检索结果-内蒙古大学图书馆

Experimental Study on Reducing the Oscillations of a Cable-Suspended Parallel robot for Video Capturing Purposes using Simulated Annealing and Path Planning

学校读者我要写书评

暂无评论

Experimental Study on Reducing the Oscillations of a Cable-S...

International Conference of Signal Processing and Intelligent Systems (ICSPIS)

作者： Mohammad Ghanatian Mohammad Reza Hairi Yazdi Mehdi Tale Masouleh Human and Robot Interaction Laboratory School of Mechanical Engineering University of Tehran Tehran Iran School of Mechanical Engineering University of Tehran Tehran Iran Human and Robot Interaction Laboratory School of Electrical and Computer Engineering University of Tehran Tehran Iran

ISBN: (纸本)9781665409391

This paper aims to smoothen the movements of an under-constrained cable-suspended parallel robot which carries a camera for video capturing purposes, especially for video capturing of football games. This goal is achieved by means of an accurate while simple PID controller optimized by the Simulated Annealing algorithm and implemented on the joint space. Moreover, a planning strategy is considered for the joint space trajectory of the robot which guarantees the zero jerk, acceleration, and velocity at the start and the end of motions. On this regard, a septic function for each joint of the robot is considered and the corresponding boundary conditions are applied to the function to make the end-effector movements less oscillatory. This method is implemented and tested on an experimental setup while the end-effector oscillation data is recorded using an IMU sensor attached to the end-effector of the robot. Applying frequency analysis on the oscillatory data of the end-effector reveals that this simple method, on average, resulted in a 33.8% reduction in the average amplitude of the end-effector oscillations. Moreover, the maximum joint space error was decreased by 76.68% when using septic joint profile compared to the ordinary linear Cartesian trajectory planning approach. Upon applying the proposed strategy, the error of the controller has been reduced by 92.26% with respect to the previous research performed on this experimental setup. Without requiring any knowledge on the dynamic model of the robot or the natural frequencies of the end-effector or using any complex controller, this method significantly increased the smoothness and accuracy of the robot movements. The proposed method can be regarded as a definitive asset when this robot is used for video capturing purposes.

关键词： Vibrations Parallel robots Trajectory planning Simulated annealing Games Aerospace electronics End effectors

robots Learn Visual Pouring Task Using Deep Reinforcement Learning with Minimal human Effort

学校读者我要写书评

暂无评论

Robots Learn Visual Pouring Task Using Deep Reinforcement Le...

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Homayoun Moradi Mehdi Tale Masouleh Behzad Moshiri Human and Robot Interaction Laboratory School of Electrical and Computer Engineering University of Tehran Tehran Iran School of Electrical and Computer Engineering College of Engineering University of Tehran Tehran Iran

ISBN: (纸本)9781665420952

In this paper, the entire pipeline for teaching a robot to perform the visual pouring task is addressed with a minimal human supervision. In this task the robot’s arm that carrying a glass, and the target container are initialized in an arbitrary position. The reinforcement learning approach is employed without the use of reward engineering where the reward function is approximated by a Convolutional Neural Network. This network, which is trained on a little set of goal images (positive samples), takes a raw RGB image as input. In data gathering step, only goal conditions are established by an expert, and positive samples are collected in a self-supervised manner. The major contribution of this paper consists in using only the RGB image, without 3D analysis which is not possible by classical approaches, and self-supervised positive sample gathering which has not been addressed in recent RL studies. Moreover, the deep metric learning approach is utilized to train the reward network in order to obtain increased robustness against unseen negative samples. The foregoing is applied since this approach classifies the negative samples as ignore class and prevents the network from being fooled by the RL algorithm. Soft Actor-Critic is the RL algorithm employed in this study. A fixed monocular camera is the single sensor used for perception and RGB image is used as the representation of the state for action selection, value estimation, and input of reward network. The action-space is defined such a way that reduces convergence time, and in this approach, the policy network generates the desired 3D position in each step. This method of defining the action-space resulted in a reduction in training steps from 100k-1M to 1k. The provided method is compared to joint-based action-space and sparse-reward to validate the good effect of this combination of action-space and reward , i.e., position-based action-space, and continuous reward. Some experiments are designed as case studies

关键词： Training Visualization Three-dimensional displays Mechatronics Pipelines Reinforcement learning robot sensing systems

A deep learning based automated hand hygiene training system

学校读者我要写书评

暂无评论

arXiv

arXiv 2021年

作者： Shahbandeh, Mobina Ghaffarpour, Fatemeh Vali, Sina Haghpanah, Mohammad Amin Torkamani, Amin Mousavi Masouleh, Mehdi Tale Kalhor, Ahmad School of Electrical and Computer Engineering University of Tehran Tehran Iran Human and Robot Interaction Laboratory University of Tehran Tehran Iran Department of Computational Engineering Lappeenranta University of Technology Lappeenranta Finland School of Electrical and Computer Engineering Human and Robot Interaction Laboratory University of Tehran Tehran Iran

Hand hygiene is crucial for preventing viruses and infections. Due to the pervasive outbreak of COVID-19, wearing a mask and hand hygiene appear to be the most effective ways for the public to curb the spread of these viruses. The World Health Organization (WHO) recommends a guideline for alcohol-based hand rub in eight steps to ensure that all surfaces of hands are entirely clean. As these steps involve complex gestures, human assessment of them lacks enough accuracy. However, Deep Neural Network (DNN) and machine vision have made it possible to accurately evaluate hand rubbing quality for the purposes of training and feedback. In this paper, an automated deep learning based hand rub assessment system with real-time feedback is presented. The system evaluates the compliance with the 8-step guideline using a DNN architecture trained on a dataset of videos collected from volunteers with various skin tones and hand characteristics following the hand rubbing guideline. Various DNN architectures were tested, and an Inception-ResNet model led to the best results with 97% test accuracy. In the proposed system, an NVIDIA Jetson AGX Xavier embedded board runs the software. The efficacy of the system is evaluated in a concrete situation of being used by various users, and challenging steps are identified. In this experiment, the average time taken by the hand rubbing steps among volunteers is 27.2 seconds, which conforms to the WHO guidelines. Copyright © 2021, The Authors. All rights reserved.

关键词： Deep neural networks

All-Printed 3d Solid-State Rechargeable Zinc-Air Micro-Batteries

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Liu, Gengsheng Ma, Zhaolei Li, Guoxian Yu, Wei Wang, Peng Meng, Chuizhou Guo, Shijie State Key Laboratory for Reliability and Intelligence of Electrical Equipment Hebei Key Laboratory of Smart Sensing and Human-Robot Interaction School of Mechanical Engineering Hebei University of Technology Tianjin300401 China

Lightweight, compact, integrated and miniaturized energy devices are under high pursuit for portable and wearable electronics. However, improving the energy density per area still remains a long-standing challenge. Herein, we report the design and fabrication of a solid-state zinc-air micro-battery (ZAmB) by a facile 3D direct printing technique. The interdigital electrodes, gel electrolyte and encapsulation frame are all printed with customized design, with composition of the printing inks optimized to obtain the best battery performance. Multiple layers of interdigital electrodes are sequentially printed with a fine overlapping to achieve an ultrahigh thickness of 2.5 mm for a remarkably increased specific areal energy of up to 45.6 mWh cm−2. To meet different powering requirement for output voltage and current, battery modules consisting of individual ZAmBs connected in series or parallel or combination of the two are printed with a facile integration to external loads. Powering of LEDs, digital watch, mini rotary motor and even charging of a smartphone by the printed ZAmB modules are successfully demonstrated. The versatile 3D direct printing technique enables the fabricated ZAmBs with adjustable form factor and integration capability with other electronics, paving the way of exploring new energy systems with diverse structures and extended functionalities. © 2022, The Authors. All rights reserved.

关键词： 3D printers

Ultrathin and Highly-Stable Rubber Electrodes Based on Island-Bridge Multi-Filler Conductive Network for Multilayer-Stacked Dielectric Elastomer Artificial Muscles

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Yu, Wei Chen, Wenqiang Yuan, Wenjing Li, Guoxian Meng, Chuizhou Guo, Shijie State Key Laboratory for Reliability and Intelligence of Electrical Equipment Engineering Research Center of Ministry of Education for Intelligent Rehabilitation Device and Detection Technology Hebei Key Laboratory of Smart Sensing and Human-Robot Interaction School of Mechanical Engineering Hebei University of Technology Tianjin300401 China School of Materials Science & Engineering Tianjin Key Laboratory of Materials Laminating Fabrication and Interface Control Technology Hebei University of Technology Tianjin300130 China

Dielectric elastomer actuators (DEAs) have shown great promise in applications ranging from millimeter-sized haptic feedback to centimeter-scaled soft robots. However, past progress has been focused on the soft dielectric elastomers, and little attention has been paid to the compliant electrodes. Being stable to maintain conductive while enduring repeated large deformations and compliant to avoid stiffness increasing for the dielectric elastomer remains challenging. Moreover, the fabrication of multilayer DEAs is complex, hindering the practicality. Herein, an ultrathin (~500 nm) and highly stable acrylate rubber electrode based on unique island-bridge multi-filler conductive network is developed with outstanding electrical conductivity of 2.7 S cm-1, excellent stretchability of over 1200%, and superior cycling stability of 500,000 cycles at 30% uniaxial strain. An efficient stacking method based on conductive and dielectric layer co-stretching (CDCS) approach is proposed for assembling multilayer-stacked DEAs. The 5-layer DEA reaches a large actuation strain of >17% for lifting a heavy 1-kg load and can serve as biceps muscle to drive an artificial lower arm by a maximum rotation angle of 8°, demonstrating its potential in actuating humanoid robot in an essentially soft way. This conductive rubber electrode and CDCS method are promising for developing soft robotics and muscle-like actuators. © 2024, The Authors. All rights reserved.

关键词： Electrodes

A Highly Sensitive Stretchable Constraint Fiber Mat Sensor with Tailored Near-Zero Poisson's Ratio for Very Early Stage of Battery Over Charge/Overdischarge Monitoring

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Liu, Guoqing Zhang, Yingying Li, Bo Guo, Shijie Yuan, Wenjing Meng, Chuizhou School of Materials Science & Engineering Tianjin Key Laboratory of Materials Laminating Fabrication and Interface Control Technology Hebei University of Technology Tianjin300130 China State Key Laboratory for Reliability and Intelligence of Electrical Equipment Engineering Research Center of Ministry of Education for Intelligent Rehabilitation Device and Detection Technology Hebei Key Laboratory of Smart Sensing and Human-Robot Interaction School of Mechanical Engineering Hebei University of Technology Tianjin300401 China

Battery state monitoring is crucial for safe operation and efficient management of lithium-ion battery (LIB) systems, but early stage of timely and accurate diagnosis remains challenging. Herein, we propose to utilize a highly sensitive stretchable strain sensor to in-situ monitor the slight strain variation of cells during charge and discharge, through which the state of the battery can be accurately reflected. The high sensing capability is realized by tailoring the Poisson’s ratio of low-modulus conductive fiber mats through a facile constraint manner. The non-uniform strain distribution together with the near-zero Poisson’s ratio synergistically result in superior sensing performance, including ultralow strain detection limit (0.006%), high strain sensitivity particularly under small strains (gauge factor of 116.1 between 0–2% strain), and high strain resolution under large deformations. When being practically utilized for LIB monitoring, the constraint fiber mat sensor can produce instant resistance response reflecting the cell capacity, and offer timely warning signal once the cell is overcharged or overdischarged. The developed stretchable strain sensor possesses comprehensive advantages of low modulus, tailorable small size, easy integration and conformality to cells, and little temperature interference, which would serve as a promising candidate for future battery state analysis and safety monitoring. © 2024, The Authors. All rights reserved.

关键词： Lithium-ion batteries

Experimental Study on Neural Network-ARX and ARMAX Actuation Identification of a 3-DoF Delta Parallel robot for Accurate Motion Controller Design

学校读者我要写书评

暂无评论

Experimental Study on Neural Network-ARX and ARMAX Actuation...

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Saeed Rahimi Hasan Jalali Mohammad Reza Hairi Yazdi Ahmad Kalhor Mehdi Tale Masouleh Human and Robot Interaction Laboratory School of Mechanical Engineering University of Tehran Tehran Iran Human and Robot Interaction Laboratory School of Electrical and Computer Engineering University of Tehran Tehran Iran School of Mechanical Engineering College of Engineering University of Tehran Tehran Iran Control and Intelligent Processing Center of Excellence School of Electrical and Computer Engineering University of Tehran Tehran Iran

ISBN: (纸本)9781665420952

This paper investigates the problem of actuator identification of a 3-DoF Delta parallel robot, by means of linear AutoRegressive Moving Average with eXogenous input (ARMAX) and nonlinear dynamic Neural Network AutoRegressive with eXogenous input (NN-ARX) methods. To this end, the ARMAX and NN-ARX approaches are used to develop a scheme which is capable of identifying a model for each actuator. Based on the ARMAX structure, an accurate model of the actuation system is derived. The model is then trained and tested using the data collected from a real robotic setup. Using a dynamic neural network capabilities, an identification and prediction scheme is designed for modeling the nonlinear dynamic behavior of the system. The NN-ARX is trained based on the collected data from the system, and the new trajectory data is used to validate both methods. By considering the results of experimental implementations, three servo motors are demonstrated to have different dynamical behavior which was expected to happen from the outset, due to uncertainty in fabrication of motors component and gearbox. In the identification and prediction stages, the Root Mean Square Errors (RMSE) index is used to validate and analyze the performance of each method using the validation data from new trajectories. In terms of predicting the output of the system, NN-ARX performed better than ARMAX with RMSE of 0.001441, compared to ARMAX with RMSE of 0.0886. Due to the high accuracy of the obtained models. thus they can be used in the design of motion controllers and modelling disturbances in the system.

关键词： Actuators Parallel robots Artificial neural networks Predictive models Mathematical models Trajectory Nonlinear dynamical systems

Effect of the configuration and concentration of B-doping on the thermal properties and structural stability of graphene: A theoretical study

学校读者我要写书评

暂无评论

Diamond and Related Materials 2025年 156卷

作者： Li, Dongbo Da, Liangguo Huang, Kaifeng Li, Dong Yu, Liang Wang, Haini Yu, Yaqi Wu, Long Tang, Yunqing School of Mechanical and Electrical Engineering Huainan Normal University Huainan232038 China Human-computer collaborative robot Joint Laboratory of Anhui Province Huainan232038 China School of Mechanical Engineering Shandong University Jinan250061 China State Key Laboratory of Advanced Equipment and Technology for Metal Forming Shandong University Jinan250061 China Key Laboratory of High Efficiency and Clean Mechanical Manufacture Ministry of Education Shandong University Jinan250061 China

The thermal properties and structural stability of B-doped graphene were investigated by employing density functional theory (DFT), ab initio molecular dynamics (AIMD), and molecular dynamics (MD) simulations, with a focus on the effects of doping configuration and concentration. DFT calculations indicate that B-doping reduces the cohesive energy of graphene, consequently diminishing its structural stability. Among the three B double-doping configurations, only the para configuration exhibits a phonon dispersion without imaginary frequencies, whereas both meta and ortho configurations show imaginary frequencies, indicating structural instability. The AIMD results further demonstrate that B-doped graphene preferentially undergoes cleavage at B[sbnd]B bonds, leading to decreased structural stability. MD results show that B-doping significantly reduces the thermal conductivity of graphene. At a doping concentration is 0.6 %, the thermal conductivity of meta and para configurations decreases to 48.08 % and 54.6 % of that of the pristine graphene, respectively. The results of phonon properties show that B-doping reduces the phonon participation rate and induces strong phonon localization at the B-doped site. These results and conclusions will provide valuable theoretical insights for the thermal design of graphene-based semiconductor devices. © 2025 Elsevier B.V.

关键词： Temperature distribution

Flexible, Stretchable, Breathable and Sweatproof All-Nanofiber Iontronic Tactile Sensor for Continuous and Comfortable Knee Joint Motion Monitoring

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Wang, Peng Liu, Jun Yu, Wei Li, Guoxian Meng, Chuizhou Guo, Shijie State Key Laboratory for Reliability and Intelligence of Electrical Equipment School of Mechanical Engineering Hebei University of Technology Tianjin300401 China Hebei Key Laboratory of Smart Sensing and Human-Robot Interaction School of Mechanical Engineering Hebei University of Technology Tianjin300401 China

Running in a wrong way for years will cause progressive damage to the knee joints, which brings about a lot of pains to exercisers. Wearable sensors for continuously monitoring the knee condition are highly required. However, conventional flexible sensors built on solid substrates are inferior in comfortability during long time wearing. Herein, a highly flexible and stretchable iontronic tactile sensor based on an electrospun thermoplastic polyurethane nanofiber platform is developed. Thanks to the supercapacitive sensing mechanism and porous microstructure, high sensitivity, wide detection range, fast response and well stability and repeatability are achieved. The multilayer interlaced nanofiber network containing numerous micro-pores and micro-channels endows the sensor with air permeability, sweat inertness, thermal conductivity and biocompatibility, which ensures a comfort wearing for a long time. Practical wearable sensing of the knee motion during running in a real-time and continuous manner is demonstrated, including the monitoring of single point pressure by 3 individual sensor units on the rim of the knee and mapping of the spatial pressure distribution by a sensor array on the whole knee region. The health status of the knee can be timely judged, which is useful in warning the knee injury to achieve a correct fitness for exercisers. © 2022, The Authors. All rights reserved.

关键词： Joints (anatomy)