the art of the theory of time delay dynamical systems.

Sign in / Sign up

Export Citation Format

Share Document