Pub Date : 2022-07-17DOI: 10.1109/RCAR54675.2022.9872220
Chunling Yang, Dong Qiu
In recent years, pedestrian detection based on image recognition has become an important research topic in vehicle assisted driving. For the question of poor detection accuracy resulted from missing detection and small targets in pedestrian detection, proposes a pedestrian detection method based on improved Faster-RCNN. First, ResNet34 residual network was used to replace VGG-16 as the backbone feature extraction network, and then SENet mechanism was introduced to further enhance and suppress the weight vector. Then, aiming at the multi-scale problem in the detection set, FPN network is added to further strengthen the feature extraction ability of the network. The k-means algorithm is introduced to generate appropriate anchors according to the characteristics of the dataset. The experimental results show that, compared with the classic network, the average precision (mAP) of the improved algorithm reaches 93.36%, which is 5.34% higher than the original Faster-RCNN algorithm, which proves the effectiveness of the algorithm.
{"title":"Pedestrian Detection Based on Improved Faster-RCNN Algorithm","authors":"Chunling Yang, Dong Qiu","doi":"10.1109/RCAR54675.2022.9872220","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872220","url":null,"abstract":"In recent years, pedestrian detection based on image recognition has become an important research topic in vehicle assisted driving. For the question of poor detection accuracy resulted from missing detection and small targets in pedestrian detection, proposes a pedestrian detection method based on improved Faster-RCNN. First, ResNet34 residual network was used to replace VGG-16 as the backbone feature extraction network, and then SENet mechanism was introduced to further enhance and suppress the weight vector. Then, aiming at the multi-scale problem in the detection set, FPN network is added to further strengthen the feature extraction ability of the network. The k-means algorithm is introduced to generate appropriate anchors according to the characteristics of the dataset. The experimental results show that, compared with the classic network, the average precision (mAP) of the improved algorithm reaches 93.36%, which is 5.34% higher than the original Faster-RCNN algorithm, which proves the effectiveness of the algorithm.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131149524","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 : 2022-07-17DOI: 10.1109/RCAR54675.2022.9872272
Yongfeng Cao, Fan Feng, Zefeng Liu, Le Xie
Continuum robots (CRs) have been developed for maxillary sinus surgery (MSS) in recent years. However, due to the curved and narrow pathway of the maxillary sinus, and the deformability of the CR, accurately approaching the target in the sinus is still a challenge. In this paper, a CR integrated with essential instruments and sensors is developed for the MSS. To improve the maneuverability of the CR during the surgery, a master-slave motion control algorithm is proposed based on the kinematic model. Two types of commonly used master devices, joystick and sidestick, are compared in the MSS. Comparative experiments are performed to verify the feasibility of the proposed scheme.
{"title":"Design and Validation of a Master-slave Continuum Robot for Maxillary Sinus Surgery","authors":"Yongfeng Cao, Fan Feng, Zefeng Liu, Le Xie","doi":"10.1109/RCAR54675.2022.9872272","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872272","url":null,"abstract":"Continuum robots (CRs) have been developed for maxillary sinus surgery (MSS) in recent years. However, due to the curved and narrow pathway of the maxillary sinus, and the deformability of the CR, accurately approaching the target in the sinus is still a challenge. In this paper, a CR integrated with essential instruments and sensors is developed for the MSS. To improve the maneuverability of the CR during the surgery, a master-slave motion control algorithm is proposed based on the kinematic model. Two types of commonly used master devices, joystick and sidestick, are compared in the MSS. Comparative experiments are performed to verify the feasibility of the proposed scheme.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131914528","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 Safety of the Intended Functionality (SOTIF) mainly solves the safety problems induced by external scenarios. However, in known international standards and practices with respect to SOTIF, there is no method to identify the triggering conditions that is agreed by most researchers because the triggering conditions come from scenarios, a relatively disordered system. Taking ACC system as an example, this paper, from the perspective of scenario elements and classification, puts forward a set of analysis methods to systematically and effectively identify SOTIF triggering conditions upon reasonable analysis of functional insufficiency.
{"title":"The Research on the Identification of ACC SOTIF Triggering Conditions Based on Scenario Analysis","authors":"Qidong Zhao, Zheng Tong, Yunshuang Zhang, Chen Chao, Qingyu Zhang, Shuai Zhao, Zhibin Du","doi":"10.1109/rcar54675.2022.9872207","DOIUrl":"https://doi.org/10.1109/rcar54675.2022.9872207","url":null,"abstract":"The Safety of the Intended Functionality (SOTIF) mainly solves the safety problems induced by external scenarios. However, in known international standards and practices with respect to SOTIF, there is no method to identify the triggering conditions that is agreed by most researchers because the triggering conditions come from scenarios, a relatively disordered system. Taking ACC system as an example, this paper, from the perspective of scenario elements and classification, puts forward a set of analysis methods to systematically and effectively identify SOTIF triggering conditions upon reasonable analysis of functional insufficiency.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129707545","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}
Generative adversarial networks are widely used in computer vision tasks like image translation and image style transfer. Most of mainstream methods including CycleGAN and pix2pix use the stacking of residual blocks to deepen the number of network layers, which makes the networks have a large number of parameters and floating point operations. This paper presents a ghost-module-based generative adversarial networks. We use the ghost module to replace the residual blocks in the traditional generative adversarial network for building lightweight generative adversarial networks. Experiments shows that our method significantly reducing the parameters and floating point operations of the generative adversarial network on the precondition of assuring the quality of the generated images.
{"title":"Lightweight Generative Adversarial Networks Based on Ghost Module","authors":"Xinyuan Xiang, Meiqin Liu, Senlin Zhang, Ping Wei, Badong Chen","doi":"10.1109/RCAR54675.2022.9872153","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872153","url":null,"abstract":"Generative adversarial networks are widely used in computer vision tasks like image translation and image style transfer. Most of mainstream methods including CycleGAN and pix2pix use the stacking of residual blocks to deepen the number of network layers, which makes the networks have a large number of parameters and floating point operations. This paper presents a ghost-module-based generative adversarial networks. We use the ghost module to replace the residual blocks in the traditional generative adversarial network for building lightweight generative adversarial networks. Experiments shows that our method significantly reducing the parameters and floating point operations of the generative adversarial network on the precondition of assuring the quality of the generated images.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125063953","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 : 2022-07-17DOI: 10.1109/RCAR54675.2022.9872285
Zhang Meng, Daoxiong Gong
Accidental falls often cause serious harm to the human body, especially for the elderly. But falls tend to be infrequent, making it difficult to collect large amounts of data for research. In reality, there is a large gap between the amount of sensor data collected by falling activities and daily activities, which will lead to class imbalance. When using machine learning algorithms to detect falls, class imbalance will cause the performance of the classifier to be biased towards most classes and reduce the detection accuracy of a few classes. When faced with the problem of binary class imbalance, selecting an effective machine learning algorithm and resampling data can effectively improve the accuracy of classification. In this paper, an ensemble learning algorithm and clustering undersampling method are used for fall detection. The ensemble learning algorithm can reduce the impact of imbalanced datasets on the training model through multiple classifier iterations. Clustering undersampling method can change the dataset distribution and balance the number of positive and negative samples. The method in this paper is evaluated on the public dataset Sisfall. Compared with the traditional machine learning algorithms, the ensemble learning has higher accuracy and faster training speed. Combined with the clustering undersampling method, the method has a higher recall and precision.
{"title":"Combining Clustering Undersample and Ensemble Learning for Wearable Fall Detection","authors":"Zhang Meng, Daoxiong Gong","doi":"10.1109/RCAR54675.2022.9872285","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872285","url":null,"abstract":"Accidental falls often cause serious harm to the human body, especially for the elderly. But falls tend to be infrequent, making it difficult to collect large amounts of data for research. In reality, there is a large gap between the amount of sensor data collected by falling activities and daily activities, which will lead to class imbalance. When using machine learning algorithms to detect falls, class imbalance will cause the performance of the classifier to be biased towards most classes and reduce the detection accuracy of a few classes. When faced with the problem of binary class imbalance, selecting an effective machine learning algorithm and resampling data can effectively improve the accuracy of classification. In this paper, an ensemble learning algorithm and clustering undersampling method are used for fall detection. The ensemble learning algorithm can reduce the impact of imbalanced datasets on the training model through multiple classifier iterations. Clustering undersampling method can change the dataset distribution and balance the number of positive and negative samples. The method in this paper is evaluated on the public dataset Sisfall. Compared with the traditional machine learning algorithms, the ensemble learning has higher accuracy and faster training speed. Combined with the clustering undersampling method, the method has a higher recall and precision.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130085562","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 : 2022-07-17DOI: 10.1109/RCAR54675.2022.9872225
Hongbo Liu, Ping Song, Youtian Qie, Yifan Li
Tiny Machine Learning (TinyML) is a new research area aimed at designing and developing machine learning (ML) techniques for embedded systems and IoT units. Due to the limited resources of embedded system, neural network pruning is widely used to reduce resource occupation. To solve the problem that the Remaining Useful Life (RUL) of the equipment is difficult to calculate accurately and in real time, a pruning method based on L1 norm weight was designed to reduce the memory footprint and computational load of the neural network, and a lightweight two-dimensional convolutional neural network was constructed. Experimental results show that compared with random pruning, this method greatly reduces the influence of neural network parameter reduction on the accuracy of inference results. Meanwhile, a retraining method based on Adam optimization was used to make the RUL curve predicted by the retrained model more close to the real RUL curve. When the weight parameters are reduced by 30%, the model still maintains good prediction accuracy, and can realize the real-time prediction of RUL in the embedded system with limited resources.
{"title":"Real-time Prediction Method of Remaining Useful Life Based on TinyML","authors":"Hongbo Liu, Ping Song, Youtian Qie, Yifan Li","doi":"10.1109/RCAR54675.2022.9872225","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872225","url":null,"abstract":"Tiny Machine Learning (TinyML) is a new research area aimed at designing and developing machine learning (ML) techniques for embedded systems and IoT units. Due to the limited resources of embedded system, neural network pruning is widely used to reduce resource occupation. To solve the problem that the Remaining Useful Life (RUL) of the equipment is difficult to calculate accurately and in real time, a pruning method based on L1 norm weight was designed to reduce the memory footprint and computational load of the neural network, and a lightweight two-dimensional convolutional neural network was constructed. Experimental results show that compared with random pruning, this method greatly reduces the influence of neural network parameter reduction on the accuracy of inference results. Meanwhile, a retraining method based on Adam optimization was used to make the RUL curve predicted by the retrained model more close to the real RUL curve. When the weight parameters are reduced by 30%, the model still maintains good prediction accuracy, and can realize the real-time prediction of RUL in the embedded system with limited resources.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123952000","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 : 2022-07-17DOI: 10.1109/RCAR54675.2022.9872188
Ruliang Feng, Huiren Tao, Canhua Ye, Guanglin Li, Xueling Bai, Lin Wang
The scoliosis with a prevalence of about 2.4% has become the top 3 major “killer” of children and adolescents’ health. Only 0.02% of scoliosis would be severe enough to require surgical intervention, while the rest was treated with orthotics or exercise training. However, the location of the orthotic forces and its effects were still not clear, resulting in a great deal of blindness in the fabrication of precise individualized orthoses and the later applied orthopedic forces. In this paper, we built a 3D spine model of a patient with idiopathic scoliosis based on CT tomography data, applied different orthopedic forces to the spine model to compare the results and clarify the relationship between them in order to determine the optimal location and magnitude of the orthopedic force, which were necessary for precise interventions in patients. The present results showed that 1) the greater the applied force, the better the correction effect (within reasonable limits) and 2) the effect of multiple forces applied for correction was better than that of a single force applied, as reflected by a greater displacement of the vertebrae and almost identical mean Von Mises stress in the discs, which could support the production of effective personalized orthopedic robots.
{"title":"A quantitative biomechanical study for precise orthopedic intervention in idiopathic scoliosis","authors":"Ruliang Feng, Huiren Tao, Canhua Ye, Guanglin Li, Xueling Bai, Lin Wang","doi":"10.1109/RCAR54675.2022.9872188","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872188","url":null,"abstract":"The scoliosis with a prevalence of about 2.4% has become the top 3 major “killer” of children and adolescents’ health. Only 0.02% of scoliosis would be severe enough to require surgical intervention, while the rest was treated with orthotics or exercise training. However, the location of the orthotic forces and its effects were still not clear, resulting in a great deal of blindness in the fabrication of precise individualized orthoses and the later applied orthopedic forces. In this paper, we built a 3D spine model of a patient with idiopathic scoliosis based on CT tomography data, applied different orthopedic forces to the spine model to compare the results and clarify the relationship between them in order to determine the optimal location and magnitude of the orthopedic force, which were necessary for precise interventions in patients. The present results showed that 1) the greater the applied force, the better the correction effect (within reasonable limits) and 2) the effect of multiple forces applied for correction was better than that of a single force applied, as reflected by a greater displacement of the vertebrae and almost identical mean Von Mises stress in the discs, which could support the production of effective personalized orthopedic robots.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128487523","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 : 2022-07-17DOI: 10.1109/RCAR54675.2022.9872281
Yue Xu, Jie Xie, Zhiwei Xing, Wenqiang Yuan, Guanqun Yu, Z. Zeng, Baoshun Zhang, Dongmin Wu
The method of implementing TDC with FPGA carry chain is widely used, but the delay time of each TDC bin is greatly affected by the changes of operating temperature. At present, the commonly used methods can’t well fit the changing trend of each delay bin in long delay line under the influence of complex temperature changes. In this paper, a neural network calibration module based on MLP is proposed, in which 128 delay time data of delay line and corresponding temperature data transmitted to the host computer are used as training samples to establish MLP. When working, the delay time of each TDC bin can be given independently by knowing current temperature condition. Through experiments, the compensation of network calibration module on temperature changes is verified, and the network can be transplanted to different types of FPGA chips and run under various temperature changes. The TDC have a precision of 34ps.
{"title":"A Bin-by-Bin Calibration with Neural Network for FPGA-Based Tapped-Delay-Line Time-to-Digital Converter","authors":"Yue Xu, Jie Xie, Zhiwei Xing, Wenqiang Yuan, Guanqun Yu, Z. Zeng, Baoshun Zhang, Dongmin Wu","doi":"10.1109/RCAR54675.2022.9872281","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872281","url":null,"abstract":"The method of implementing TDC with FPGA carry chain is widely used, but the delay time of each TDC bin is greatly affected by the changes of operating temperature. At present, the commonly used methods can’t well fit the changing trend of each delay bin in long delay line under the influence of complex temperature changes. In this paper, a neural network calibration module based on MLP is proposed, in which 128 delay time data of delay line and corresponding temperature data transmitted to the host computer are used as training samples to establish MLP. When working, the delay time of each TDC bin can be given independently by knowing current temperature condition. Through experiments, the compensation of network calibration module on temperature changes is verified, and the network can be transplanted to different types of FPGA chips and run under various temperature changes. The TDC have a precision of 34ps.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129053222","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 : 2022-07-17DOI: 10.1109/RCAR54675.2022.9872177
Gang Li, Xin Ma, Zhi Li, Yibin Li
Rotary cranes usually transfer the heavy mass and large payloads with distributed mass beam in practice. As a typical underactuated system, the dynamic model is very complicated and challenging due to the actuated boom luffing motion and underactuated distributed mass beam payload swing vibration. To simplify the dynamic model, the existing control method regards the payload as the mass point, which ignores the geometry of the payload. This paper proposes an energy shaping based nonlinear anti-swing controller for double-pendulum rotary crane with distributed-mass beams. The rotary crane dynamic model is established based on the Lagrange’s method. After analysis the energy function, a nonlinear anti-swing controller is designed for the rotary crane. It successfully solves the boom positioning and payload anti-swing problems with a distributed mass beam payload. LaSalle’s invariance theorem and Lyapunov technology are helped to analysis the stability of the rotary crane control system. Finally, the proposed controller shows effective and robust control performance in the simulation verification.
{"title":"Energy Shaping Based Nonlinear Anti-Swing Controller for Double-Pendulum Rotary Crane with Distributed-Mass Beams","authors":"Gang Li, Xin Ma, Zhi Li, Yibin Li","doi":"10.1109/RCAR54675.2022.9872177","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872177","url":null,"abstract":"Rotary cranes usually transfer the heavy mass and large payloads with distributed mass beam in practice. As a typical underactuated system, the dynamic model is very complicated and challenging due to the actuated boom luffing motion and underactuated distributed mass beam payload swing vibration. To simplify the dynamic model, the existing control method regards the payload as the mass point, which ignores the geometry of the payload. This paper proposes an energy shaping based nonlinear anti-swing controller for double-pendulum rotary crane with distributed-mass beams. The rotary crane dynamic model is established based on the Lagrange’s method. After analysis the energy function, a nonlinear anti-swing controller is designed for the rotary crane. It successfully solves the boom positioning and payload anti-swing problems with a distributed mass beam payload. LaSalle’s invariance theorem and Lyapunov technology are helped to analysis the stability of the rotary crane control system. Finally, the proposed controller shows effective and robust control performance in the simulation verification.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114474303","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 distributed real-time system composed of industrial robots usually needs to meet the support of low-latency, high-reliability data transmission and service invocation. Based on the Data Distribution Service (DDS) publish-subscribe communication paradigm, this paper designs a service integration framework model called Dynamic-DDS-RPC with a request-response mechanism to achieve low-latency and flexible configuration of the robot service remote procedure call function. First of all, we use JSON to implement a set of service description specifications to support service definition and program interface implementation, and uses dynamic library automatic loading technology to solve the problem that DDS-RPC does not support dynamic loading and discovery of services. Secondly, we implement a topic-centric service request-response mechanism based on the publish-subscribe communication mode of DDS, and realize low-latency data communication. Finally, we develop a robot middleware software based on the framework, and use the middleware to compare the performance of the service-request response speed with the WebService-based middleware and the DDS-RPC-based middleware. The results show that the service integration framework proposed in this paper can realize the remote procedure call function with low latency and high flexible configuration, which verifies the effectiveness of the framework.
{"title":"Design and Implementation of Robot Middleware Service Integration Framework Based on DDS","authors":"Xiaowen Zhang, Xiaogang Zhang, Shaoyuan Wang, Xiao Ping","doi":"10.1109/RCAR54675.2022.9872212","DOIUrl":"https://doi.org/10.1109/RCAR54675.2022.9872212","url":null,"abstract":"The distributed real-time system composed of industrial robots usually needs to meet the support of low-latency, high-reliability data transmission and service invocation. Based on the Data Distribution Service (DDS) publish-subscribe communication paradigm, this paper designs a service integration framework model called Dynamic-DDS-RPC with a request-response mechanism to achieve low-latency and flexible configuration of the robot service remote procedure call function. First of all, we use JSON to implement a set of service description specifications to support service definition and program interface implementation, and uses dynamic library automatic loading technology to solve the problem that DDS-RPC does not support dynamic loading and discovery of services. Secondly, we implement a topic-centric service request-response mechanism based on the publish-subscribe communication mode of DDS, and realize low-latency data communication. Finally, we develop a robot middleware software based on the framework, and use the middleware to compare the performance of the service-request response speed with the WebService-based middleware and the DDS-RPC-based middleware. The results show that the service integration framework proposed in this paper can realize the remote procedure call function with low latency and high flexible configuration, which verifies the effectiveness of the framework.","PeriodicalId":304963,"journal":{"name":"2022 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114821709","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：