2021 IEEE International Conference on Mechatronics and Automation (ICMA)最新文献_第6页

YOLO Target Detection Algorithm with Deformable Convolution Kernel 基于可变形卷积核的YOLO目标检测算法

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512626

Hui Wang, Shuai Zhang, Lijun Yu, Ce Shi

引用次数: 1

Research on Underwater Target Detection Method Based on Improved MSRCP and YOLOv3 基于改进MSRCP和YOLOv3的水下目标检测方法研究

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512827

Tongxu Guo, Yanhui Wei, Hong Shao, Bo Ma

{"title":"Research on Underwater Target Detection Method Based on Improved MSRCP and YOLOv3","authors":"Tongxu Guo, Yanhui Wei, Hong Shao, Bo Ma","doi":"10.1109/ICMA52036.2021.9512827","DOIUrl":"https://doi.org/10.1109/ICMA52036.2021.9512827","url":null,"abstract":"With the increasing demand for marine resources, the development of underwater visual robot technology has become more and more important as the main tool for developing marine resources. Aiming at the problem of insufficient illumination in the underwater environment and slow detection speed in the detection of marine biological targets, a rapid marine organisms detection scheme based on the improved MSRCP image enhancement algorithm and YOLOv3 is designed. First, in order to solve the problem of fuzzy and low contrast of underwater images, an improved MSRCP image enhancement algorithm based on a self-adaptive algorithm is proposed to clarify the collected underwater degraded images. We chose YOLOv3 as our detection method,which has high processing speed and high detection accuracy. A large-scale underwater biological data set is established,and we use the Darknet-53 network to train the data set. Finally, we use the trained network to detect the enhanced image and compare it with other target detection algorithms. Experimental results show that the proposed scheme can quickly and accurately detect underwater targets.","PeriodicalId":339025,"journal":{"name":"2021 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133240219","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}

引用次数: 8

A System to Identify Walking Pattern Using Machine Learning for a Load Reduction Exoskeleton Robot 基于机器学习的减载外骨骼机器人行走模式识别系统

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512780

Hanqing Zhao, Hidetaka Nambo

引用次数: 0

Active Noise Cancellation Techniques to Enhance Audition in Noisy Cities 在嘈杂的城市中提高听力的主动降噪技术

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512571

J. A. Calderón Ch., Johnathan Lozano, Benjamín Barriga, J. Tafur, J. C. Lengua, Gonzalo Solano

引用次数: 1

An Elastic Biomimetic Fish Tail and Its Undulation Fitting Method of Body Wave 弹性仿生鱼尾及其体波波动拟合方法

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512706

Xiaocun Liao, Yuzhuo Fu, Ben Lu, Qianqian Zou, Zhuoliang Zhang, Chao Zhou

{"title":"An Elastic Biomimetic Fish Tail and Its Undulation Fitting Method of Body Wave","authors":"Xiaocun Liao, Yuzhuo Fu, Ben Lu, Qianqian Zou, Zhuoliang Zhang, Chao Zhou","doi":"10.1109/ICMA52036.2021.9512706","DOIUrl":"https://doi.org/10.1109/ICMA52036.2021.9512706","url":null,"abstract":"This paper presents a novel wire-driven fish tail with two joints, using spring-steel-sheet to mimic fish spine, which makes the whole fish tail compliant and can simulate real fish better. This mechanism can also achieve C-shape and S-shape swing similar to natural fish tail. To predict the attitude of fish tail, kinematics model is established theoretically in this paper. The simulation experiments show that the maximum error between the amplitude predicted by the kinematics model and expected amplitude at the end of the fish tail is 5 mm, indicating the correctness of the proposed kinematics model. In order to solve the control coupling problem of servomotors, this paper further proposes a based-vision biomimetic undulation fitting method, which is both simple and efficient. Finally, the proposed method is verified by two groups of experiments involving forward motion and turn motion. Experimental results are in good agreement with the expected values. When the frequency is 0.1 Hz, the average errors of forward motion and turn motion are 0.59° and 0.7° respectively, which proves that the proposed method is reliable and effective.","PeriodicalId":339025,"journal":{"name":"2021 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":"73 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115569932","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}

引用次数: 1

Trajectory Planning of an Underactuated Cable-Driven Planar Device for the Trunk 一种欠驱动平面主干缆索驱动装置的轨迹规划

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512724

Lailu Li, Shuoyu Wang, Guang Yang

引用次数: 0

Classification of Aquatic Animals by the Spherical Amphibian Robot based on Transfer Learning 基于迁移学习的球形两栖机器人对水生动物的分类

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512599

Shuxiang Guo, Shaolong Wang, Jian Guo, Jigang Xu

{"title":"Classification of Aquatic Animals by the Spherical Amphibian Robot based on Transfer Learning","authors":"Shuxiang Guo, Shaolong Wang, Jian Guo, Jigang Xu","doi":"10.1109/ICMA52036.2021.9512599","DOIUrl":"https://doi.org/10.1109/ICMA52036.2021.9512599","url":null,"abstract":"The spherical robot is mainly used for normal observation of aquaculture biology. The performance of aquatic biological image recognition mainly depends on the feature extraction and the selected classifier. Traditional manual extraction methods often cannot meet actual application requirements, and have problems such as poor accuracy and weak generalization ability. To solve the above problems, a small data set aquatic animal classification model based on convolutional neural network and transfer learning is proposed in the spherical robot. First, the original images of aquatic animals is preprocessed, and the data set is enhanced using the data increment method. Second, The original CNN model is then improved by embedding the SE module and using the triplet loss function to replace the softmax loss function. Finally, Transfer learning a deep pre-trained model of the ImageNet image data set. Training and fitting parameter distributions on aquatic image data sets. Experimental results show that the model optimizes the accuracy of aquatic animal target recognition, and the test accuracy reaches 93.11%.The model has good stability and high precision in aquaculture environment.","PeriodicalId":339025,"journal":{"name":"2021 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121464640","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}

引用次数: 1

Vehicle Pose Estimation System Base on Pressure Sensor Array for Clamping Parking Robot 基于压力传感器阵列的泊车机器人姿态估计系统

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512772

Juzhong Zhang, Yuyi Chu, Zhisen Wang, Tingfeng Ye, Liming Cai, Hongbo Yang

{"title":"Vehicle Pose Estimation System Base on Pressure Sensor Array for Clamping Parking Robot","authors":"Juzhong Zhang, Yuyi Chu, Zhisen Wang, Tingfeng Ye, Liming Cai, Hongbo Yang","doi":"10.1109/ICMA52036.2021.9512772","DOIUrl":"https://doi.org/10.1109/ICMA52036.2021.9512772","url":null,"abstract":"Vehicle pose estimation is one of the most essential functions for the clamping parking robot. In order to detect the parameter information of a car accurately, laser scanners are used widely for their highly precise distance measurements. However, the price of laser scanners, especially 3D laser scanners, is usually very expensive, and the pose estimating algorithm is complex. In this paper, an innovative vehicle pose estimation system based on pressure sensor array is proposed. In order to improve the estimation accuracy, a rectangle fitting based on the least square method are applied. The suggested algorithm is simple and effective, it can not only estimate the pose of a car, but also achieve the parameters of wheel width, wheel tread, wheel base, weight and gravity center position. In addition, whether the steering wheel is back can also be detected. All the information above is important for clamping parking robot. What is more, it's much cheaper than laser scanners. In order to verify the algorithm, two experiment prototypes were built. The results proved that the proposed system provides an effective method of vehicle pose estimation for clamping parking robot.","PeriodicalId":339025,"journal":{"name":"2021 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124800122","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}

引用次数: 1

Trajectory Tracking Control of an Amphibious Spherical Robot Using MPC Approach 基于MPC方法的两栖球形机器人轨迹跟踪控制

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512688

Meng Liu, Shuxiang Guo, Liwei Shi, Xihuan Hou, He Yin, Ao Li, Zan Li, Debin Xia, Mugen Zhou

引用次数: 1

Development of a Knee Extensors Training System with Active Resistance Component 具有主动阻力成分的膝关节伸肌训练系统的开发

2021 IEEE International Conference on Mechatronics and Automation (ICMA) Pub Date : 2021-08-08 DOI: 10.1109/ICMA52036.2021.9512677

Z. Gu, Ko Matsuhiro, Jiei Yanagi, Kazuhiko Mizukami, Ryuya Watanabe, Hiroaki Eto, S. Cosentino, A. Takanishi

{"title":"Development of a Knee Extensors Training System with Active Resistance Component","authors":"Z. Gu, Ko Matsuhiro, Jiei Yanagi, Kazuhiko Mizukami, Ryuya Watanabe, Hiroaki Eto, S. Cosentino, A. Takanishi","doi":"10.1109/ICMA52036.2021.9512677","DOIUrl":"https://doi.org/10.1109/ICMA52036.2021.9512677","url":null,"abstract":"Knee extensors play a key role in maintaining balance and stabilizing gait. Insufficient training of knee extensors leads to muscle atrophy and progressive overall body instability. However, training knee extensors requires specific exercises which are not easily integrated in activities of daily life. For example, even if walking is one of the most popular daily exercise, its training effects are limited to lower leg muscles. To better integrate knee extensor training in daily life we are developing a training device which can apply additional mechanical load on the knee joint during daily normal walking, to extend the training effects of simple walking to knee extensors. We projected a longterm study to develop an integrated system which combines a training function with an assisting function. Integrated sensors measure the biological data of muscular activity to automatically select the proper mode in real-time. In 2019, we designed a prototype of the training device and evaluated its effects on the knee extensor muscle with preliminary experiments. The results validated the feasibility of the training system and showed that the training load need to be adapted to the gait phase during walking. In this study, we improved the training device with active resistance component to adjust the training load in real-time.","PeriodicalId":339025,"journal":{"name":"2021 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":"112 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127968153","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}

引用次数: 0