Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536125
Lin Lang, Hui Peng, Junhao Xiao, Huimin Lu, Zongtan Zhou
Human performance augmentation exoskeletons have a wide range of applications in the military, industry, construction and other environments. The control methods of augmentation exoskeletons are crucial to ensure effective assistance and the safety of the systems. We proposed a scale force control method that considers both cognitive human-robot interaction and physical human-robot interaction. Experiments on an elbow joint exoskeleton verified the feasibility of the method.
{"title":"Scale Force Control of a Robot Bearing Augmentation Exoskeleton","authors":"Lin Lang, Hui Peng, Junhao Xiao, Huimin Lu, Zongtan Zhou","doi":"10.1109/ICARM52023.2021.9536125","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536125","url":null,"abstract":"Human performance augmentation exoskeletons have a wide range of applications in the military, industry, construction and other environments. The control methods of augmentation exoskeletons are crucial to ensure effective assistance and the safety of the systems. We proposed a scale force control method that considers both cognitive human-robot interaction and physical human-robot interaction. Experiments on an elbow joint exoskeleton verified the feasibility of the method.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124958702","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536158
Hao Wu, Tianya You, Xiangrong Xu, A. Rodic, P. Petrovic
In this paper, an imaginary part of Gabor features based feature selection and classification method for electronic component solder joint inspection have been proposed. Using an image acquisition system, the images of component solder can be obtained. The RGB color solder joint image is transformed into HSI (Hue-Saturation-Intensity) model space and the image Gabor features is extracted. Then based on the algorithm of principal component analysis (PCA), feature selection is conducted. finally, the solder joint is classified using the support vector machine (SVM). The proposed inspection scheme improves efficiency and recognition, since it extracts the Gabor features and reduces the input image dimension through feature selection.
{"title":"Solder Joint Inspection Using Imaginary Part of Gabor Features","authors":"Hao Wu, Tianya You, Xiangrong Xu, A. Rodic, P. Petrovic","doi":"10.1109/ICARM52023.2021.9536158","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536158","url":null,"abstract":"In this paper, an imaginary part of Gabor features based feature selection and classification method for electronic component solder joint inspection have been proposed. Using an image acquisition system, the images of component solder can be obtained. The RGB color solder joint image is transformed into HSI (Hue-Saturation-Intensity) model space and the image Gabor features is extracted. Then based on the algorithm of principal component analysis (PCA), feature selection is conducted. finally, the solder joint is classified using the support vector machine (SVM). The proposed inspection scheme improves efficiency and recognition, since it extracts the Gabor features and reduces the input image dimension through feature selection.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130308456","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536053
Xuefeng Wang, Yu Kang, Yang Cao
Modeling and control of nonlinear robotic systems have been challenging tasks. If a linear approximate embed-ding space for nonlinear dynamical robotic systems can be constructed, well-established techniques in the field of linear systems are expected to be used to deal with this problem. The Koopman theory suggests that a data-driven approach can be used to construct a suitable set of observation functions to map the nonlinear system into an equivalent linear model in the embedding space. We use deep neural networks to construct more adaptive sets of observation functions, treat the control inputs as generalized states, learn the input-Koopman operator of the controlled nonlinear robotic system, and construct the embedded linear state model DKoopman-predictor. The learned linear state model is then used to design the model prediction controller (DKoopman-MPC) to control the original nonlinear robotic system. The proposed approach is easy to implement and is data-driven without the need for a priori knowledge of model dynamics. Our experiments on mobile robot modeling and control show that the proposed method has higher model fidelity than existing local linearization methods, achieving 79.27% error reduction in the prediction task and has good convergence properties in the control task.
{"title":"Deep Koopman Operator Based Model Predictive Control for Nonlinear Robotics Systems","authors":"Xuefeng Wang, Yu Kang, Yang Cao","doi":"10.1109/ICARM52023.2021.9536053","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536053","url":null,"abstract":"Modeling and control of nonlinear robotic systems have been challenging tasks. If a linear approximate embed-ding space for nonlinear dynamical robotic systems can be constructed, well-established techniques in the field of linear systems are expected to be used to deal with this problem. The Koopman theory suggests that a data-driven approach can be used to construct a suitable set of observation functions to map the nonlinear system into an equivalent linear model in the embedding space. We use deep neural networks to construct more adaptive sets of observation functions, treat the control inputs as generalized states, learn the input-Koopman operator of the controlled nonlinear robotic system, and construct the embedded linear state model DKoopman-predictor. The learned linear state model is then used to design the model prediction controller (DKoopman-MPC) to control the original nonlinear robotic system. The proposed approach is easy to implement and is data-driven without the need for a priori knowledge of model dynamics. Our experiments on mobile robot modeling and control show that the proposed method has higher model fidelity than existing local linearization methods, achieving 79.27% error reduction in the prediction task and has good convergence properties in the control task.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129134839","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536110
Zongyi Mu, Y. Ran, Genbao Zhang, Guangqi Ying
Design for key Meta-action unit is an effective method to reduce the design difficulty of mechanical products. Reliability is an important design index for the Meta-action unit. Traditional reliability evaluation method evaluates the Meta-action unit’s reliability through the fault data after it fails. It means that traditional reliability evaluation method cannot evaluate the Meta-action unit’s reliability during its running process and provide reference for the initial design or reliability condition monitoring. To solve the above problems, this paper evaluated the Meta-action unit’s reliability by using the running data on the basis of its all fault modes. In this paper, firstly, the mechanical system is decomposed by FMA decomposition method and the Meta-action units of the mechanical system is obtained. Then, reliability of the Meta-action unit is evaluated by combining its all fault modes and running data. The reliability evaluation method based on all fault modes can provides references for the initial design and reliability condition monitoring of the Meta-action unit.
{"title":"Reliability Evaluation Method for Meta-action Unit Based on the All Fault Modes","authors":"Zongyi Mu, Y. Ran, Genbao Zhang, Guangqi Ying","doi":"10.1109/ICARM52023.2021.9536110","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536110","url":null,"abstract":"Design for key Meta-action unit is an effective method to reduce the design difficulty of mechanical products. Reliability is an important design index for the Meta-action unit. Traditional reliability evaluation method evaluates the Meta-action unit’s reliability through the fault data after it fails. It means that traditional reliability evaluation method cannot evaluate the Meta-action unit’s reliability during its running process and provide reference for the initial design or reliability condition monitoring. To solve the above problems, this paper evaluated the Meta-action unit’s reliability by using the running data on the basis of its all fault modes. In this paper, firstly, the mechanical system is decomposed by FMA decomposition method and the Meta-action units of the mechanical system is obtained. Then, reliability of the Meta-action unit is evaluated by combining its all fault modes and running data. The reliability evaluation method based on all fault modes can provides references for the initial design and reliability condition monitoring of the Meta-action unit.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129853813","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536102
Tao Zhang, Zhijun Li, Qinjian Li, Ying Feng
This paper proposes a trajectory generation of prosthesis to help amputee crossing over obstacles. Dynamical movement primitives (DMPs) are popular methods of reproducing trajectory for learning control. In the basic of DMPs, a novel term of obstacle is added to generate the trajectory in real-time. This term includes multiple point obstacle sources to reflect spatial size of obstacles. Each point obstacle is used to calculate direction relative to current position. Moreover, the direction vector and velocity vector are considered into term of obstacle. Besides, DMPs can generate effective trajectory through obstacle term parameter adjusting regardless of the obstacles in the front, middle or behind position in a step. Therefore, our method solves the inability crossing over obstacles under some scenarios and helps the prosthesis adapt to the various environment. Gaussian mixture regression is used to conjunction with DMPs for movement representation, which can reproduce new trajectory from multiple sets of original trajectories. Finally, simulations for DMPs were performed to demonstrate the flexibility of DMPs with obstacle term, which can generate desired trajectory of crossing over obstacles in various situations, indicating our method has a potential to facilitating prosthesis motion control.
{"title":"Dynamical Movement Primitives of Crossing over Obstacles for a Lower-Limb Prosthesis","authors":"Tao Zhang, Zhijun Li, Qinjian Li, Ying Feng","doi":"10.1109/ICARM52023.2021.9536102","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536102","url":null,"abstract":"This paper proposes a trajectory generation of prosthesis to help amputee crossing over obstacles. Dynamical movement primitives (DMPs) are popular methods of reproducing trajectory for learning control. In the basic of DMPs, a novel term of obstacle is added to generate the trajectory in real-time. This term includes multiple point obstacle sources to reflect spatial size of obstacles. Each point obstacle is used to calculate direction relative to current position. Moreover, the direction vector and velocity vector are considered into term of obstacle. Besides, DMPs can generate effective trajectory through obstacle term parameter adjusting regardless of the obstacles in the front, middle or behind position in a step. Therefore, our method solves the inability crossing over obstacles under some scenarios and helps the prosthesis adapt to the various environment. Gaussian mixture regression is used to conjunction with DMPs for movement representation, which can reproduce new trajectory from multiple sets of original trajectories. Finally, simulations for DMPs were performed to demonstrate the flexibility of DMPs with obstacle term, which can generate desired trajectory of crossing over obstacles in various situations, indicating our method has a potential to facilitating prosthesis motion control.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126372882","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536114
Ting-Bo Chen, Wenlian Huang, Xiaonan Hu, Zun Liu, Jie Chen, Zhuangzhuang Chen, Jianqiang Li, Junxin Liu, Xiao-Fan Ye
Person counting at the construction site is of great significance to person scheduling and progress supervision. This work aims to develop a person counting framework, which is composed of three modules: person detection network, person re-identification module and person counting module. In order to solve the problem of repeated counting caused by Unmanned Aerial Vehicles (UAVs) taking multiple shots of the same person, we propose a Graph Similarity-based Person Counting Network (GSPCN), which can re-identify persons and avoid counting the same person multiple times. We tested the proposed framework both on simulation environment and real datasets (our data was collected from multiple construction sites in Shenzhen), It is empirically superior to the most advanced methods available.
{"title":"Person Counting Based On Graph Relation Network Using UAV","authors":"Ting-Bo Chen, Wenlian Huang, Xiaonan Hu, Zun Liu, Jie Chen, Zhuangzhuang Chen, Jianqiang Li, Junxin Liu, Xiao-Fan Ye","doi":"10.1109/ICARM52023.2021.9536114","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536114","url":null,"abstract":"Person counting at the construction site is of great significance to person scheduling and progress supervision. This work aims to develop a person counting framework, which is composed of three modules: person detection network, person re-identification module and person counting module. In order to solve the problem of repeated counting caused by Unmanned Aerial Vehicles (UAVs) taking multiple shots of the same person, we propose a Graph Similarity-based Person Counting Network (GSPCN), which can re-identify persons and avoid counting the same person multiple times. We tested the proposed framework both on simulation environment and real datasets (our data was collected from multiple construction sites in Shenzhen), It is empirically superior to the most advanced methods available.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121446126","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536085
Xiuheng Zhang, Heng Zhang, Zhen Li, Guibin Bian
In neurosurgery, bipolar forceps is used to coagulate and dissect brain tissues. As in robot-assisted neurosurgical scenario, force perception on bipolar forceps can quantify intraoperative contacting forces between tips of forceps and brain tissues. The quantified tool-tissue contacting force information can help quantify experts experience and ensure safety of robot-assisted surgery. It is challengeable to facilitate the forceps with forces sensing ability because of the limited space for installing sensors. The craniocerebral tissues, interstitial fluids and currents passing though polar of forceps also bring restrictions on force sensing methods. In this article, we propose a planar force perception method to collect force information on bipolar forceps tips in real-time by installing two Fiber Bragg Grating Sensors (FBGs) perpendicularly on each polar of bipolar forceps. FBGs are free from electromagnetic interference. The slim configuration and silicon materials of FBGs make it a suitable sensing method for little disturbance on working state of bipolar forceps. In experiments, axial forces and planar forces were monitored in real-time. It is possible to sense forces at a minimum of 0.01 N. The calibrating revolution on x, y axis is 0.01, 0.03 N, separately.
{"title":"The Planar Force Perception on Robotic Bipolar Forceps","authors":"Xiuheng Zhang, Heng Zhang, Zhen Li, Guibin Bian","doi":"10.1109/ICARM52023.2021.9536085","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536085","url":null,"abstract":"In neurosurgery, bipolar forceps is used to coagulate and dissect brain tissues. As in robot-assisted neurosurgical scenario, force perception on bipolar forceps can quantify intraoperative contacting forces between tips of forceps and brain tissues. The quantified tool-tissue contacting force information can help quantify experts experience and ensure safety of robot-assisted surgery. It is challengeable to facilitate the forceps with forces sensing ability because of the limited space for installing sensors. The craniocerebral tissues, interstitial fluids and currents passing though polar of forceps also bring restrictions on force sensing methods. In this article, we propose a planar force perception method to collect force information on bipolar forceps tips in real-time by installing two Fiber Bragg Grating Sensors (FBGs) perpendicularly on each polar of bipolar forceps. FBGs are free from electromagnetic interference. The slim configuration and silicon materials of FBGs make it a suitable sensing method for little disturbance on working state of bipolar forceps. In experiments, axial forces and planar forces were monitored in real-time. It is possible to sense forces at a minimum of 0.01 N. The calibrating revolution on x, y axis is 0.01, 0.03 N, separately.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"216 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124299728","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536121
Yangxin Xu, Keyu Li, Ziqi Zhao, M. Meng
Wireless capsule endoscopy is a painless tool for screening the entire gastrointestinal tract, and magnetic localization is a promising technology for tracking the capsule in the human intestine. In this paper, we propose a novel design approach of the optimal mobile sensor array for tracking the slowly moving capsule. A grid and combinatorial mathematics based method is presented to enumerate all feasible 3D layouts of the four-layer sensor array. The optimal layout with a valid localization range is selected from all layouts by localization accuracy evaluation. The valid localization range of the optimal layout is "confidence-box". Many simulation experiments on different trajectories with different noise levels show the feasibility of our proposed design approach and demonstrate the robustness and accuracy performance of the mobile sensor array based large-scale tracking. The moving capsule can be tracked with the accuracy of 1.9mm and 1.19° in position and orientation, respectively, and the valid localization depth reaches −12.88cm.
{"title":"A Design Approach of 3D Optimal Mobile Sensor Array for Confidence-box based Tracking of a Magnetic Capsule","authors":"Yangxin Xu, Keyu Li, Ziqi Zhao, M. Meng","doi":"10.1109/ICARM52023.2021.9536121","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536121","url":null,"abstract":"Wireless capsule endoscopy is a painless tool for screening the entire gastrointestinal tract, and magnetic localization is a promising technology for tracking the capsule in the human intestine. In this paper, we propose a novel design approach of the optimal mobile sensor array for tracking the slowly moving capsule. A grid and combinatorial mathematics based method is presented to enumerate all feasible 3D layouts of the four-layer sensor array. The optimal layout with a valid localization range is selected from all layouts by localization accuracy evaluation. The valid localization range of the optimal layout is \"confidence-box\". Many simulation experiments on different trajectories with different noise levels show the feasibility of our proposed design approach and demonstrate the robustness and accuracy performance of the mobile sensor array based large-scale tracking. The moving capsule can be tracked with the accuracy of 1.9mm and 1.19° in position and orientation, respectively, and the valid localization depth reaches −12.88cm.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124508509","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536128
Ruoxing Wang, Zhihua Chen, Kang Xu, Shou-kun Wang, Junzheng Wang, Bin Li
The stable operation with passing obstacles is the main challenge for wheel-legged robots in irregular terrains, especially in the gully terrain. An obstacle-surmounting method of wheel-legged hybrid gait based on perception system is proposed in this paper for gully terrain. Firstly, the system composition and communication mode of the wheel-legged robot are described. Secondly, the kinematics and dynamics models are established, and on this basis, the strategy of wheel-legged compound gait crossing gully landform is designed. Finally, the co-simulation based on MATLAB and V-REP is carried out. The simulation results show that the body of robot can maintain horizontal stability in the process of crossing gully terrain, which verifies the effectiveness of the obstacle-surmounting strategy.
{"title":"Hybrid Obstacle-Surmounting Gait for Hexapod Wheel-Legged Robot in Special Terrain","authors":"Ruoxing Wang, Zhihua Chen, Kang Xu, Shou-kun Wang, Junzheng Wang, Bin Li","doi":"10.1109/ICARM52023.2021.9536128","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536128","url":null,"abstract":"The stable operation with passing obstacles is the main challenge for wheel-legged robots in irregular terrains, especially in the gully terrain. An obstacle-surmounting method of wheel-legged hybrid gait based on perception system is proposed in this paper for gully terrain. Firstly, the system composition and communication mode of the wheel-legged robot are described. Secondly, the kinematics and dynamics models are established, and on this basis, the strategy of wheel-legged compound gait crossing gully landform is designed. Finally, the co-simulation based on MATLAB and V-REP is carried out. The simulation results show that the body of robot can maintain horizontal stability in the process of crossing gully terrain, which verifies the effectiveness of the obstacle-surmounting strategy.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117332885","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}
Pub Date : 2021-07-03DOI: 10.1109/ICARM52023.2021.9536160
Yu Chen, Xinde Li, S. Ge
Target positioning of a large inclination angle in aerial images is challenging for camera distortion, which makes it difficult to obtain a positioning model. Oblique images produce a larger positioning error if a traditional positioning algorithm is directly used to locate the target. To address this problem, this study uses a BP neural network to automatically calculate the high-accuracy positioning of the target. The location algorithm not only does not require camera calibration in advance but also eliminates the impact of camera distortion on target positioning. Through positioning experiments on the collected aerial dataset, the results demonstrate that the average positioning error of the target is about 1m, which has a high-precision positioning result and algorithm robustness.
{"title":"Research on the Algorithm of Target Location in Aerial Images under a Large Inclination Angle","authors":"Yu Chen, Xinde Li, S. Ge","doi":"10.1109/ICARM52023.2021.9536160","DOIUrl":"https://doi.org/10.1109/ICARM52023.2021.9536160","url":null,"abstract":"Target positioning of a large inclination angle in aerial images is challenging for camera distortion, which makes it difficult to obtain a positioning model. Oblique images produce a larger positioning error if a traditional positioning algorithm is directly used to locate the target. To address this problem, this study uses a BP neural network to automatically calculate the high-accuracy positioning of the target. The location algorithm not only does not require camera calibration in advance but also eliminates the impact of camera distortion on target positioning. Through positioning experiments on the collected aerial dataset, the results demonstrate that the average positioning error of the target is about 1m, which has a high-precision positioning result and algorithm robustness.","PeriodicalId":367307,"journal":{"name":"2021 6th IEEE International Conference on Advanced Robotics and Mechatronics (ICARM)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131308911","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：