Pub Date : 2018-07-01DOI: 10.23919/CHICC.2018.8482837
Xing Jin, Zhiwen Yao, Jingjing Zhang
Considering the difficulties in estimating depth from single image, in this paper, we propose a method to obtain the absolute scale depth map by combining the convolution neural network and depth filter. We compute relative transformation between consecutive frames by direct tracking features, which are extracted from RGB images and whose depthes are predicted by deep network, and then optimize relative motion by searching for a better feature alignment in epipolar line, and finally update every pixel depth of the reference frame by depth filter. We evaluate the proposed method on the open dataset comparison against the state of the art in depth estimation to evaluate our method.
{"title":"Dense Depth Estimation with Absolute Scale","authors":"Xing Jin, Zhiwen Yao, Jingjing Zhang","doi":"10.23919/CHICC.2018.8482837","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8482837","url":null,"abstract":"Considering the difficulties in estimating depth from single image, in this paper, we propose a method to obtain the absolute scale depth map by combining the convolution neural network and depth filter. We compute relative transformation between consecutive frames by direct tracking features, which are extracted from RGB images and whose depthes are predicted by deep network, and then optimize relative motion by searching for a better feature alignment in epipolar line, and finally update every pixel depth of the reference frame by depth filter. We evaluate the proposed method on the open dataset comparison against the state of the art in depth estimation to evaluate our method.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128755359","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8483296
Hao Xia, Zengle Li, Xiluo Yang
Alarm is an important method to detect the abnormal situations in modern industrial plants. Reducing the number of false alarms and missed alarms is significant for improving the performance of the alarm systems. In this paper, the optimal alarm threshold under time-varying operating condition has been studied. An alarm system model is introduced first, then the computational method of optimal alarm threshold without any data processing techniques is discussed. Moving average filter is then used to reduce the impact of measurement noise and its effect on the threshold design is further explained. An alarm design procedure based on these analysis is presented for the fast computation of alarm threshold. An example is provided to illustrate the effectiveness of the proposed method.
{"title":"Optimal alarm threshold under time-varying operating conditions","authors":"Hao Xia, Zengle Li, Xiluo Yang","doi":"10.23919/CHICC.2018.8483296","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8483296","url":null,"abstract":"Alarm is an important method to detect the abnormal situations in modern industrial plants. Reducing the number of false alarms and missed alarms is significant for improving the performance of the alarm systems. In this paper, the optimal alarm threshold under time-varying operating condition has been studied. An alarm system model is introduced first, then the computational method of optimal alarm threshold without any data processing techniques is discussed. Moving average filter is then used to reduce the impact of measurement noise and its effect on the threshold design is further explained. An alarm design procedure based on these analysis is presented for the fast computation of alarm threshold. An example is provided to illustrate the effectiveness of the proposed method.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128807510","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8483428
F. Wang, Baihai Zhang, S. Chai, Lingguo Cui, Fenxi Yao
The community structure is one of the most attractive properties of a complex network. This structure has been fundamental to advancements in various scientific branches. Numerous tools that involve community detection algorithms have been used in recent studies. In this paper, we propose a lightweight support vector clustering method. It surpasses traditional support vector approaches in terms of accuracy and complexity on account of its innovative design of distance calculations and the utilization of stable equilibrium points in the community assignment process. Extensive experiments are undertaken in computer-generated networks as well as real-world datasets. The results illustrate the competitive performance of the proposed algorithm compared to its community detection counterparts.
{"title":"Lightweight Support Vector Clustering Algorithm for Community Detection in Complex Networks","authors":"F. Wang, Baihai Zhang, S. Chai, Lingguo Cui, Fenxi Yao","doi":"10.23919/CHICC.2018.8483428","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8483428","url":null,"abstract":"The community structure is one of the most attractive properties of a complex network. This structure has been fundamental to advancements in various scientific branches. Numerous tools that involve community detection algorithms have been used in recent studies. In this paper, we propose a lightweight support vector clustering method. It surpasses traditional support vector approaches in terms of accuracy and complexity on account of its innovative design of distance calculations and the utilization of stable equilibrium points in the community assignment process. Extensive experiments are undertaken in computer-generated networks as well as real-world datasets. The results illustrate the competitive performance of the proposed algorithm compared to its community detection counterparts.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128575728","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8483276
Chu Wu, H. Fang, Xianlin Zeng
This paper addresses a distributed projected control algorithm for networked mobile manipulators to achieve a task of object transport with optimal manipulable coordination between robot arm and mobile platform. The task is designed as a distributed non-convex optimization problem with time-varying coupled equality constraints. To eliminate the time-varying term in the optimization problem, the task variables are first introduced. The non-convex cost function is then proved convex under a sufficient condition by selecting a proper displacement vector. Moreover, a modified Lagrangian function containing local multipliers and a nonsmooth penalty function is constructed to handle the coupled equality constraints in a distributed manner. Therefore, a fully distributed projected control algorithm is proposed to achieve the object transport task based on the primal-dual subgradient dynamics. We prove the convergence of the algorithm utilizing the Lyapunov stability theory and the invariance principle. Under the proposed control algorithm, the object transport is accomplished with optimal manipulable coordination, which is further validated through numerical simulations.
{"title":"Distributed Object Transport of Mobile Manipulators with Optimal Manipulable Coordination","authors":"Chu Wu, H. Fang, Xianlin Zeng","doi":"10.23919/CHICC.2018.8483276","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8483276","url":null,"abstract":"This paper addresses a distributed projected control algorithm for networked mobile manipulators to achieve a task of object transport with optimal manipulable coordination between robot arm and mobile platform. The task is designed as a distributed non-convex optimization problem with time-varying coupled equality constraints. To eliminate the time-varying term in the optimization problem, the task variables are first introduced. The non-convex cost function is then proved convex under a sufficient condition by selecting a proper displacement vector. Moreover, a modified Lagrangian function containing local multipliers and a nonsmooth penalty function is constructed to handle the coupled equality constraints in a distributed manner. Therefore, a fully distributed projected control algorithm is proposed to achieve the object transport task based on the primal-dual subgradient dynamics. We prove the convergence of the algorithm utilizing the Lyapunov stability theory and the invariance principle. Under the proposed control algorithm, the object transport is accomplished with optimal manipulable coordination, which is further validated through numerical simulations.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128585786","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8482855
Q. Jin, Chen Wang, Hehe Wang, Wu Cai, Yaxu Niu
In this paper, we study the problem of MIMO Hammerstein systems identification under heavy-tailed noises. As far as we know, there is no effective method to solve this problem. Inspired by this, we firstly introduced fuzzy logic and the nonlinear stochastic search (NLJ) algorithm to modify cuckoo search algorithm (CS) and proposed a novel CS algorithm (HFCS). According to Taylor expansion formula, the nonlinear block of the Hammerstein model is approximated by a class of polynomial family. Then, HFCS is used to estimate parameters of the model. The simulation results verify the efficiency of the proposed method.
{"title":"Hybrid Fuzzy Cuckoo Search Algorithm for MIMO Hammerstein Model Identification Under Heavy-Tailed Noises","authors":"Q. Jin, Chen Wang, Hehe Wang, Wu Cai, Yaxu Niu","doi":"10.23919/CHICC.2018.8482855","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8482855","url":null,"abstract":"In this paper, we study the problem of MIMO Hammerstein systems identification under heavy-tailed noises. As far as we know, there is no effective method to solve this problem. Inspired by this, we firstly introduced fuzzy logic and the nonlinear stochastic search (NLJ) algorithm to modify cuckoo search algorithm (CS) and proposed a novel CS algorithm (HFCS). According to Taylor expansion formula, the nonlinear block of the Hammerstein model is approximated by a class of polynomial family. Then, HFCS is used to estimate parameters of the model. The simulation results verify the efficiency of the proposed method.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129332905","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8483963
Yahui Li, Jun Liu, Li-li Wang
Deep learning is a field that has attracted a great concern in recent years, and plays an important role in computer vision. Traditional object detection methods failed to adapt to the increasingly complex application environment. While, deep learning, because of the powerful feature extraction capabilities, shows strong ability in object detection tasks in recent years. However, intensive and complex calculations of the deep network are very demanding for the hardware, which makes it will be difficult to deploy on the common hardware devices. In this case, lightweight network technology comes into being. Firstly, this paper analyzes the limitations of deep learning and the necessity of lightweight network technology. Then, According to the existing technology, the methods of lightweight network are summarized and analyzed. In addition, lightweight network methods are compared and analyzed, and the advantages and disadvantages of these methods are pointed out. Finally, we summarize the problems to be faced by the lightweight network approach and the direction of deep learning technology development.
{"title":"Lightweight Network Research Based on Deep Learning: A Review","authors":"Yahui Li, Jun Liu, Li-li Wang","doi":"10.23919/CHICC.2018.8483963","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8483963","url":null,"abstract":"Deep learning is a field that has attracted a great concern in recent years, and plays an important role in computer vision. Traditional object detection methods failed to adapt to the increasingly complex application environment. While, deep learning, because of the powerful feature extraction capabilities, shows strong ability in object detection tasks in recent years. However, intensive and complex calculations of the deep network are very demanding for the hardware, which makes it will be difficult to deploy on the common hardware devices. In this case, lightweight network technology comes into being. Firstly, this paper analyzes the limitations of deep learning and the necessity of lightweight network technology. Then, According to the existing technology, the methods of lightweight network are summarized and analyzed. In addition, lightweight network methods are compared and analyzed, and the advantages and disadvantages of these methods are pointed out. Finally, we summarize the problems to be faced by the lightweight network approach and the direction of deep learning technology development.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124543938","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8482671
Meiliu Li, Guangjun Wang, Jinhua She, Zhentao Liu, Danyun Li
To meet the needs of high speed and good environmental adaptability for data transmission in industrial robot servo systems, this paper presents a wireless transmission technology on a data acquisition control terminal. The system is based on STM32F103RET6 processor and reads cache data from the SRAM of FPGA in DMA method, and communicates with a WIFI chip, Marvell 88w8686, through the SDIO interface. A host computer segments and filters the data received by the WIFI chip and sends control commands back to STM32. The functions of the hardware and software of the system contains data compression and storage, wireless data transmission, data display, and control-command transmission. An experimental platform has been built. It carries out real-time transmission of 8–16 digits based on the TCP/IP protocol with expected performance.
{"title":"Control of Wireless Network Communication for Industrial Robot Servo Systems","authors":"Meiliu Li, Guangjun Wang, Jinhua She, Zhentao Liu, Danyun Li","doi":"10.23919/CHICC.2018.8482671","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8482671","url":null,"abstract":"To meet the needs of high speed and good environmental adaptability for data transmission in industrial robot servo systems, this paper presents a wireless transmission technology on a data acquisition control terminal. The system is based on STM32F103RET6 processor and reads cache data from the SRAM of FPGA in DMA method, and communicates with a WIFI chip, Marvell 88w8686, through the SDIO interface. A host computer segments and filters the data received by the WIFI chip and sends control commands back to STM32. The functions of the hardware and software of the system contains data compression and storage, wireless data transmission, data display, and control-command transmission. An experimental platform has been built. It carries out real-time transmission of 8–16 digits based on the TCP/IP protocol with expected performance.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124616611","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8482897
Shuaijie Ouyang, Zhongjian Dai, C. Yan, Peng Wei
Aiming at the problem that it is difficult to quantitatively evaluate the combat effectiveness of the C4ISR system, this paper proposes a method to evaluate the combat effectiveness of the marine C4ISR system based on system dynamics. By analyzing the feedback mechanism of the maritime C4ISR system and the causal relationship between the red and blue maritime C4ISR systems under complex electronic warfare conditions, a system dynamics model and related equations of the maritime C4ISR system in counter condition are established. The degree of the damage from Red to Blue is used as a measure of the operational effectiveness of the maritime C4ISR system. The simulation results show that this method is helpful to improve the combat plan and improve the combat effectiveness of the C4ISR system.
{"title":"Operational Effectiveness Evaluation of Maritime C4ISR System Based on System Dynamics","authors":"Shuaijie Ouyang, Zhongjian Dai, C. Yan, Peng Wei","doi":"10.23919/CHICC.2018.8482897","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8482897","url":null,"abstract":"Aiming at the problem that it is difficult to quantitatively evaluate the combat effectiveness of the C4ISR system, this paper proposes a method to evaluate the combat effectiveness of the marine C4ISR system based on system dynamics. By analyzing the feedback mechanism of the maritime C4ISR system and the causal relationship between the red and blue maritime C4ISR systems under complex electronic warfare conditions, a system dynamics model and related equations of the maritime C4ISR system in counter condition are established. The degree of the damage from Red to Blue is used as a measure of the operational effectiveness of the maritime C4ISR system. The simulation results show that this method is helpful to improve the combat plan and improve the combat effectiveness of the C4ISR system.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124639582","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8484015
Yongyue Zhang, Yali Jin, Weihua Cao, Zezhong Li, Yan Yuan
In the continuous annealing line heating process, it is hard to get an accurate predict result only by a steady state model as it is a complex, strongly time-delayed and confounding process. This study provides a method for building a dynamic model. First analyzes the mechanism of the annealing line to get the main parameters, and then use the data-driven modeling method to get a steady state model, finally combines with dynamic algorithm to establish a dynamic model. This modeling method improves the accuracy of predict result to guarantee the efficiency of enterprises.
{"title":"A Dynamic Data-driven Model for Predicting Strip Temperature in Continuous Annealing Line Heating Process","authors":"Yongyue Zhang, Yali Jin, Weihua Cao, Zezhong Li, Yan Yuan","doi":"10.23919/CHICC.2018.8484015","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8484015","url":null,"abstract":"In the continuous annealing line heating process, it is hard to get an accurate predict result only by a steady state model as it is a complex, strongly time-delayed and confounding process. This study provides a method for building a dynamic model. First analyzes the mechanism of the annealing line to get the main parameters, and then use the data-driven modeling method to get a steady state model, finally combines with dynamic algorithm to establish a dynamic model. This modeling method improves the accuracy of predict result to guarantee the efficiency of enterprises.","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"186 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124750244","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 : 2018-07-01DOI: 10.23919/CHICC.2018.8483051
Hongwei Wang, Xia Hao, Jie Lian
This paper is motivated by the practical control considerations that non-uniformly sampled nonlinear systems are abundant in industrial process. The corresponding input-output relationship of non-uniformly sampled nonlinear systems is obtained by using the weighted combination of the multiple local lifted linear models acquired from non-uniformly sampled measurements. Further, fuzzy model is derived by constructing the fuzzy membership degree functions as the weighted combination representation. On this basis, we propose a fuzzy identification algorithm using GK fuzzy clustering and recursive least squared method. Finally, the simulation example is studied to demonstrate the effectiveness of the proposed method..
{"title":"Fuzzy Identification of Non-uniformly Sampled Data Nonlinear Systems Based on Clustering Method","authors":"Hongwei Wang, Xia Hao, Jie Lian","doi":"10.23919/CHICC.2018.8483051","DOIUrl":"https://doi.org/10.23919/CHICC.2018.8483051","url":null,"abstract":"This paper is motivated by the practical control considerations that non-uniformly sampled nonlinear systems are abundant in industrial process. The corresponding input-output relationship of non-uniformly sampled nonlinear systems is obtained by using the weighted combination of the multiple local lifted linear models acquired from non-uniformly sampled measurements. Further, fuzzy model is derived by constructing the fuzzy membership degree functions as the weighted combination representation. On this basis, we propose a fuzzy identification algorithm using GK fuzzy clustering and recursive least squared method. Finally, the simulation example is studied to demonstrate the effectiveness of the proposed method..","PeriodicalId":158442,"journal":{"name":"2018 37th Chinese Control Conference (CCC)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129907204","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：