NEURAL NETWORK BASED DYNAMIC CONTROLLERS FOR INDUSTRIAL ROBOTS

1995 ◽  
Vol 06 (03) ◽  
pp. 257-271
Author(s):  
SE-YOUNG OH ◽  
WEON-CHANG SHIN ◽  
HYO-GYU KIM
Keyword(s):  
Neural Network ◽  
Real Time ◽  
Network Architecture ◽  
Industrial Robot ◽  
Industrial Robots ◽  
System Stability ◽  
Robot Dynamics ◽  
Time Dynamic ◽  
Dynamic Controller ◽  
And Control

The industrial robot’s dynamic performance is frequently measured by positioning accuracy at high speeds and a good dynamic controller is essential that can accurately compute robot dynamics at a servo rate high enough to ensure system stability. A real-time dynamic controller for an industrial robot is developed here using neural networks. First, an efficient time-selectable hidden layer architecture has been developed based on system dynamics localized in time, which lends itself to real-time learning and control along with enhanced mapping accuracy. Second, the neural network architecture has also been specially tuned to accommodate servo dynamics. This not only facilitates the system design through reduced sensing requirements for the controller but also enhances the control performance over the control architecture neglecting servo dynamics. Experimental results demonstrate the controller’s excellent learning and control performances compared with a conventional controller and thus has good potential for practical use in industrial robots.

Interfacing an industrial robot and MATLAB for predictive visual servoing

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Megha G. Krishnan ◽  
Abhilash T. Vijayan ◽  
Ashok S.
Keyword(s):  
Real Time ◽  
Control Algorithm ◽  
Visual Servoing ◽  
Industrial Robot ◽  
Effective Communication ◽  
Industrial Robots ◽  
Content Type ◽  
Robot Controller ◽  
Predictive Controller ◽  
And Control

Purpose Real-time implementation of sophisticated algorithms on robotic systems demands a rewarding interface between hardware and software components. Individual robot manufacturers have dedicated controllers and languages. However, robot operation would require either the knowledge of additional software or expensive add-on installations for effective communication between the robot controller and the computation software. This paper aims to present a novel method of interfacing the commercial robot controllers with most widely used simulation platform, e.g. MATLAB in real-time with a demonstration of visual predictive controller. Design/methodology/approach A remote personal computer (PC), running MATLAB, is connected with the IRC5 controller of an ABB robotic arm through the File Transfer Protocol (FTP). FTP server on the IRC5 responds to a request from an FTP client (MATLAB) on a remote computer. MATLAB provides the basic platform for programming and control algorithm development. The controlled output is transferred to the robot controller through Ethernet port as files and, thereby, the proposed scheme ensures connection and control of the robot using the control algorithms developed by the researchers without the additional cost of buying add-on packages or mastering vendor-specific programming languages. Findings New control strategies and contrivances can be developed with numerous conditions and constraints in simulation platforms. When the results are to be implemented in real-time systems, the proposed method helps to establish a simple, fast and cost-effective communication with commercial robot controllers for validating the real-time performance of the developed control algorithm. Practical implications The proposed method is used for real-time implementation of visual servo control with predictive controller, for accurate pick-and-place application with different initial conditions. The same strategy has been proven effective in supervisory control using two cameras and artificial neural network-based visual control of robotic manipulators. Originality/value This paper elaborates a real-time example using visual servoing for researchers working with industrial robots, enabling them to understand and explore the possibilities of robot communication.

scholarly journals Data-Driven Stability Assessment of Multilayer Long Short-Term Memory Networks

2021 ◽  
Vol 11 (4) ◽  
pp. 1829
Author(s):  
Davide Grande ◽  
Catherine A. Harris ◽  
Giles Thomas ◽  
Enrico Anderlini
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Short Term Memory ◽  
Moving Average ◽  
Machine Learning Algorithms ◽  
Physical Models ◽  
Data Driven ◽  
The Stability ◽  
And Control ◽  
Nonlinear Autoregressive

Recurrent Neural Networks (RNNs) are increasingly being used for model identification, forecasting and control. When identifying physical models with unknown mathematical knowledge of the system, Nonlinear AutoRegressive models with eXogenous inputs (NARX) or Nonlinear AutoRegressive Moving-Average models with eXogenous inputs (NARMAX) methods are typically used. In the context of data-driven control, machine learning algorithms are proven to have comparable performances to advanced control techniques, but lack the properties of the traditional stability theory. This paper illustrates a method to prove a posteriori the stability of a generic neural network, showing its application to the state-of-the-art RNN architecture. The presented method relies on identifying the poles associated with the network designed starting from the input/output data. Providing a framework to guarantee the stability of any neural network architecture combined with the generalisability properties and applicability to different fields can significantly broaden their use in dynamic systems modelling and control.

scholarly journals Robust Approach to Supervised Deep Neural Network Training for Real-Time Object Classification in Cluttered Indoor Environment

2021 ◽  
Vol 11 (15) ◽  
pp. 7148
Author(s):  
Bedada Endale ◽  
Abera Tullu ◽  
Hayoung Shi ◽  
Beom-Soo Kang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
Network Architecture ◽  
Input Data ◽  
Deep Neural Network ◽  
Data Augmentation ◽  
Object Classification ◽  
Training Data ◽  
Gradient Descent Algorithm

Unmanned aerial vehicles (UAVs) are being widely utilized for various missions: in both civilian and military sectors. Many of these missions demand UAVs to acquire artificial intelligence about the environments they are navigating in. This perception can be realized by training a computing machine to classify objects in the environment. One of the well known machine training approaches is supervised deep learning, which enables a machine to classify objects. However, supervised deep learning comes with huge sacrifice in terms of time and computational resources. Collecting big input data, pre-training processes, such as labeling training data, and the need for a high performance computer for training are some of the challenges that supervised deep learning poses. To address these setbacks, this study proposes mission specific input data augmentation techniques and the design of light-weight deep neural network architecture that is capable of real-time object classification. Semi-direct visual odometry (SVO) data of augmented images are used to train the network for object classification. Ten classes of 10,000 different images in each class were used as input data where 80% were for training the network and the remaining 20% were used for network validation. For the optimization of the designed deep neural network, a sequential gradient descent algorithm was implemented. This algorithm has the advantage of handling redundancy in the data more efficiently than other algorithms.

scholarly journals Real-Time Detection of Dictionary DGA Network Traffic Using Deep Learning

2021 ◽  
Vol 2 (2) ◽  
Author(s):  
Kate Highnam ◽  
Domenic Puzio ◽  
Song Luo ◽  
Nicholas R. Jennings
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
Network Traffic ◽  
Short Term Memory ◽  
Domain Names ◽  
Control Networks ◽  
Detection Techniques ◽  
Lstm Network ◽  
And Control

AbstractBotnets and malware continue to avoid detection by static rule engines when using domain generation algorithms (DGAs) for callouts to unique, dynamically generated web addresses. Common DGA detection techniques fail to reliably detect DGA variants that combine random dictionary words to create domain names that closely mirror legitimate domains. To combat this, we created a novel hybrid neural network, Bilbo the “bagging” model, that analyses domains and scores the likelihood they are generated by such algorithms and therefore are potentially malicious. Bilbo is the first parallel usage of a convolutional neural network (CNN) and a long short-term memory (LSTM) network for DGA detection. Our unique architecture is found to be the most consistent in performance in terms of AUC, $$F_1$$ F 1 score, and accuracy when generalising across different dictionary DGA classification tasks compared to current state-of-the-art deep learning architectures. We validate using reverse-engineered dictionary DGA domains and detail our real-time implementation strategy for scoring real-world network logs within a large enterprise. In 4 h of actual network traffic, the model discovered at least five potential command-and-control networks that commercial vendor tools did not flag.

Robotics Application in Remote Data Acquisition and Control for Solar Ponds

2012 ◽  
Vol 253-255 ◽  
pp. 705-715 ◽  
Author(s):  
Mohamed Elbanhawi ◽  
Milan Simic
Keyword(s):  
Data Acquisition ◽  
Real Time ◽  
Energy Production ◽  
Green Energy ◽  
Industrial Robots ◽  
Sampling Station ◽  
Monitoring And Control ◽  
Solar Ponds ◽  
And Control ◽  
Remote Data

This paper presents one application of industrial robots in the automation of renewable energy production. The robot supports remote performance monitoring and maintenance of salinity gradient solar ponds. The details of the design, setup and the use of the robot sampling station and the remote Data Acquisition (DAQ) system are given here. The use of a robot arm, to position equipment and sensors, provides accurate and reliable real time data needed for autonomous monitoring and control of this type of green energy production. Robot upgrade of solar ponds can be easily integrated with existing systems. Data logged by the proposed system can be remotely accessed, plotted and analysed. Thus the simultaneous and remote monitoring of a large scale network of ponds can be easily implemented. This provides a fully automated solution to the monitoring and control of green energy production operations, which can be used to provide heat and electricity to buildings. Remote real time monitoring will facilitate the setup and operations of several solar ponds around cities.

Finding Features of Positioning Error for Large Industrial Robots Based on Convolutional Neural Network

2021 ◽  
Author(s):  
Daiki Kato ◽  
Kenya Yoshitugu ◽  
Naoki Maeda ◽  
Toshiki Hirogaki ◽  
Eiichi Aoyama ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Production Systems ◽  
Industrial Robot ◽  
Three Dimensional ◽  
Industrial Robots ◽  
Numerical Control ◽  
Robot Motion ◽  
Positioning Error ◽  
End Effector

Abstract Most industrial robots are taught using the teaching playback method; therefore, they are unsuitable for use in variable production systems. Although offline teaching methods have been developed, they have not been practiced because of the low accuracy of the position and posture of the end-effector. Therefore, many studies have attempted to calibrate the position and posture but have not reached a practical level, as such methods consider the joint angle when the robot is stationary rather than the features during robot motion. Currently, it is easy to obtain servo information under numerical control operations owing to the Internet of Things technologies. In this study, we propose a method for obtaining servo information during robot motion and converting it into images to find features using a convolutional neural network (CNN). Herein, a large industrial robot was used. The three-dimensional coordinates of the end-effector were obtained using a laser tracker. The positioning error of the robot was accurately learned by the CNN. We extracted the features of the points where the positioning error was extremely large. By extracting the features of the X-axis positioning error using the CNN, the joint 1 current is a feature. This indicates that the vibration current in joint 1 is a factor in the X-axis positioning error.

Sign in / Sign up

Export Citation Format

Share Document