Due to the complex background of flowers and the similarity between their own categories, the traditional method of image recognition is to extract features manually, which can not solve this problem well. With the development and progress of science and technology, deep learning has gradually entered the image recognition problem and achieved good results. This paper proposes the flower recognition based on transfer learning and Adam deep learning optimization algorithm for the defects of the current mainstream convolutional neural network with deep depth and long parameters, long training time and slow convergence. The VGG16 model is modified and supplemented. At the same time, the transfer learning method and the Adam optimization algorithm are used to accelerate network convergence. Thirty kinds of flower image data sets were established by 102 Category Flower Dataset partial images and 17 Category Flower Dataset. The experimental results show that the accuracy of the test set in this paper is 98.99%. Compared with the traditional image recognition algorithm, it has the characteristics of fast convergence and high recognition accuracy.
{"title":"Flower Recognition Based on Transfer Learning and Adam Deep Learning Optimization Algorithm","authors":"Jing Feng, Zhiwen Wang, Min Zha, Xinliang Cao","doi":"10.1145/3366194.3366301","DOIUrl":"https://doi.org/10.1145/3366194.3366301","url":null,"abstract":"Due to the complex background of flowers and the similarity between their own categories, the traditional method of image recognition is to extract features manually, which can not solve this problem well. With the development and progress of science and technology, deep learning has gradually entered the image recognition problem and achieved good results. This paper proposes the flower recognition based on transfer learning and Adam deep learning optimization algorithm for the defects of the current mainstream convolutional neural network with deep depth and long parameters, long training time and slow convergence. The VGG16 model is modified and supplemented. At the same time, the transfer learning method and the Adam optimization algorithm are used to accelerate network convergence. Thirty kinds of flower image data sets were established by 102 Category Flower Dataset partial images and 17 Category Flower Dataset. The experimental results show that the accuracy of the test set in this paper is 98.99%. Compared with the traditional image recognition algorithm, it has the characteristics of fast convergence and high recognition accuracy.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117137704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the emergence of human-computer interaction, human-machine collaboration and human-computer integration, The robot's perception and motion response of external interactive movement or interactive force has become the key technology for human-computer integration. The identification method of interactive guiding motion the is studied and a robot end sensor with the perception function of external interaction motion is designed. Based on the sensor designed, the recognition principle of the perceptron is analyzed and modeled based on the micro-displacement model. A neural network fusion model between array sensing unit micro-displacement and six-axis response displacement is constructed. In order to ensure the recognition accuracy of the sensing unit, the method of fuzzy information fusion is used to preprocess the sensing unit information. In order to improve and verify the recognition performance of the sensor, the relevant design parameters of the sensor are optimized and analyzed.
{"title":"Research on the identification method of interactive guiding movement by the end sensor of robot","authors":"Haibo Xu, Rui Wang, L. xilinx Wang, Yingkun Qian","doi":"10.1145/3366194.3366289","DOIUrl":"https://doi.org/10.1145/3366194.3366289","url":null,"abstract":"With the emergence of human-computer interaction, human-machine collaboration and human-computer integration, The robot's perception and motion response of external interactive movement or interactive force has become the key technology for human-computer integration. The identification method of interactive guiding motion the is studied and a robot end sensor with the perception function of external interaction motion is designed. Based on the sensor designed, the recognition principle of the perceptron is analyzed and modeled based on the micro-displacement model. A neural network fusion model between array sensing unit micro-displacement and six-axis response displacement is constructed. In order to ensure the recognition accuracy of the sensing unit, the method of fuzzy information fusion is used to preprocess the sensing unit information. In order to improve and verify the recognition performance of the sensor, the relevant design parameters of the sensor are optimized and analyzed.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115624509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Infrared small target detection plays a significant role in modern infrared search and track systems(IRST) and has also received much attention. Based on this background, a novel approach to detect infrared small target based on salient region extraction (SRE) and gradient vector processing (GVP) is proposed in this paper. This approach consists of three parts: pre-processing based on fast median filtering, SRE based on spectral residual algorithm (SRA) processing and similarity analysis algorithm (SAA) processing, and target detection in infrared gradient vector field (IGVF). Both theory and experimental have shown that the proposed algorithm can detect the target more robustly and accurately, and spend less time processing compared with the existing infrared small target detection algorithm. It is a small infrared target detection method with superior performance and good application prospect.
{"title":"Infrared small target detection based on salient region extraction and gradient vector processing","authors":"Junchang Zhang, Bochuan Zhang, Pu Liu","doi":"10.1145/3366194.3366269","DOIUrl":"https://doi.org/10.1145/3366194.3366269","url":null,"abstract":"Infrared small target detection plays a significant role in modern infrared search and track systems(IRST) and has also received much attention. Based on this background, a novel approach to detect infrared small target based on salient region extraction (SRE) and gradient vector processing (GVP) is proposed in this paper. This approach consists of three parts: pre-processing based on fast median filtering, SRE based on spectral residual algorithm (SRA) processing and similarity analysis algorithm (SAA) processing, and target detection in infrared gradient vector field (IGVF). Both theory and experimental have shown that the proposed algorithm can detect the target more robustly and accurately, and spend less time processing compared with the existing infrared small target detection algorithm. It is a small infrared target detection method with superior performance and good application prospect.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127455855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The visual-based localization (VBL) obtains the corresponding pose estimation in the localization system by utilizing various useful information in the surrounding environment, such as images, point cloud models, geometric information, semantic information. In recent years, visual-based localization (VBL) has been widely concerned by scientists, mainly because the commonly used GPS localization system cannot be effectively used in various environments. When GPS localization fails in some scenes such as very messy environments and severe signal occlusion, we can consider using visual-based localization to obtain the pose of the query images. Visual-based localization (VBL) has been widely used in the field of visual tasks, such as augmented reality, unmanned vehicle navigation, robotics, closed-loop detection, SFM (Structure from Motion) models. After years of development, the methods of visual-based localization (VBL) have been enriched and developed, In order to better understand the latest developments in VBL, overall research status and possible future development trends, we need make a systematic detailed classification of VBL. Although the predecessors have summarized the methods of VBL, due to the many new breakthroughs in VBL in recent years, the original summary is not perfect enough. So this paper will make a new and more detailed review of VBL in recent years. This paper divides the visual-based localization methods into three categories: image-based localization, localization based on learning model and localization based on 3D structure. And we also detail the principle, development of methods and the advantages and disadvantages of each method and future development trends.
基于视觉的定位(visual based localization, VBL)是利用周围环境中的各种有用信息，如图像、点云模型、几何信息、语义信息等，在定位系统中获得相应的姿态估计。近年来，基于视觉的定位(visual-based localization, VBL)受到了科学家们的广泛关注，主要原因是目前常用的GPS定位系统无法在各种环境下有效使用。当GPS定位在一些场景中失败时，例如非常混乱的环境和严重的信号遮挡，我们可以考虑使用基于视觉的定位来获得查询图像的位姿。基于视觉的定位(VBL)已广泛应用于增强现实、无人驾驶车辆导航、机器人、闭环检测、SFM (Structure from Motion)模型等视觉任务领域。经过多年的发展，基于视觉的定位方法得到了丰富和发展，为了更好地了解基于视觉的定位的最新发展、总体研究现状和未来可能的发展趋势，我们有必要对基于视觉的定位进行系统详细的分类。虽然前人对VBL的方法进行了总结，但由于近年来VBL出现了许多新的突破，原有的总结还不够完善。因此，本文将对近年来VBL的研究进行新的、更详细的综述。本文将基于视觉的定位方法分为三类:基于图像的定位、基于学习模型的定位和基于三维结构的定位。并详细介绍了各种方法的原理、发展、优缺点以及未来的发展趋势。
{"title":"A review of Visual-Based Localization","authors":"Xing Xin, Jie Jiang, Y. Zou","doi":"10.1145/3366194.3366211","DOIUrl":"https://doi.org/10.1145/3366194.3366211","url":null,"abstract":"The visual-based localization (VBL) obtains the corresponding pose estimation in the localization system by utilizing various useful information in the surrounding environment, such as images, point cloud models, geometric information, semantic information. In recent years, visual-based localization (VBL) has been widely concerned by scientists, mainly because the commonly used GPS localization system cannot be effectively used in various environments. When GPS localization fails in some scenes such as very messy environments and severe signal occlusion, we can consider using visual-based localization to obtain the pose of the query images. Visual-based localization (VBL) has been widely used in the field of visual tasks, such as augmented reality, unmanned vehicle navigation, robotics, closed-loop detection, SFM (Structure from Motion) models. After years of development, the methods of visual-based localization (VBL) have been enriched and developed, In order to better understand the latest developments in VBL, overall research status and possible future development trends, we need make a systematic detailed classification of VBL. Although the predecessors have summarized the methods of VBL, due to the many new breakthroughs in VBL in recent years, the original summary is not perfect enough. So this paper will make a new and more detailed review of VBL in recent years. This paper divides the visual-based localization methods into three categories: image-based localization, localization based on learning model and localization based on 3D structure. And we also detail the principle, development of methods and the advantages and disadvantages of each method and future development trends.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126761359","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we researched and put forward a novel algorithm named equivalent static transformation based on strain energy in the loading process. The algorithm could reduce the working time and make the dynamic analysis and optimization simpler. In order to verify advantage of the method, first, we investigated the strain energy dependence of the load when the structure deformation occurred during loading process theoretically. Then, we identified the key point according to spectral element discrete of the solution space, on which the dynamic load reached its maximum. Thirdly, on the key point, we proposed the model of equivalent static transformation from dynamic loading process, and gave the solution of finding this equivalent static load using DIRECT algorithm. Finally, we analyzed the 10-bar, 124-bar and 2-storey trusses mechanical system under various dynamic loading using this algorithm, and we found that the algorithm was valid on equivalent static transformation.
{"title":"The New Algorithm of Equivalent Transformation from Dynamic Loads Based on Energy","authors":"Yan-gang Zhang","doi":"10.1145/3366194.3366330","DOIUrl":"https://doi.org/10.1145/3366194.3366330","url":null,"abstract":"In this paper, we researched and put forward a novel algorithm named equivalent static transformation based on strain energy in the loading process. The algorithm could reduce the working time and make the dynamic analysis and optimization simpler. In order to verify advantage of the method, first, we investigated the strain energy dependence of the load when the structure deformation occurred during loading process theoretically. Then, we identified the key point according to spectral element discrete of the solution space, on which the dynamic load reached its maximum. Thirdly, on the key point, we proposed the model of equivalent static transformation from dynamic loading process, and gave the solution of finding this equivalent static load using DIRECT algorithm. Finally, we analyzed the 10-bar, 124-bar and 2-storey trusses mechanical system under various dynamic loading using this algorithm, and we found that the algorithm was valid on equivalent static transformation.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123083697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Stock price forecasting is a challenging task. Stock prices are predicted by Twin Support Vector Regression (TWSVR) with two different kernel functions in this paper. The two kernel functions are linear kernel function and polynomial kernel function. The parameters of TWSVR models were selected by genetic algorithm (GA). With the optimized parameters, these models are used to predict the closing prices of the stock in the next day. The predicted results are compared with those obtained by traditional SVR models. The results shown that the TWSVR model with polynomial kernel function has higher accuracy than twin support vector regression with linear kernel. The time consumed by TWSVR is less than that of traditional SVR in prediction.
{"title":"Application of TWSVR Models in Stock Price Forecast","authors":"Haofeng Cui, Xiangfeng Yin, Xueting Wen","doi":"10.1145/3366194.3366200","DOIUrl":"https://doi.org/10.1145/3366194.3366200","url":null,"abstract":"Stock price forecasting is a challenging task. Stock prices are predicted by Twin Support Vector Regression (TWSVR) with two different kernel functions in this paper. The two kernel functions are linear kernel function and polynomial kernel function. The parameters of TWSVR models were selected by genetic algorithm (GA). With the optimized parameters, these models are used to predict the closing prices of the stock in the next day. The predicted results are compared with those obtained by traditional SVR models. The results shown that the TWSVR model with polynomial kernel function has higher accuracy than twin support vector regression with linear kernel. The time consumed by TWSVR is less than that of traditional SVR in prediction.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130787005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abdou Yahouza M. Sani, Tao He, Wenlong Zhao, Tingting Yao
Underwater Robots play an important role in a number of shallow and deep-water missions. In recent years, Underwater Robots system design has been an active field of engineering researches. This paper proposes new system design for underwater vehicles combining the best features of both ROV (remotely operated vehicle) and AUV (autonomous underwater vehicle) technologies. The system based on ROS platform and Ardupilot software is equipped with two types of navigation system setup (autonomous navigation and teleoperation navigation) capable of realizing precise motion control, navigating while performing SLAM (Simultaneous Localization and Mapping) path-planning and obstacles avoidance.
水下机器人在许多浅水和深水任务中发挥着重要作用。近年来，水下机器人系统设计一直是一个活跃的工程研究领域。本文提出了一种新的水下航行器系统设计方案，结合了ROV(遥控航行器)和AUV(自主水下航行器)技术的优点。该系统基于ROS平台和Ardupilot软件，配备自主导航和遥操作导航两种导航系统设置，能够实现精确的运动控制，在导航的同时进行SLAM (Simultaneous Localization and Mapping)路径规划和避障。
{"title":"Hybrid Underwater Robot System Based on ROS","authors":"Abdou Yahouza M. Sani, Tao He, Wenlong Zhao, Tingting Yao","doi":"10.1145/3366194.3366264","DOIUrl":"https://doi.org/10.1145/3366194.3366264","url":null,"abstract":"Underwater Robots play an important role in a number of shallow and deep-water missions. In recent years, Underwater Robots system design has been an active field of engineering researches. This paper proposes new system design for underwater vehicles combining the best features of both ROV (remotely operated vehicle) and AUV (autonomous underwater vehicle) technologies. The system based on ROS platform and Ardupilot software is equipped with two types of navigation system setup (autonomous navigation and teleoperation navigation) capable of realizing precise motion control, navigating while performing SLAM (Simultaneous Localization and Mapping) path-planning and obstacles avoidance.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132840269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Short-term electricity load has the characteristics of timing and non-linearity, which is affected by many factors such as temperature, humidity. This paper proposes a hybrid prediction algorithm based on K-means clustering and long short-term memory (LSTM). K-means, which clusters the highest temperature, the lowest temperature, humidity and other characteristics of the electricity load, divides the data set into K classes. LSTM is utilized to solve nonlinear regressive and time series problem. Simulation results based on real load data show its priority to LSTM without clustering.
{"title":"Effective LSTM with K-means Clustering Algorithm for Electricity Load Prediction","authors":"Dawei Geng, Haifeng Zhang, Ting Xu","doi":"10.1145/3366194.3366279","DOIUrl":"https://doi.org/10.1145/3366194.3366279","url":null,"abstract":"Short-term electricity load has the characteristics of timing and non-linearity, which is affected by many factors such as temperature, humidity. This paper proposes a hybrid prediction algorithm based on K-means clustering and long short-term memory (LSTM). K-means, which clusters the highest temperature, the lowest temperature, humidity and other characteristics of the electricity load, divides the data set into K classes. LSTM is utilized to solve nonlinear regressive and time series problem. Simulation results based on real load data show its priority to LSTM without clustering.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131010107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A remote human-machine collaboration for the positioning arm for interventional operation is proposed. The existing research orientation of the arm is combined for further study. 1) Operator's end position is detected based on the determined position and posture of the target, and the position and posture mapping mechanism between the operator and the end effector of the arm is researched based on the virtual motion prediction. 2) The strategy of telepresence is realized based on the bidirectional force mapping between the operator and the positioning arm. 3) The mechanism of man-machine cooperative control and safety control mechanism of the precise positioning of the arm is studied. 4) The recognition, tracking, force feedback and other verification and evaluation work of the robotic arm positioning process based on remote motion following is completed. The key problem is the realization of the telepresence of telecontrol and teleoperation of the interventional operation positioning arm, and bi-directional perception of human in the loop. Remote precise positioning of the arm based on human action recognition, action following and force telepresence based on operational force detection and collision force feedback is implemented. The research work is directly satisfied with the clinical needs.
{"title":"Construction of a Teleoperational Interventional Surgery Robot System","authors":"X. Yang, Zhaocui Han, Yingjun Li","doi":"10.1145/3366194.3366292","DOIUrl":"https://doi.org/10.1145/3366194.3366292","url":null,"abstract":"A remote human-machine collaboration for the positioning arm for interventional operation is proposed. The existing research orientation of the arm is combined for further study. 1) Operator's end position is detected based on the determined position and posture of the target, and the position and posture mapping mechanism between the operator and the end effector of the arm is researched based on the virtual motion prediction. 2) The strategy of telepresence is realized based on the bidirectional force mapping between the operator and the positioning arm. 3) The mechanism of man-machine cooperative control and safety control mechanism of the precise positioning of the arm is studied. 4) The recognition, tracking, force feedback and other verification and evaluation work of the robotic arm positioning process based on remote motion following is completed. The key problem is the realization of the telepresence of telecontrol and teleoperation of the interventional operation positioning arm, and bi-directional perception of human in the loop. Remote precise positioning of the arm based on human action recognition, action following and force telepresence based on operational force detection and collision force feedback is implemented. The research work is directly satisfied with the clinical needs.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114161653","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The magnetic circuit environment is the most important factor restricting the normal operation of the magnetorheological fluid (MRF) device. A good magnetic circuit environment can make the MRF dynamometer more economical and practical. In this paper, a new MRF dynamometer is used as the research object. The magnetic field simulation is carried out by Ansoft Maxwell software to study the influence of magnetic isolation ring, magnetic permeability material, Working gap and Excitation current on the magnetic field performance of MRF dynamometer. The research shows that the magnetic induction line distribution in the working gap of the MRF dynamometer is relatively uniform, the magnetic induction intensity is in the range of 120-162 mT, For materials with higher permeability, the stronger the magnetic induction intensity in the working gap is. This research has certain guiding significance for the magnetic circuit optimization of MRF dynamometer.
{"title":"Influencing Factors for Magnetic Circuit Environment of the Magnetorheological Fluid Dynamometer","authors":"Yameng Chen, Yiping Luo, Bin Yang","doi":"10.1145/3366194.3366307","DOIUrl":"https://doi.org/10.1145/3366194.3366307","url":null,"abstract":"The magnetic circuit environment is the most important factor restricting the normal operation of the magnetorheological fluid (MRF) device. A good magnetic circuit environment can make the MRF dynamometer more economical and practical. In this paper, a new MRF dynamometer is used as the research object. The magnetic field simulation is carried out by Ansoft Maxwell software to study the influence of magnetic isolation ring, magnetic permeability material, Working gap and Excitation current on the magnetic field performance of MRF dynamometer. The research shows that the magnetic induction line distribution in the working gap of the MRF dynamometer is relatively uniform, the magnetic induction intensity is in the range of 120-162 mT, For materials with higher permeability, the stronger the magnetic induction intensity in the working gap is. This research has certain guiding significance for the magnetic circuit optimization of MRF dynamometer.","PeriodicalId":105852,"journal":{"name":"Proceedings of the 2019 International Conference on Robotics, Intelligent Control and Artificial Intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114367377","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：