{"title":"Accumulation of adversarial examples for underwater visual object tracking","authors":"Yu Zhang, Jin Li, Chenghao Zhang","doi":"10.1109/ICMA54519.2022.9855996","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9855996","url":null,"abstract":"Recent studies show that small perturbations in video frames could misguide the deep learning-based visual object trackers. In this paper, we first attempt to generate an accumulation of adversarial examples for underwater VOT (Visual object tracking). This is the first attempt at underwater VOT attack. The data used are the URPC (Underwater Robot Picking Contest) in 2017 and 2018, and the fish2 subset of the VOT2019(Visual Object Tracking Challenge in 2019). We generate adversarial examples by minimizing the L2 total loss function which we designed. Experiments show that this attack method can achieve an effective attack, which can reduce the success rate of at least 48.6% of the DaSiamRPN(Distractor-aware SiamRPN) trackers.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125248466","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}
{"title":"Research on Rainfall Intensity Level Retrieval Based on Wave-Number Energy Spectrum from Rain-Contaminated X-band Marine Radar Image","authors":"Lei Sun, Zhizhong Lu, Hong Liu, Runbo Zhang","doi":"10.1109/ICMA54519.2022.9855919","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9855919","url":null,"abstract":"Rainfall can change the wavenumber distribution in the energy spectrum of radar image. Based on this feature, a wave-number energy spectrum method for retrieving rainfall intensity level in the rain-contaminated radar image is proposed in this paper. A large number of parameter information in the energy spectrum under different rainfall intensities are used to determine the rainfall intensity fitting function, and the rainfall intensity level is retrieved in real time according to the classification standard. Compared with the rainfall recorded synchronously by rain gauge, the experimental results demonstrate that the proposed method has satisfactory retrieval performance of rainfall intensity level.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125220962","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}
{"title":"A Fuzzy Indrect Adaptive Robust Control for Upper Extremity Exoskeleton Driven by Pneumatic Artificial Muscle","authors":"Siyuan Dan, Haoshu Cheng, Yong Zhang, Hao Liu","doi":"10.1109/ICMA54519.2022.9856383","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9856383","url":null,"abstract":"Pneumatic artificial muscle (PAM), a new type of flexible and high force-to-weight ratio actuator, is widely used in the fields of exoskeleton and bionic robot. Despite these advantages, the dynamics of the flow and pressure as well as the time-varying behavior have increased the nonlinearity of it, which affects the control accuracy of the system significantly. Therefore, high-efficiency and high-precision control algorithms have become the key to improve the motion control performance of PAM. In this paper, a new fuzzy indirect adaptive robust control algorithm (FIARC) is proposed for an upper extremity exoskeleton driven by PAM. The FIARC introduces a modified fuzzy radial basis function (RBF) and a fuzzy control logic integrated with indirect adaptive robust control (IARC) to reduce the control system error effectively. The modified fuzzy RBF is used to compensate the model uncertainty and the deviation of true control paraments, while the fuzzy control logic is used for real-time flow coefficient correction, which combines the flow hysteresis characteristics of the high speed on/off valve with the inlet and exhaust flow model of pneumatic muscle. To verify the performance of the proposed controller, the FIARC is applied to the motion control of elbow joint in the upper extremity exoskeleton. Experimental results show that the FIARC is more advanced and effective in tracking the reference signals compared with conventional method.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114005209","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}
Yi Wei, Qingfang Zhang, Xueshan Gao, Peng Liang, Mingkang Li, Kejie Li
{"title":"Aerodynamic Analysis of a Wall-Climbing Robot with Dual-propeller","authors":"Yi Wei, Qingfang Zhang, Xueshan Gao, Peng Liang, Mingkang Li, Kejie Li","doi":"10.1109/ICMA54519.2022.9855936","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9855936","url":null,"abstract":"A wall-climbing robot is a mobile robot that operates on a vertical wall. A wall-climbing robot with dual-propeller is proposed in this paper, in which the aerodynamics of the propeller is an important research part. Therefore, the aerodynamics of the propeller is simulated by ANSYS Fluent and experimentally analyzed in this paper. First, the aerodynamic simulation analysis of the single rotor and the double rotor is carried out in this paper, and then the tensile test experiment of the single propeller is carried out by using the pulling machine. Finally, an experimental platform for the whole machine was built, and the tensile test experiment was carried out on the double propellers. It is concluded that the aerodynamic forces generated by the two rotors are not linearly superimposed due to the airflow interference, and the rotational speed of the rotors and the distance between the rotor platforms all affect the magnitude of the force.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131368844","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}
Mingkang Li, Xueshan Gao, Qingfang Zhang, Peng Liang, Yi Wei, Kejie Li
{"title":"Force Analysis and Verification of Wall-climbing Robot with Rotor-propellers","authors":"Mingkang Li, Xueshan Gao, Qingfang Zhang, Peng Liang, Yi Wei, Kejie Li","doi":"10.1109/ICMA54519.2022.9856259","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9856259","url":null,"abstract":"Wall-climbing robot can adapt to the wall surface of a variety of different materials, and can perform a variety of tasks, such as cleaning glass curtain walls and inspecting bridges. In this paper, a rotor-propeller type wall-climbing robot is researched and developed. The model aircraft propeller is used as the adsorption unit. The rotor platform has one roll degree of freedom, one pitch degree of freedom. Based on the vector superposition method, this paper considers the difference between the support forces of each wheel of the wall-climbing robot, and analyzes the relationship between the wheel and the wall of the wall-climbing robot. Furthermore, the effects of the anisotropic friction of the wheels and the roll joints on the lateral resultant force of the wall-climbing robot are studied. The results show that the roll joint is beneficial to the stable adsorption of the robot on the wall. Finally, a wall motion experiment is carried out in an indoor environment to verify the feasibility of the robot and the correctness of the theory.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131642844","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}
{"title":"Research on EMG Signal of Human Lower Limbs Based on Empirical Mode Decomposition","authors":"Jun-yao Wang, Yue-hong Dai, Xiaxi Si","doi":"10.1109/ICMA54519.2022.9856078","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9856078","url":null,"abstract":"To recognize Electromyography (EMG) signal of lower limbs more effectively, this paper analyzed EMG signal of gastrocnemius muscle during dorsal flexion, plantar flexion, knee flexion, knee flexion and hip flexion; Based on advantages of Empirical Mode Decomposition (EMD) in nonlinear and non-stationary signal analysis, Intrinsic Mode Function (IMF) was utilized as eigenvalue of EMG signal of gastrocnemius muscle; Support Vector Machine (SVM) was applied to verify influence on different orders of IMF in the classification effect. Results shown that IMF of gastrocnemius muscle under different movements is different. The orders of ankle dorsal flexion and plantar flexion are 5, knee flexion and extension are 6, and hip flexion is 7; when eigenvalues of higher order are discarded, the recognition rate for 5 movements is low (81.46%); The value is higher when eigenvalue matrix is supplemented by 0 (85.3%). the recognition rate is the highest when combining this two methods (90.42%).","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133591765","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}
{"title":"Design and Optimization of Manipulator for Live-line Working Robot Based on Manipulability","authors":"Y. Li, Afen Zhou, Huan Yao, Yuxiang Zheng, Erbao Dong, Yu Feng, Kai-Kai Wu, Shaolei Wu, Hao Zhang, Xuming Tang","doi":"10.1109/ICMA54519.2022.9855928","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9855928","url":null,"abstract":"Live-line working robot on distribution network always is a hot topic of research. However, limited by structural dimensions and safety considerations of insulation, manipulator arms, as robot actuators, have always been less dexterous and redundant. In this paper, the manipulator arm is modeled base on MD-H conventions, and the manipulability is used as the evaluation index for structural design and optimization. Then, the parameters requirement of motors for each joint is calculated based on fifth-order polynomial interpolation, and the motors are selected reasonably. And the two designed manipulator arms are integrated into a same platform to obtain a dual-arm live working robot in V-shaped configuration for greater workspace and flexibility. For safety reasons, a 10 centimeters long insulated section is added at the end of the manipulator arm to meet the insulation requirement and tasks for the 10kV live working. And finite element insulation analysis withstand voltage simulation is performed to prove its validity. Finally, functional tests of the manipulator arm were conducted to prove that the performance of the manipulator arm and the design of insulation protection meet the requirements of live working robot on distribution network. Based on the dual-arm live working robot with the designed manipulator, outdoor operation experiments are performed to verify the feasibility and flexibility of the designed manipulator with optimization of manipulability.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134346910","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}
{"title":"Circle Area Detection Based on Convolutional Neural Networks","authors":"Tiantian Hao, De Xu","doi":"10.1109/ICMA54519.2022.9856373","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9856373","url":null,"abstract":"It is difficult for the traditional circle detection methods to locate the circle areas in an image. In this paper, we develop an end-to-end convolutional neural network (CNN) to detect circle areas for objects. In the CNN, an edge feature extraction module is designed to extract the edge feature. Then the edge and high-level features are fused in order to locate the circle areas accurately. A circle dataset with 4024 images is built for the CNN training. Several circles are embedded into the image, which will be used to train the CNN, in order to realize data augment simply and efficiently. The comparison experiments to the state-of-art CNN-based methods are well conducted. Our circle detector achieves the average precision 92.7% and recall rate 95.8% on datasets, and the time-cost is at the same level with others. The effectiveness of the proposed method is validated by a series of experiments.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133136241","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}
Mengqi Cheng, Nan Xiao, H. Yuan, Haoyang Liu, Pinzheng Ni
{"title":"A Novel Guide Wire Rotation and Delivery Structure for Interventional Surgical Robot","authors":"Mengqi Cheng, Nan Xiao, H. Yuan, Haoyang Liu, Pinzheng Ni","doi":"10.1109/ICMA54519.2022.9856010","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9856010","url":null,"abstract":"In recent years, with the development of vascular interventional surgery robots, the use of surgical robots for angiography surgery has begun to be accepted by more and more doctors. With this tool, the doctor can operate the catheter and guide wire several meters away from the patient to complete the operation, which greatly reduces the radiation received by the radiologist and protects the doctor’s health. However, the traditional interventional surgery robot still has the following problems. Because the guide wire surface is too smooth, when encountering resistance, the guide wire is easy to slide, resulting in the reduction of the accuracy of guide wire rotation and delivery. The guide wire rotation and delivery mechanism of traditional interventional surgery robots are connected with the robot body, and the sterile environment is not completely realized. In this study, a new guide wire rotation and delivery transmission device is proposed, which can improve the accuracy of guide wire rotation and delivery without damaging the guide wire. Through experiments, we have verified that the guide wire rotation and delivery device meets the functional integrity and the accuracy of the guide wire delivery function. The experimental results show that the accuracy of the guide wire rotation and delivery transmission device can meet the requirements of vascular interventional surgery.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133389285","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}
{"title":"Inter-subject Variability Evaluation of Continuous Elbow Angle from sEMG using BPNN","authors":"Hengrui Li, Shuxiang Guo, Dongdong Bu, Hanze Wang","doi":"10.1109/ICMA54519.2022.9856005","DOIUrl":"https://doi.org/10.1109/ICMA54519.2022.9856005","url":null,"abstract":"As a non-invasive approach, surface electromyographic (sEMG) signal has great potential for application in human-robot interfaces, such as the upper-limb exoskeleton rehabilitation device. However, due to the differences in activity level of muscles, there exists high inter-subject variability. In this work, the influence of inter-subject variability for elbow continuous motion is evaluated through a shallow neural network (BPNN), and user-dependent and user-independent models are established respectively. In user-dependent model, training and testing sets are from the same subject, new set of the same person as during training is used as the input of network. The user-independent models are constructed by the same user and another additional user to determine inter-subject variability in model construction. To evaluate the degree of inter-subject variability, evaluation criteria and statistical method are adopted. Through the prediction results, and further the value of evaluation criteria and the plot of statistical method, it can be seen that the inter-subject variability on sEMG has a huge impact on the regression of elbow continuous angle, which can provide reference for the future study of building sEMG generalized modeling to estimate elbow angles.","PeriodicalId":120073,"journal":{"name":"2022 IEEE International Conference on Mechatronics and Automation (ICMA)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114568545","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}