scholarly journals Pallet Recognition with Multi-Task Learning for Automated Guided Vehicles

2021 ◽  
Vol 11 (24) ◽  
pp. 11808
Author(s):  
Chunghyup Mok ◽  
Insung Baek ◽  
Yoonsang Cho ◽  
Younghoon Kim ◽  
Seoungbum Kim
Keyword(s):  
Deep Learning ◽  
Control System ◽  
Control Systems ◽  
Travel Time ◽  
Manufacturing Systems ◽  
Automated Guided Vehicles ◽  
Vision Sensor ◽  
Single Task ◽  
Laser Sensors ◽  
Proposed Model

As the need for efficient warehouse logistics has increased in manufacturing systems, the use of automated guided vehicles (AGVs) has also increased to reduce travel time. The AGVs are controlled by a system using laser sensors or floor-embedded wires to transport pallets and their loads. Because such control systems have only predefined palletizing strategies, AGVs may fail to engage incorrectly positioned pallets. In this study, we consider a vision sensor-based method to address this shortcoming by recognizing a pallet’s position. We propose a multi-task deep learning architecture that simultaneously predicts distances and rotation based on images obtained from a visionary sensor. These predictions complement each other in learning, allowing a multi-task model to learn and execute tasks impossible with single-task models. The proposed model can accurately predict the rotation and displacement of the pallets to derive information necessary for the control system. This information can be used to optimize a palletizing strategy. The superiority of the proposed model was verified by an experiment on images of stored pallets that were collected from a visionary sensor attached to an AGV.

‘New Lamps for Old!’ A Method to Rejuvenate Old Robots through the Use of a Simple Set of Microcontroller-Based Control Boards to Replace Outdated Control Systems

2005 ◽  
Vol 33 (4) ◽  
pp. 339-348 ◽  
Author(s):  
P. Brunn ◽  
A. W. Labib
Keyword(s):  
Control System ◽  
Control Systems ◽  
Manufacturing Systems ◽  
Low Cost ◽  
Welding Robot ◽  
Design Development ◽  
Simple Set

The paper describes the design, development, testing and use of a microcontroller- and PC-based control system which was used to repair and enhance an ASEA IRB6 welding robot in the authors' laboratory. The principles described could be applied to any robot of similar age and to provide a low-cost route to revitalise any working robot hardware that is limited by an outdated control system. The proposed approach addresses a problem within many manufacturing systems operating in industry.

scholarly journals HEADING CONTROL OF AUTOMATED GUIDED VEHICLES

2021 ◽  
Vol 6 (3) ◽  
Author(s):  
Abhishek Bhardwaj ◽  
Kshitij Anand ◽  
Murtaza Khasamwala
Keyword(s):  
Fuzzy Logic ◽  
Control System ◽  
Control Systems ◽  
Pid Control ◽  
Control Strategies ◽  
Optimal Path ◽  
Automated Guided Vehicles ◽  
Holistic View ◽  
Heading Control ◽  
Automation Control

With rapid developments in industry wide automation control systems, there have been numerous proposed design and control strategies as alternatives to conventional methods. This term paper deals with the same in the context of Automated Guided Vehicles (AGV’s) with a review of the current literature for proposed design improvements as well as trajectory control systems that can improve productivity and reduce conflicts compared to current methods. The results of these proposed methods are also compared in contrast to each other to obtain holistic view of the merits and demerits of the same. It is found that a fuzzy logic based PID control is a definite upgrade over conventional PID control systems and that the application of LIDAR and Camera based technologies can help enable free roaming AGV’s lending to increased robustness and flexibility. For a control system consisting of a single AGV and multiple processing stations, a genetic algorithm enhanced centralized fuzzy logic control system is the most optimal path planning algorithm whereas in the cases of multiple robots, a dynamic priority allocated free roaming AGVsystem with semi-decentralized control is found to be much more efficient at simultaneously solving paths and reducingconflicts with other robots in real time compared to the purely centralized systems.

scholarly journals Flexible Productive Systems Modeling and Control Tools: Literature Review

2020 ◽  
pp. 91-98
Author(s):  
URC Mazzoni ◽  
OL Asato ◽  
FY Nakamoto
Keyword(s):  
Control System ◽  
Control Systems ◽  
Manufacturing Systems ◽  
New Technologies ◽  
System Modeling ◽  
Production Costs ◽  
Modeling Tools ◽  
Modeling And Control ◽  
Productive Systems ◽  
And Control

The challenges imposed on Manufacturing Systems (MS), given the demands of a dynamic and competitive market, instigates the development of new technologies to promote the reduction of production costs, increase productivity and ensure the level of quality established by the company. Such technologies applied in MS create demands for new paradigms for the design of control systems, mainly about the integration of automated systems, such as multifunction machines, flexible machining centers, intelligent robotic conveyor systems, and the integration of information systems, production planning and management, and manufacturing execution. The main purpose of control system modeling is to represent a real system using conceptual models to visualize, predict and simulate the desired dynamic behavior of the system. This article presents some modeling tools for control systems capable of adequately representing a manufacturing system with all its requirements and intrinsic characteristics, supported by formal methods for structured modeling of the control system.

Using Heuristic Searching Algorithm for Optimizing Intelligent Transportation Controls

2014 ◽  
Vol 602-605 ◽  
pp. 937-941
Author(s):  
Hua Tao ◽  
Li Wei Zhu
Keyword(s):  
Optimal Control ◽  
Control System ◽  
Control Systems ◽  
Travel Time ◽  
Traffic Safety ◽  
Optimal Solution ◽  
Speed Control ◽  
Intelligent Transportation ◽  
Evolution Process ◽  
Novel Technique

Previously, intelligent transportation control systems have been widely used to improve traffic safety on roadways. Among them, dynamic speed control is a novel technique that has been increasingly used for reducing collision risks. However, previous studies did not consider using any heuristic searching algorithm to obtain the optimal solution to improve traffic safety. The objective of this study is to evaluate the application of using heuristic searching algorithm to optimize the safety effects of dynamic speed control system. The results showed that the evolution process in the heuristic searching algorithm can promote the strategy towards better solutions for the optimization of the safety effects of dynamic speed control. The optimal control reduces collision risks by 12.7% and increased travel time only by 1.4%.

scholarly journals Detection and Classification of Different Weapon Types Using Deep Learning

2021 ◽  
Vol 11 (16) ◽  
pp. 7535
Author(s):  
Volkan Kaya ◽  
Servet Tuncer ◽  
Ahmet Baran
Keyword(s):  
Deep Learning ◽  
Control Systems ◽  
Training Process ◽  
Security Forces ◽  
New Approach ◽  
New Model ◽  
Automatic Control Systems ◽  
Proposed Model ◽  
Computer Environment

Today, with the increasing number of criminal activities, automatic control systems are becoming the primary need for security forces. In this study, a new model is proposed to detect seven different weapon types using the deep learning method. This model offers a new approach to weapon classification based on the VGGNet architecture. The model is taught how to recognize assault rifles, bazookas, grenades, hunting rifles, knives, pistols, and revolvers. The proposed model is developed using the Keras library on the TensorFlow base. A new model is used to determine the method required to train, create layers, implement the training process, save training in the computer environment, determine the success rate of the training, and test the trained model. In order to train the model network proposed in this study, a new dataset consisting of seven different weapon types is constructed. Using this dataset, the proposed model is compared with the VGG-16, ResNet-50, and ResNet-101 models to determine which provides the best classification results. As a result of the comparison, the proposed model’s success accuracy of 98.40% is shown to be higher than the VGG-16 model with 89.75% success accuracy, the ResNet-50 model with 93.70% success accuracy, and the ResNet-101 model with 83.33% success accuracy.

Improving Contouring Accuracy by Using Gain Matching Method

2016 ◽  
Vol 693 ◽  
pp. 1658-1666
Author(s):  
Wen Xia Cao ◽  
Xiao Qing Tian ◽  
Yong Hua Wang
Keyword(s):  
Control System ◽  
Control Systems ◽  
Motion Control ◽  
Control Loop ◽  
Circular Trajectory ◽  
Matching Method ◽  
Matching Model ◽  
Contouring Error ◽  
Proposed Model ◽  
Contouring Accuracy

Motion axes should coordinate exactly in order to get high contouring accuracy. Gain matching is one of the coordinating problems. In this paper, gain matching effects on contouring accuracy are analyzed for both full-closed and half-closed control loop structures. A simple gain matching model is proposed in order to reduce the contouring error caused by unmatched gains of a multi-axis motion control system. According to this model, the contouring error can be reduced by tuning the proportional gain in the position loop of each axis. Finally, experiments are made to verify the proposed model, and the results show that the gain matching model can effectively reduce the contouring error of both linear and circular trajectory of motion control systems.

scholarly journals Spatiotemporal Features—Extracted Travel Time Prediction Leveraging Deep-Learning-Enabled Graph Convolutional Neural Network Model

2021 ◽  
Vol 13 (3) ◽  
pp. 1253
Author(s):  
Xiantong Li ◽  
Hua Wang ◽  
Pengcheng Sun ◽  
Hongquan Zu
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Travel Time ◽  
Coarse Grained ◽  
Travel Time Prediction ◽  
Spatiotemporal Features ◽  
Traffic Conditions ◽  
Time Prediction ◽  
Proposed Model

Travel time prediction is one of the most important parameters to forecast network-wide traffic conditions. Travelers can access traffic roadway networks and arrive in their destinations at the lowest costs guided by accurate travel time estimation on alternative routes. In this study, we propose a long short-term memory (LSTM)-based deep learning model, deep learning on spatiotemporal features with Convolution Neural Network (DLSF-CNN), to extract the spatial–temporal correlation of travel time on different routes to accurately predict route travel time. Specifically, this model utilizes network-wide travel time, considering its topological structure as inputs, and combines convolutional neural network and LSTM techniques to accurately predict travel time. In addition to their spatial dependence, both coarse-grained and fine-grained temporal dependences are fully considered among the road segments along a route as well. The shift problem is formulated in the coarse-grained granularity to predict the route travel time in the next time interval. The experimental tests were conducted using real route travel time obtained by taxi trajectories in Harbin. The test results show that the travel time prediction accuracy of DLSF-CNN is above 90%. Meanwhile, the proposed model outperformed the other machine learning models based on multiple evaluation criteria. The RMSE (Root Mean Squard Error) and R2 (R Squared) increased by 18.6% and 22.46%, respectively. The results indicate the proposed model performs reasonably well under prevailing traffic conditions.

Pengaruh Sistem Pengendalian Internal Pemerintah dan Kompetensi Pegawai Terhadap Kinerja Instansi Pemerintah Kabupaten Karimun

2020 ◽  
Vol 5 (1) ◽  
pp. 101-106
Author(s):  
Doni Putra Utama
Keyword(s):  
Control System ◽  
Control Systems ◽  
Internal Control ◽  
Statistical Tests ◽  
Government Agencies ◽  
Government Performance ◽  
Significant Positive Effect ◽  
Internal Control System ◽  
Employee Competencies ◽  
Positive Effect

This research is a causality study with the title "Effect of Government Internal Control Systems and Employee Competence on the Performance of Government Agencies in Karimun Regency." The purpose of this study was to determine the effect of the implementation of the Government's Internal Control System on the performance of Karimun Regency government agencies and to determine the effect of employee competence on the performance of Karimun Regency government agencies. Data collection using a questionnaire where the questionnaire contained questions about the Government's Internal Control System, employee competencies and agency performance. Data were tested using multiple linear regression statistical tests. Based on the results of the study, it can be concluded that the governmental internal control system has a significant positive effect on Government Agency Performance with the results of statistical tests that show a sig value of 0.016 <0.05 (alpha 5%). Employee Competency has a significant positive effect on Government Performance with the results of statistical tests showing a sig value of 0,000.

Human Activity Recognition using Fourier Transform Inspired Deep Learning Combination Model

2019 ◽  
Vol 9 (1) ◽  
pp. 16-31
Author(s):  
Kyungkoo Jun
Keyword(s):  
Fourier Transform ◽  
Deep Learning ◽  
Short Term Memory ◽  
Window Size ◽  
Sensor Data ◽  
Data Sets ◽  
Data Set ◽  
Proposed Model ◽  
Testing Data ◽  
Labeling Scheme

Background & Objective: This paper proposes a Fourier transform inspired method to classify human activities from time series sensor data. Methods: Our method begins by decomposing 1D input signal into 2D patterns, which is motivated by the Fourier conversion. The decomposition is helped by Long Short-Term Memory (LSTM) which captures the temporal dependency from the signal and then produces encoded sequences. The sequences, once arranged into the 2D array, can represent the fingerprints of the signals. The benefit of such transformation is that we can exploit the recent advances of the deep learning models for the image classification such as Convolutional Neural Network (CNN). Results: The proposed model, as a result, is the combination of LSTM and CNN. We evaluate the model over two data sets. For the first data set, which is more standardized than the other, our model outperforms previous works or at least equal. In the case of the second data set, we devise the schemes to generate training and testing data by changing the parameters of the window size, the sliding size, and the labeling scheme. Conclusion: The evaluation results show that the accuracy is over 95% for some cases. We also analyze the effect of the parameters on the performance.

Sign in / Sign up

Export Citation Format

Share Document