A New Hybrid Learning Control System for Robots Based on Spiking Neural Networks DOI
Vahid Azimirad,

S. Yaser Khodkam,

Amir Bolouri

et al.

Published: Jan. 1, 2023

This paper presents a new hybrid learning and control method that can tune their parameters based on reinforcement learning. In the proposed method, nonlinear controllers are considered multi-input multi-output functions then replaced with SNNs algorithms. Dopamine-modulated spike-timing-dependent plasticity (STDP) is used for manipulating synaptic weights between input output of neuronal groups (for parameter adjustment). Details presented some case studies done such as Fractional Order PID (FOPID) Feedback Linearization. The structure dynamic equations presented, algorithm tested robots results compared other works. addition, to verify applicability it another controller named Linearization, which provides acceptable results. Results show has better performance in terms Integral Absolute Error (IAE) highly useful hardware implementation due its low energy consumption, high speed, accuracy. be generalized systems like robots.

Language: Английский

A Smart Factory Architecture Based on Industry 4.0 Technologies: Open-Source Software Implementation DOI Creative Commons
Jesús Anselmo Fortoul-Diaz, Luis Antonio Carrillo-Martinez, Adolfo Centeno-Téllez

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 101727 - 101749

Published: Jan. 1, 2023

The Smart Factory has been a concept studied during the last decade that not standardized yet; for this reason, academy and industry have developed wide variety of new architectures describe integration elements digitization interconnection. present research aims to introduce architecture proposal migrating traditional (automation) smart (digitization) factories, implemented through open-source software. proposed is integrated, first time, by interconnection six main elements: cyber-physical systems, edge computing, artificial intelligence, cloud data analytics, cybersecurity; describes in detail their definitions, sub-elements, between elements, minimum requirements implementation. test factory was done scale pilot testing pick place process, where assembly wood pieces from geometric Tangram's puzzle required; includes six-degree-of-freedom robot arm, conveyor, vision system, storage area. case study conducted allowed four puzzles (fish, house, rocket, swan) were assembled with different batches pieces. implementation flexibility adaptability. final reports included status assembly, number assembled, stored, sequence, time. Similarly, development SCADA system asset control as well monitoring. KPIs process measured productivity (OTD) time tracking (ATCT TA) 16 tests, founding manufacturing cell fully integrated repeatability; SF represents an alternative small medium automated factories achieve digitization, it ready be tested more complex scenario.

Language: Английский

Citations

26
IoT Real-Time Production Monitoring and Automated Process Transformation in Smart Manufacturing DOI Open Access
Xiangqian Wang, Haifeng Hu, Yuyao Wang

et al.

Journal of Organizational and End User Computing, Journal Year: 2024, Volume and Issue: 36(1), P. 1 - 25

Published: Jan. 17, 2024

Conventional automobile manufacturing plants involve intricate assembly, testing, and debugging processes heavily reliant on manual operations. This study aims to explore the application of industrial internet things (IIoT) deep learning algorithms achieve process automation in manufacturing. Firstly, utilizing IIoT technology, OPC UA, point cloud fitting techniques, a comprehensive modeling most equipment materials within factory is conducted, constructing digital twin (DT) model as virtual representation actual equipment. Subsequently, innovatively introduces Q network algorithm, facilitating automatic transition production improving efficiency. Through comparison with ten baseline models, proposed demonstrates an improvement efficiency at least four percentage points compared other models. Experimental validation confirms effectiveness smart for electric vehicle

Language: Английский

Citations

6
A Brief Review on Flexible Electronics for IoT: Solutions for Sustainability and New Perspectives for Designers DOI Creative Commons
Graziella Scandurra,

Antonella Arena,

C. Ciofi

et al.

Sensors, Journal Year: 2023, Volume and Issue: 23(11), P. 5264 - 5264

Published: June 1, 2023

The Internet of Things (IoT) is gaining more and popularity it establishing itself in all areas, from industry to everyday life. Given its pervasiveness considering the problems that afflict today's world, must be carefully monitored addressed guarantee a future for new generations, sustainability technological solutions focal point activities researchers field. Many these are based on flexible, printed or wearable electronics. choice materials therefore becomes fundamental, just as crucial provide necessary power supply green way. In this paper we want analyze state art flexible electronics IoT, paying particular attention issue sustainability. Furthermore, considerations will made how skills required designers such circuits, features design tools characterization electronic circuits changing.

Language: Английский

Citations

14
From Smart Devices to Smarter Systems: The Evolution of Artificial Intelligence of Things (AIoT) with Characteristics, Architecture, Use Cases and Challenges DOI
Veena Parihar,

Ayasha Malik,

Bhawna

et al.

Springer eBooks, Journal Year: 2023, Volume and Issue: unknown, P. 1 - 28

Published: Jan. 1, 2023

Language: Английский

Citations

12
Introducing an improved rime algorithm combined with gate current unit as an innovative stability monitoring and controlling model for flexible biogas-to-hydrogen/methanol system DOI
Tao Tan,

Zetao Huang,

Zuhao Li

et al.

Renewable Energy, Journal Year: 2025, Volume and Issue: unknown, P. 123032 - 123032

Published: April 1, 2025

Language: Английский

Citations

0
A Review: Image Processing Techniques’ Roles towards Energy-Efficient and Secure IoT DOI Creative Commons
Abbas M. Al-Ghaili, Hairoladenan Kasim,

Zainuddin Hassan

et al.

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(4), P. 2098 - 2098

Published: Feb. 6, 2023

The goal of this review paper is to highlight the image processing techniques’ role in Internet Things (IoT), aiming attain an energy-efficient and secure IoT. IoT-dependent systems (IoTSs) cause heavy usage energy. This one biggest issues associated with IoTSs. Another issue that security digital content a big challenge difficulty. Image has recently played essential resolving these difficulties. Several researchers have made efforts improve future IoTSs, which are summarized article. Day-by-day, proposed methods developed, thus IoT deployment been plainly engaged our everyday activities. efficient image-processing techniques can be utilized by IoTSs overcome such proposed. aims those make contributions direction. Thus, study numerous research studies on subject. looks at 36 publications relevant several types innovative work provide readers map suitable used certain (i.e., scenarios). Both methodology analysis come out suggested mind highlighting number solutions techniques) help design energy-efficient, secure, intelligent system. We some conclusions projections for work.

Language: Английский

Citations

8
A new hybrid learning control system for robots based on spiking neural networks DOI Creative Commons
Vahid Azimirad,

S. Yaser Khodkam,

Amir Bolouri

et al.

Neural Networks, Journal Year: 2024, Volume and Issue: 180, P. 106656 - 106656

Published: Aug. 23, 2024

This paper presents a new hybrid learning and control method that can tune their parameters based on reinforcement learning. In the proposed method, nonlinear controllers are considered multi-input multi-output functions then replaced with SNNs algorithms. Dopamine-modulated spike-timing-dependent plasticity (STDP) is used for manipulating synaptic weights between input output of neuronal groups (for parameter adjustment). Details presented some case studies done such as Fractional Order PID (FOPID) Feedback Linearization. The structure dynamic equations presented, algorithm tested robots results compared other works. Moreover, to demonstrate effectiveness SNNFOPID, we conducted rigorous testing variety systems including two-wheel mobile robot, double inverted pendulum, four-link manipulator robot. revealed impressively low errors 0.01 m, 0.03 rad, rad each system, respectively. another controller named Linearization, which provides acceptable results. Results show has better performance in terms Integral Absolute Error (IAE) highly useful hardware implementation due its energy consumption, high speed, accuracy. duration necessary achieving full stable proficiency various robotic using SNNFOPD, SNNFL an Asus Core i5 system within Simulink's Simscape environment follows: - Two-link robot SNNFOPID: 19.85656 hours SNNFL: 0.45828 Double pendulum 3.455 Mobile 3.71948 Four-link 16.6789 hours. be generalized like robots.

Language: Английский

Citations

1
Optimization and design of electromechanical control automation based on dual motor control algorithm DOI Creative Commons
Lu Wei

Frontiers in Mechanical Engineering, Journal Year: 2024, Volume and Issue: 10

Published: Dec. 6, 2024

Introduction In response to the high demand for dynamic characteristics and control in current electromechanical automatic systems. Methods This study first analyzes dual motor system. A novel automation model based on a algorithm is proposed through strategy of backlash elimination digital proportional integral derivative algorithm. Results Discussion The results indicated that optimization had promoting effect performance Compared with other popular models same type, research method was best. During no-load start-up phase, maximum tracking error synchronization speed new showed significant decreasing trend, between two motors being only 0.02%. Under steady-state sudden load, could reach stable state within 3 s, errors ±5%. Conclusion summary, combining can provide theoretical basis practical guidance designing implementing efficient

Language: Английский

Citations

0
Improved Path Tracking Control in Mobile Robots Using a Hybrid FOPID Controller with Backstepping Technique: An Experimental Study DOI Open Access

Rafik Euldji,

Redha Rebhi, Mohammed Ayad Alkhafaji

et al.

Journal Européen des Systèmes Automatisés, Journal Year: 2023, Volume and Issue: 56(2), P. 173 - 186

Published: April 30, 2023

This study aims to address the challenge of low-cost hardware implementation a combined backstepping with fractional order PID (FOPID) controller for mobile robots in real-time applications.Moreover, this work proposes self-designed robot prototype that is easy realize, low cost, spares time, and reduces human effort.This platform was equipped two DC motors quadratic encoders passive wheels, controlled by an Arduino mega, where software code developed Matlab-Simulink environment, using Simulink support package Arduino.Four case studies were conducted demonstrate effectiveness suggested methodology.Experimental results improved trajectory tracking performance less error smooth control efforts, capable handling trajectories continuous non-continuous gradients.

Language: Английский

Citations

1
A New Hybrid Learning Control System for Robots Based on Spiking Neural Networks DOI
Vahid Azimirad,

S. Yaser Khodkam,

Amir Bolouri

et al.

Published: Jan. 1, 2023

This paper presents a new hybrid learning and control method that can tune their parameters based on reinforcement learning. In the proposed method, nonlinear controllers are considered multi-input multi-output functions then replaced with SNNs algorithms. Dopamine-modulated spike-timing-dependent plasticity (STDP) is used for manipulating synaptic weights between input output of neuronal groups (for parameter adjustment). Details presented some case studies done such as Fractional Order PID (FOPID) Feedback Linearization. The structure dynamic equations presented, algorithm tested robots results compared other works. addition, to verify applicability it another controller named Linearization, which provides acceptable results. Results show has better performance in terms Integral Absolute Error (IAE) highly useful hardware implementation due its low energy consumption, high speed, accuracy. be generalized systems like robots.

Language: Английский

Citations

0