Article
Computer Science, Information Systems
Yong Tao, Haitao Liu, Shuo Chen, Jiangbo Lan, Qi Qi, Wenlei Xiao
Summary: In this paper, an accuracy compensation algorithm for the absolute positioning of industrial robots is proposed based on deep belief networks. A position error mapping model is proposed to realize the absolute positioning accuracy compensation of industrial robots. Experimental results showed that the proposed algorithm significantly improved the absolute positioning accuracy of industrial robots.
Article
Computer Science, Information Systems
Kunming Zheng, Qiuju Zhang, Li Peng, Shuisheng Zeng
Summary: This study proposes an adaptive memetic differential evolution-back propagation-fuzzy neural network (AMDE-BP-FNN) control method for efficient and precise control of robots with complex dynamic characteristics, while reducing control costs.
INFORMATION SCIENCES
(2023)
Article
Engineering, Aerospace
Bo LI, Wei Tian, Chufan Zhang, Fangfang Hua, Guangyu Cui, Yufei LI
Summary: This paper proposes a neural-network-based approach to improve the positioning accuracy of industrial robots. By optimizing a neural network to model and predict positioning errors, the predicted errors are utilized for compensation in the robot's workspace. Experimental results demonstrate a significant reduction in positioning errors and an increase in the robot's accuracy.
CHINESE JOURNAL OF AERONAUTICS
(2022)
Article
Engineering, Electrical & Electronic
Vytautas Bucinskas, Andrius Dzedzickis, Marius Sumanas, Ernestas Sutinys, Sigitas Petkevicius, Jurate Butkiene, Darius Virzonis, Inga Morkvenaite-Vilkonciene
Summary: Positioning accuracy in robotics is crucial for the manufacturing process, and one way to enhance accuracy is through machine learning. This paper introduces an online deep Q-learning method to improve positioning accuracy at key points, with experiments conducted on a KUKA-YouBot robot. Results show that implementing this ML-based compensation method led to a significant decrease in positioning error at critical trajectory points.
Article
Engineering, Mechanical
Congcong Ye, Jixiang Yang, Han Ding
Summary: Industrial robots are useful for machining large complex structural parts due to their flexibility and low cost, but stiffness deformation affects milling accuracy. Traditional stiffness models lack accuracy due to ignored factors. A simulation-driven transfer learning method is proposed for accurate deformation prediction with minimal real data compared to conventional models.
INTERNATIONAL JOURNAL OF MECHANICAL SCIENCES
(2022)
Article
Mathematics
Manu Centeno-Telleria, Ekaitz Zulueta, Unai Fernandez-Gamiz, Daniel Teso-Fz-Betono, Adrian Teso-Fz-Betono
Summary: This paper presents a methodology for tuning optimal parameters in the differential evolution algorithm using an artificial neural network. Experimental results on 24 test problems reveal three distinct cases for optimal parameter tuning, and a comparison with other tuning rules is conducted for performance validation.
Article
Engineering, Electrical & Electronic
Yong Ye, Yuting Liu, Weihan Yin, Jiahao Deng, Xiaofeng Zhu
Summary: This paper proposes a new method for hand distance estimation and positioning using a combination of capacitive sensor array and machine learning. Experimental results show that the RBF neural network algorithm performs best in hand distance estimation, and in the positioning tests, the proposed method outperforms the Y-shaped sensor arrangement method with an accuracy improvement of over 98.8%.
MICROSYSTEM TECHNOLOGIES-MICRO-AND NANOSYSTEMS-INFORMATION STORAGE AND PROCESSING SYSTEMS
(2021)
Article
Engineering, Mechanical
Peng Huang, Hong-Zhong Huang, Yan-Feng Li, He Li
Summary: This study proposes a new method for positioning accuracy reliability analysis based on differential kinematics and saddle-point approximation, which is demonstrated to outperform existing methods in terms of accuracy and efficiency through comparative analysis.
MECHANISM AND MACHINE THEORY
(2021)
Review
Agricultural Engineering
Ayon Tarafdar, Ranjna Sirohi, Vivek Kumar Gaur, Sunil Kumar, Poonam Sharma, Sunita Varjani, Hari Om Pandey, Raveendran Sindhu, Aravind Madhavan, Reshmy Rajasekharan, Sang Jun Sim
Summary: Commercial enzyme production is gaining popularity globally due to its wide applications in traditional and modern industrial sectors. Research efforts focus on improving efficiency, reducing costs, and ensuring sustainability. New engineering interventions and optimization strategies are being introduced to meet the growing demand for industrial enzymes.
BIORESOURCE TECHNOLOGY
(2021)
Article
Engineering, Multidisciplinary
Punit Gupta, Pradeep Singh Rawat, Dinesh Kumar Saini, Ankit Vidyarthi, Meshal Alharbi
Summary: Cloud computing has become essential in recent years and task scheduling plays a critical role in its performance and efficiency. This is particularly important during the pandemic where healthcare services heavily rely on cloud infrastructure. Efficient resource management algorithms are necessary to handle the task loads effectively.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Multidisciplinary Sciences
Yueyang Liu, Likun Hu, Zhihuan Ma
Summary: This paper proposes a new adaptive Differential Evolution algorithm (KBPDE) combined with K-modes clustering, BP neural network, and other strategies to overcome the challenge of planning a globally optimal path and performing dynamic path planning in a complex environment. The experimental results demonstrate that KBPDE can obtain the globally optimal path in a complex environment, and EKBPDE can successfully plan a path in a partially known environment.
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
(2023)
Article
Mathematics, Applied
Changxin Qiu, Aaron Bendickson, Joshua Kalyanapu, Jue Yan
Summary: In this paper, the residual neural network (ResNet) method is investigated for solving ordinary differential equations (ODEs). The accuracy order of the ResNet ODE solver is verified to match the accuracy order of the data. By using Forward Euler, Runge-Kutta2, and Runge-Kutta4 finite difference schemes, three learning data sets are generated and applied to train three independent ResNet ODE solvers. The well-trained ResNet solvers achieve 2nd, 3rd, and 5th orders of one-step errors and exhibit behavior similar to their finite difference method counterparts for linear and nonlinear ODEs with regular solutions. Architecture study, target study, and solution trajectory simulations are conducted to demonstrate the accuracy and capability of the ResNet solver.
JOURNAL OF SCIENTIFIC COMPUTING
(2023)
Article
Computer Science, Information Systems
Anlong Zhang, Zhiyun Lin, Bo Wang, Zhimin Han
Summary: In this study, a nonlinear model predictive control technique using recurrent neural network and differential evolution optimization was proposed for the position control of a single-link flexible-joint robot. The approach demonstrated high efficiency and performance in controlling precision, while effectively suppressing overshoots and residual vibrations as verified through numerical simulations.
Article
Computer Science, Artificial Intelligence
Aiqin Liu, Yuezhong Zhang, Honghua Zhao, Shi Wang, Dianmin Sun
Summary: Attitude detection is crucial for cooperative robots to improve safety, accuracy, and efficiency in their work. It can estimate configuration for robots with unknown parameters, facilitate further kinematics and dynamics analysis. With attitude detection, kinematic parameters of an economical cooperative robot can be corrected.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Huitaek Yun, Hanjun Kim, Young Hun Jeong, Martin B. G. Jun
Summary: Sound and vibration analysis are important tools for machine health diagnosis, and neural networks have potential for finding complex relationships. This paper presents an autoencoder-based anomaly detection framework using an internal sound sensor for industrial robot arms.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)
Article
Engineering, Multidisciplinary
Huining Zhao, Liandong Yu, Haojie Xia, Weishi Li, Yizhou Jiang, Huakun Jia
MEASUREMENT SCIENCE AND TECHNOLOGY
(2018)
Article
Chemistry, Analytical
Liling Han, Huining Zhao, Haojie Xia, Chengliang Pan, Yizhou Jiang, Weishi Li, Liandong Yu
Article
Chemistry, Multidisciplinary
Liling Han, Liandong Yu, Chengliang Pan, Huining Zhao, Yizhou Jiang
APPLIED SCIENCES-BASEL
(2018)
Article
Chemistry, Multidisciplinary
Hua-Kun Jia, Lian-Dong Yu, Hui-Ning Zhao, Yi-Zhou Jiang
APPLIED SCIENCES-BASEL
(2019)
Article
Chemistry, Analytical
Hua-Kun Jia, Lian-Dong Yu, Yi-Zhou Jiang, Hui-Ning Zhao, Jia-Ming Cao
Article
Instruments & Instrumentation
Huining Zhao, Liandong Yu, Jiaming Cao, Huakun Jia, Yizhou Jiang
REVIEW OF SCIENTIFIC INSTRUMENTS
(2020)
Article
Engineering, Electrical & Electronic
Haojie Xia, Lanlin Ni, Songtao Chang, Weishi Li, Jin Zhang, Huining Zhao
Summary: The method improved the Mach-Zehnder interferometer to accurately measure the magnitude and sign of the topological charge of a beam carrying an optical vortex, up to +/- 120.
IEEE PHOTONICS TECHNOLOGY LETTERS
(2022)
Article
Instruments & Instrumentation
Huining Zhao, Ruru Niu, Mengyao Fan, Haojie Xia
Summary: This paper introduces a new method of absolute position measurement based on the hybrid encoding principle for a two-dimensional micro displacement table. The method combines pseudo-random sequences, binary codes, and checkerboards to design a 2D encoding method for precise measurement. A 2D encoding disk is used for measuring absolute position, and the accuracy of the measurement is verified using a modular approach. Experimental results demonstrate that the average position deviation of the measurement system is less than +/- 1 μm.
REVIEW OF SCIENTIFIC INSTRUMENTS
(2023)
Article
Instruments & Instrumentation
Huining Zhao, Wenjing Ding, Mengyao Fan, Haojie Xia, Liandong Yu
Summary: The paper presents an optical method based on auto-collimation to simultaneously measure five degrees of freedom error motions of the rotary axis with high precision. The mathematical model is established by optical ray tracing, and an adjusting mechanism for X-Y micro-displacement is designed to adjust the installation eccentricity error of the high-precision steel ball.
REVIEW OF SCIENTIFIC INSTRUMENTS
(2022)
Article
Engineering, Multidisciplinary
Huining Zhao, Zuo Zhang, Mengyao Fan, Haojie Xia, Liandong Yu
Summary: This paper proposes an inner shape measurement method based on circular structured light for large-scale workpieces. The method extracts the center of the circular laser stripe using a modified Steger method and calibrates the optical plane using a multi-diameter concentric calibration gauge design. The experimental results demonstrate the effectiveness of the proposed method, with a relative error of less than 0.012%.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2023)