scholarly journals Comparison of Different Training Algorithms for the Leg Extension Training with an Industrial Robot

2018 ◽  
Vol 4 (1) ◽  
pp. 17-20 ◽  
Author(s):  
Maike Ketelhut ◽  
Fabian Göll ◽  
Björn Braunstein ◽  
Kirsten Albracht ◽  
Dirk Abel
Keyword(s):  
Industrial Robot ◽  
Force Plate ◽  
Training Algorithms ◽  
Training Algorithm ◽  
Systematic Analysis ◽  
Constant Weight ◽  
Leg Extension ◽  
Motion Trajectories ◽  
High Payload ◽  
Robotic Research

AbstractIn the past, different training scenarios have been developed and implemented on robotic research platforms, but no systematic analysis and comparison have been done so far. This paper deals with the comparison of an isokinematic (motion with constant velocity) and an isotonic (motion against constant weight) training algorithm. Both algorithms are designed for a robotic research platform consisting of a 3D force plate and a high payload industrial robot, which allows leg extension training with arbitrary six-dimensional motion trajectories. In the isokinematic as well as the isotonic training algorithm, individual paths are defined i n C artesian s pace by sufficient s upport p oses. I n t he i sotonic t raining s cenario, the trajectory is adapted to the measured force as the robot should only move along the trajectory as long as the force applied by the user exceeds a minimum threshold. In the isotonic training scenario however, the robot’s acceleration is a function of the force applied by the user. To validate these findings, a simulative experiment with a simple linear trajectory is performed. For this purpose, the same force path is applied in both training scenarios. The results illustrate that the algorithms differ in the force dependent trajectory adaption.

scholarly journals Forecasting Wind Power Generation Using Artificial Neural Network: “Pawan Danawi”—A Case Study from Sri Lanka

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Amila T. Peiris ◽  
Jeevani Jayasinghe ◽  
Upaka Rathnayake
Keyword(s):  
Neural Network ◽  
Power Generation ◽  
Wind Power ◽  
Wind Farm ◽  
Mean Squared Error ◽  
Wind Power Generation ◽  
Training Algorithms ◽  
Training Algorithm ◽  
Squared Error ◽  
Ann Models

Wind power, as a renewable energy resource, has taken much attention of the energy authorities in many countries, as it is used as one of the major energy sources to satisfy the ever-increasing energy demand. However, careful attention is needed in identifying the wind power potential in a particular area due to climate changes. In this sense, forecasting both wind power generation and wind power potential is essential. This paper develops artificial neural network (ANN) models to forecast wind power generation in “Pawan Danawi”, a functioning wind farm in Sri Lanka. Wind speed, wind direction, and ambient temperature of the area were used as the independent variable matrices of the developed ANN models, while the generated wind power was used as the dependent variable. The models were tested with three training algorithms, namely, Levenberg-Marquardt (LM), Scaled Conjugate Gradient (SCG), and Bayesian Regularization (BR) training algorithms. In addition, the model was calibrated for five validation percentages (5% to 25% in 5% intervals) under each algorithm to identify the best training algorithm with the most suitable training and validation percentages. Mean squared error (MSE), coefficient of correlation (R), root mean squared error ratio (RSR), Nash number, and BIAS were used to evaluate the performance of the developed ANN models. Results revealed that all three training algorithms produce acceptable predictions for the power generation in the Pawan Danawi wind farm with R > 0.91, MSE < 0.22, and BIAS < 1. Among them, the LM training algorithm at 70% of training and 5% of validation percentages produces the best forecasting results. The developed models can be effectively used in the prediction of wind power at the Pawan Danawi wind farm. In addition, the models can be used with the projected climatic scenarios in predicting the future wind power harvest. Furthermore, the models can acceptably be used in similar environmental and climatic conditions to identify the wind power potential of the area.

scholarly journals Learning in Feedforward Neural Networks Accelerated by Transfer Entropy

Entropy ◽  
2020 ◽  
Vol 22 (1) ◽  
pp. 102 ◽  
Author(s):  
Adrian Moldovan ◽  
Angel Caţaron ◽  
Răzvan Andonie
Keyword(s):  
Neural Networks ◽  
Time Series ◽  
Information Transfer ◽  
Feedforward Neural Networks ◽  
Theoretical Method ◽  
Transfer Entropy ◽  
Training Algorithms ◽  
Training Algorithm ◽  
Neural Connections ◽  
Feedback Connections

Current neural networks architectures are many times harder to train because of the increasing size and complexity of the used datasets. Our objective is to design more efficient training algorithms utilizing causal relationships inferred from neural networks. The transfer entropy (TE) was initially introduced as an information transfer measure used to quantify the statistical coherence between events (time series). Later, it was related to causality, even if they are not the same. There are only few papers reporting applications of causality or TE in neural networks. Our contribution is an information-theoretical method for analyzing information transfer between the nodes of feedforward neural networks. The information transfer is measured by the TE of feedback neural connections. Intuitively, TE measures the relevance of a connection in the network and the feedback amplifies this connection. We introduce a backpropagation type training algorithm that uses TE feedback connections to improve its performance.

scholarly journals A New Type of Industrial Robot Trajectory Generation Component Based on Motion Modularity Technology

2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Zhaoming Liu ◽  
Nailong Liu ◽  
Hongwei Wang ◽  
Shen Tian ◽  
Ning Bai ◽  
...  
Keyword(s):  
Industrial Robot ◽  
Trajectory Generation ◽  
Motion Trajectories ◽  
Robot Controller ◽  
Robot Trajectory ◽  
Fieldbus System ◽  
Sampling Points ◽  
Dynamic Movement Primitives ◽  
Complex Trajectories ◽  
The Brain

Motion modularity is the main method of motion control for higher animals. That means the complex movements of the muscles are made up of basic motion primitives, and the brain or central nervous system does not care about the specific details of the movement. However, the industrial robot control system does not adopt the technical roadmap of motion modularity, it generates complex trajectories by providing a large number of sampling points. This approach is equivalent to using the brain to directly guide the specific movement of the muscle and has to rely on a faster Fieldbus system to obtain complex motion trajectories. This work constructs a modularized industrial robot trajectory generation component based on Dynamic Movement Primitives (DMP) theory. With this component, the robot controller can generate complex trajectories without increasing the sampling points and can obtain good trajectory accuracy. Finally, the rationality of this system is proved by simulations and experiments.

Control System Design for High Payload Industrial Robot via High Speed Communication Bus and Real-Time System

2011 ◽  
Vol 464 ◽  
pp. 272-278 ◽  
Author(s):  
Wei You ◽  
Min Xiu Kong ◽  
Li Ning Sun ◽  
Chan Chan Guo
Keyword(s):  
Control System ◽  
Real Time ◽  
High Speed ◽  
Industrial Robot ◽  
Engineering Application ◽  
Industrial Robots ◽  
Centralized Control ◽  
Time System ◽  
Real Time System ◽  
High Payload

In this paper, aiming at solving the problems of dynamic coupling effects and flexibility of joints and links, a kind of control system specialized for high payload industrial robots is proposed . After the comparisons between the control systems in all kinds of robots and numerical machines, industrial PC with TwinCAT real-time system is chosen as the motion control unit, EtherCAT is used for command transmitting. The whole control system has a decoupled and centralized control structure. The proposed control system is applied in control of a kind of high payload material handling robots with complex compound control algorithms. The final results shows that the control commands can be easily calculated and transmitted in one sample unit. The proposed control scheme is meaningful to real engineering application.

Isokinematic leg extension training with an industrial robot

2016 ◽  
Author(s):  
Melanie Kolditz ◽  
Thivaharan Albin ◽  
Kirsten Albracht ◽  
Gert-Peter Bruggemann ◽  
Dirk Abel
Keyword(s):  
Industrial Robot ◽  
Leg Extension

scholarly journals A Framework for Selection of Training Algorithm of Neuro-Statistic Model for Prediction of Pig Breeds in India

2020 ◽  
Vol 08 (01) ◽  
pp. 153-175
Author(s):  
Satyendra Nath Mandal ◽  
Pritam Ghosh ◽  
Nanigopal Shit ◽  
Dilip Kumar Hajra ◽  
Santanu Banik
Keyword(s):  
Neural Network ◽  
Statistical Parameter ◽  
Training Algorithms ◽  
Training Algorithm ◽  
Algorithm Selection ◽  
Test Set ◽  
Pig Breeds ◽  
Validation Set ◽  
Hidden Neurons ◽  
Selection Of

Various training algorithms are used in artificial neural networks for updating the weights during training the network. But, the selection of the appropriate training algorithm is dependent on the input–output mapping of dataset for which the network is constructed. In this paper, a framework has been proposed consisting of five modules to select the optimal training algorithm for predicting pig breeds from their images. The individual pig images from five pig-breeds have been captured using inbuilt camera of mobile phone and the contour of pig has been segmented from each captured image by HUE-based segmentation algorithm. In Statistical Parameter and Color Component retrieval module, parameters like entropy, standard deviation, variance, mean, median, and mode and color properties like hue, saturation, value (HSV) extracted from the content of each segmented image. Values of all extracted parameters have been transferred into Training Algorithm Selection Module. In this module, a fitting neural network with different numbers of hidden neurons has been executed by feeding all extracted values from pig images for mapping their breeds. Ten training algorithms have been applied on the same extracted dataset separately for five epochs each keeping other network parameters constants. The mean square error (MSE) and correlation coefficient ([Formula: see text]) for the validation set have been calculated after adjustment of weights and biases in each connection of the neurons. One training algorithm among 10 and its suitable number of hidden neurons has been selected based on comparative analysis for getting lower MSE and higher [Formula: see text] in the validation set. Then, the fitting network with selected training algorithm has been run on the same extracted datasets until the stopping condition is reached. Then the test set images are fed into the network and the network output has been categorized to class which has been assigned to each breed of pig in Breed Prediction Module. The proposed framework has been able to predict breeds with 96.00% accuracy, achieved by the trial with 50 images of the test set. It may be concluded that the Neuro Statistic Neural Network model may be used for breed prediction of pigs by using images of individual pigs.

scholarly journals Empirical analysis of software quality prediction using a TRAINBFG algorithm

2018 ◽  
Vol 7 (2.6) ◽  
pp. 259 ◽  
Author(s):  
Saumendra Pattnaik ◽  
Binod Kumar Pattanayak
Keyword(s):  
Neural Network ◽  
Software Quality ◽  
Fuzzy Model ◽  
Rule Base ◽  
Training Algorithms ◽  
Training Algorithm ◽  
Data Set ◽  
The Neural Network ◽  
Neuro Fuzzy ◽  
Neural Network Input

Software quality plays a major role in software fault proneness. That’s why prediction of software quality is essential for measuring the anticipated faults present in the software. In this paper we have proposed a Neuro-Fuzzy model for prediction of probable values for a predefined set of software characteristics by virtue of using a rule base. In course of it, we have used several training algorithms among which TRAINBFG algorithm is observed to be the best one for the purpose. There are various training algorithm available in MATLAB for training the neural network input data set. The prediction using fuzzy logic and neural network provides better result in comparison with only neural network. We find out from our implementation that TRAINBFG algorithm can provide better predicted value as compared to other algorithm in MATLAB. We have validated this result using the tools like SPSS and MATLAB. 

scholarly journals Investigating the performance of neural network backpropagation algorithms for TEC estimations using South African GPS data

2012 ◽  
Vol 30 (5) ◽  
pp. 857-866 ◽  
Author(s):  
J. B. Habarulema ◽  
L.-A. McKinnell
Keyword(s):  
Neural Network ◽  
South African ◽  
Electron Content ◽  
Training Algorithms ◽  
Network Simulator ◽  
Training Algorithm ◽  
Total Electron ◽  
Unseen Data ◽  
Time Required ◽  
Gps Observations

Abstract. In this work, results obtained by investigating the application of different neural network backpropagation training algorithms are presented. This was done to assess the performance accuracy of each training algorithm in total electron content (TEC) estimations using identical datasets in models development and verification processes. Investigated training algorithms are standard backpropagation (SBP), backpropagation with weight delay (BPWD), backpropagation with momentum (BPM) term, backpropagation with chunkwise weight update (BPC) and backpropagation for batch (BPB) training. These five algorithms are inbuilt functions within the Stuttgart Neural Network Simulator (SNNS) and the main objective was to find out the training algorithm that generates the minimum error between the TEC derived from Global Positioning System (GPS) observations and the modelled TEC data. Another investigated algorithm is the MatLab based Levenberg-Marquardt backpropagation (L-MBP), which achieves convergence after the least number of iterations during training. In this paper, neural network (NN) models were developed using hourly TEC data (for 8 years: 2000–2007) derived from GPS observations over a receiver station located at Sutherland (SUTH) (32.38° S, 20.81° E), South Africa. Verification of the NN models for all algorithms considered was performed on both "seen" and "unseen" data. Hourly TEC values over SUTH for 2003 formed the "seen" dataset. The "unseen" dataset consisted of hourly TEC data for 2002 and 2008 over Cape Town (CPTN) (33.95° S, 18.47° E) and SUTH, respectively. The models' verification showed that all algorithms investigated provide comparable results statistically, but differ significantly in terms of time required to achieve convergence during input-output data training/learning. This paper therefore provides a guide to neural network users for choosing appropriate algorithms based on the availability of computation capabilities used for research.

Sign in / Sign up

Export Citation Format

Share Document