Pub Date : 2020-07-01DOI: 10.23919/CCC50068.2020.9188890
Feng Liu, Changgu Chen, Zhihua Li, Z. Guan, Hua O. Wang
In this paper, to avoid the problem of local optimization and slow convergence in complex environment, a reinforcement learning algorithm is proposed to solve the problem. A robot path planning model is built and its feasibility is verified by simulation. In addition, this paper proposes a deep environment to neural network for robot camera to establish a deep reinforcement learning path planning model, and establishes a deep recursive Q-network (DRQN) and Deep Dueling Q-network(DDQN) respectively. In the comparison of the final simulation results, DRQN needs to consume more computation time, but can achieve better results with higher accuracy.
{"title":"Research on path planning of robot based on deep reinforcement learning","authors":"Feng Liu, Changgu Chen, Zhihua Li, Z. Guan, Hua O. Wang","doi":"10.23919/CCC50068.2020.9188890","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9188890","url":null,"abstract":"In this paper, to avoid the problem of local optimization and slow convergence in complex environment, a reinforcement learning algorithm is proposed to solve the problem. A robot path planning model is built and its feasibility is verified by simulation. In addition, this paper proposes a deep environment to neural network for robot camera to establish a deep reinforcement learning path planning model, and establishes a deep recursive Q-network (DRQN) and Deep Dueling Q-network(DDQN) respectively. In the comparison of the final simulation results, DRQN needs to consume more computation time, but can achieve better results with higher accuracy.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"37 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122827447","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 : 2020-07-01DOI: 10.23919/CCC50068.2020.9188913
Huang Wenye, Zhang Yumin, Sheng Wei, W. Xiaogang, Liu Lipeng
Aiming at the engineering environment that requires a wide range of laying, long working hours, high accuracy and reliability, a sensor network framework model was proposed, which can effectively acquire the collected information and meet the real-time data transmission function. At the same time, a data processing and fusion scheme was proposed to remove trend and gross error terms from the static collected data. The trend test and ADF test data are used to meet the stability requirements. Then a multi-redundancy data fusion mechanism is applied for data fusion of preprocessed data. Finally, via the Allan variance method, the random noise characteristics of the gyro was analyzed, and the data quality and fusion effect are evaluated. Experimental results show that the designed sensor network can effectively collect real-time information, and the data processing and fusion working mechanism can effectively reduce noise and improve data quality.
{"title":"Data Processing and Fusion Working Mechanism Scheme of MIMU Sensor Network","authors":"Huang Wenye, Zhang Yumin, Sheng Wei, W. Xiaogang, Liu Lipeng","doi":"10.23919/CCC50068.2020.9188913","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9188913","url":null,"abstract":"Aiming at the engineering environment that requires a wide range of laying, long working hours, high accuracy and reliability, a sensor network framework model was proposed, which can effectively acquire the collected information and meet the real-time data transmission function. At the same time, a data processing and fusion scheme was proposed to remove trend and gross error terms from the static collected data. The trend test and ADF test data are used to meet the stability requirements. Then a multi-redundancy data fusion mechanism is applied for data fusion of preprocessed data. Finally, via the Allan variance method, the random noise characteristics of the gyro was analyzed, and the data quality and fusion effect are evaluated. Experimental results show that the designed sensor network can effectively collect real-time information, and the data processing and fusion working mechanism can effectively reduce noise and improve data quality.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122911577","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 permutation flowshop scheduling problem with preventive maintenance operator (PM/DPFSP) is closely related to modem industry. This paper presents an improved discrete artificial bee colony (IDABC) algorithm for solving this problem. The criterion to be optimized is the makespan. An improved NEH heuristic method is proposed to initialize the population effectively. We adapted a local search method with insertion and swap operator to produce neighboring solutions in employ bee phase and onlooker bee phase. A global search method with destruction and reconstruction phase is introduced to avoid local optima in scout bee phase. The parameters for the proposed IDABC are calibrated by means of a design of experiments and analysis of variance. We conduct extensive experiments to test the performance of IDABC. Computational results indicate that IDABC has promising advantages on PM/DPFSP.
{"title":"An improved discrete artificial bee colony algorithm for the distributed permutation flowshop scheduling problem with preventive maintenance","authors":"Jia-Yang Mao, Xiaolu Hu, Q. Pan, Zhonghua Miao, Chuangxin He, M. Tasgetiren","doi":"10.23919/CCC50068.2020.9188697","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9188697","url":null,"abstract":"The distributed permutation flowshop scheduling problem with preventive maintenance operator (PM/DPFSP) is closely related to modem industry. This paper presents an improved discrete artificial bee colony (IDABC) algorithm for solving this problem. The criterion to be optimized is the makespan. An improved NEH heuristic method is proposed to initialize the population effectively. We adapted a local search method with insertion and swap operator to produce neighboring solutions in employ bee phase and onlooker bee phase. A global search method with destruction and reconstruction phase is introduced to avoid local optima in scout bee phase. The parameters for the proposed IDABC are calibrated by means of a design of experiments and analysis of variance. We conduct extensive experiments to test the performance of IDABC. Computational results indicate that IDABC has promising advantages on PM/DPFSP.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122963852","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 : 2020-07-01DOI: 10.23919/CCC50068.2020.9189303
Wenhao Jia, Jinzhi Wang
In this paper, we investigate distributed fault detection (FD) problem for a group of heterogeneous agents with internal model controllers. In order to avoid utilizing solutions of complicated matrix equations in the fault detection (FD) module, we design a distributed internal model control law with a state estimator additionally, and design observers for the MAS closed-loop system formed by the designed control law to generate residuals to detect the faulty agent, where the observer feedback information just contains itself and its neighbors’ state and state estimate information, then the utilization of solutions of matrix equations is avoided. Finally, a numerical example is utilized to verify effectiveness of the proposed distributed FD scheme, where the faulty agent will be detected in time, and the designed control law will realize cooperative output regulation (COR) if no fault occurs.
{"title":"Distributed Fault Detection for A Class of Heterogeneous Agents with Internal Model Controllers","authors":"Wenhao Jia, Jinzhi Wang","doi":"10.23919/CCC50068.2020.9189303","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9189303","url":null,"abstract":"In this paper, we investigate distributed fault detection (FD) problem for a group of heterogeneous agents with internal model controllers. In order to avoid utilizing solutions of complicated matrix equations in the fault detection (FD) module, we design a distributed internal model control law with a state estimator additionally, and design observers for the MAS closed-loop system formed by the designed control law to generate residuals to detect the faulty agent, where the observer feedback information just contains itself and its neighbors’ state and state estimate information, then the utilization of solutions of matrix equations is avoided. Finally, a numerical example is utilized to verify effectiveness of the proposed distributed FD scheme, where the faulty agent will be detected in time, and the designed control law will realize cooperative output regulation (COR) if no fault occurs.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114488374","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 : 2020-07-01DOI: 10.23919/CCC50068.2020.9189373
Haitao Ma, Shihong Yue, Jia Li
In the existing environmental air quality evaluation index system, the AQI depends on the concentration of the dominant pollutant in the six types of air pollutants. However, the interaction among different pollutants and the effect of the combination of pollutants on AQI is rarely studied. In this paper, concentrations of six types of air pollutants and corresponding AQI of Tianjin in 2018 were taken as samples, and trained by Choquet integral. The Shapley interaction index of six types of pollutants are obtained by nonadditive measure. Based on the Shapley interaction index, the weights of six air pollutants on AQI and the interaction among different pollutants were analyzed. The results show that the main air pollutants in Tianjin in 2018 are PM2.5 and O3, and there is a negative interaction between them. Finally, the trained model is used to recalculate AQI of Tianjin in the first three months of 2019, providing a reference for air quality evaluation.
{"title":"Air Quality Evaluation Method Based on Data Analysis","authors":"Haitao Ma, Shihong Yue, Jia Li","doi":"10.23919/CCC50068.2020.9189373","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9189373","url":null,"abstract":"In the existing environmental air quality evaluation index system, the AQI depends on the concentration of the dominant pollutant in the six types of air pollutants. However, the interaction among different pollutants and the effect of the combination of pollutants on AQI is rarely studied. In this paper, concentrations of six types of air pollutants and corresponding AQI of Tianjin in 2018 were taken as samples, and trained by Choquet integral. The Shapley interaction index of six types of pollutants are obtained by nonadditive measure. Based on the Shapley interaction index, the weights of six air pollutants on AQI and the interaction among different pollutants were analyzed. The results show that the main air pollutants in Tianjin in 2018 are PM2.5 and O3, and there is a negative interaction between them. Finally, the trained model is used to recalculate AQI of Tianjin in the first three months of 2019, providing a reference for air quality evaluation.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121880838","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 : 2020-07-01DOI: 10.23919/CCC50068.2020.9189352
Zhi-Shuang Xue, Wenjie Chen, Jing Li
In recent years, deep convolutional neural networks have made breakthrough progress in object recognition and object detection tasks in the field of computer vision, and have achieved great results both in accuracy and speed. However, the detection of small objects is still difficult in the field of object detection, and the accuracy on the common dataset MS COCO is very low. This paper briefly reviews some work in multi-scale object detection algorithms, and then proposes a method of feature enhancement and fusion based on multi-scale feature maps, improving detection accuracy of small objects on MS COCO.
{"title":"Enhancement and Fusion of Multi-Scale Feature Maps for Small Object Detection","authors":"Zhi-Shuang Xue, Wenjie Chen, Jing Li","doi":"10.23919/CCC50068.2020.9189352","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9189352","url":null,"abstract":"In recent years, deep convolutional neural networks have made breakthrough progress in object recognition and object detection tasks in the field of computer vision, and have achieved great results both in accuracy and speed. However, the detection of small objects is still difficult in the field of object detection, and the accuracy on the common dataset MS COCO is very low. This paper briefly reviews some work in multi-scale object detection algorithms, and then proposes a method of feature enhancement and fusion based on multi-scale feature maps, improving detection accuracy of small objects on MS COCO.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116592076","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}
In this paper, a forest data map returned by Beidou satellites is combined with classic theories such as graph theory and mathematical planning, and the Dijkstra algorithm, which is the most widely used in practice, is the core to propose a forest patrol optimal path planning model. And by comparing with the Floyd path planning algorithm, the correctness and implementability of the model are verified. The experimental results show that the algorithm effectively solves the problem of optimal path decision from a patrol base to an arbitrary forest area in a certain forest, and verifies that the Dijkstra algorithm can be accurately and reliably applied to the optimal path planning field of forest patrol.
{"title":"Analysis and Decision of Optimal Path of Forest Disaster Patrol Based on Beidou Navigation","authors":"Jinyi Ma, Lifu Hu, Bini Pan, Zhongzhi Li, Yuqing Tian, Chunyu Chen","doi":"10.23919/CCC50068.2020.9188560","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9188560","url":null,"abstract":"In this paper, a forest data map returned by Beidou satellites is combined with classic theories such as graph theory and mathematical planning, and the Dijkstra algorithm, which is the most widely used in practice, is the core to propose a forest patrol optimal path planning model. And by comparing with the Floyd path planning algorithm, the correctness and implementability of the model are verified. The experimental results show that the algorithm effectively solves the problem of optimal path decision from a patrol base to an arbitrary forest area in a certain forest, and verifies that the Dijkstra algorithm can be accurately and reliably applied to the optimal path planning field of forest patrol.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117076004","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 : 2020-07-01DOI: 10.23919/CCC50068.2020.9189393
Jing-song Bai, Liping Yan, Yuanqing Xia, Bo Xiao
Infrared and visible image fusion technology helps to improve the spatial resolution. It mainly preserves the features and details of the source images and generates a fusion image with visual enhancement. In this paper, based on the gradient features and intensity information of the source images, an optimization model for image fusion is built. Firstly, the pre-fused gradient of the source images is obtained by combining the structure tensor and the proposed local gradient similarity, where local gradient similarity is used to make the fused gradient direction more accurately. Secondly, the source images are reconstructed into salient and non-salient images according to the comparison of the pixel intensity. A weight map before the non-salient image in the optimization model makes the effective details preserved, so that the pre-fused images consist of the salient image and the non-salient image with a weight map. Finally, an optimization model is constructed to constrain the gradient and intensity of the final fused image close to the pre-fused gradient and the pre-fused images. The final fused image is obtained from solving the optimization model by use of the variational method. The experimental results are evaluated from subjective and objective assessments, which show the effectiveness of the proposed algorithm.
{"title":"Image Fusion based on Variational Method for Maintenance of Gradient and Intensity","authors":"Jing-song Bai, Liping Yan, Yuanqing Xia, Bo Xiao","doi":"10.23919/CCC50068.2020.9189393","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9189393","url":null,"abstract":"Infrared and visible image fusion technology helps to improve the spatial resolution. It mainly preserves the features and details of the source images and generates a fusion image with visual enhancement. In this paper, based on the gradient features and intensity information of the source images, an optimization model for image fusion is built. Firstly, the pre-fused gradient of the source images is obtained by combining the structure tensor and the proposed local gradient similarity, where local gradient similarity is used to make the fused gradient direction more accurately. Secondly, the source images are reconstructed into salient and non-salient images according to the comparison of the pixel intensity. A weight map before the non-salient image in the optimization model makes the effective details preserved, so that the pre-fused images consist of the salient image and the non-salient image with a weight map. Finally, an optimization model is constructed to constrain the gradient and intensity of the final fused image close to the pre-fused gradient and the pre-fused images. The final fused image is obtained from solving the optimization model by use of the variational method. The experimental results are evaluated from subjective and objective assessments, which show the effectiveness of the proposed algorithm.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129593019","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 : 2020-07-01DOI: 10.23919/CCC50068.2020.9189020
Jikun Rong, S. Arrigoni, N. Luan, F. Braghin
Sampling-based motion planning(SMPs) approach has been very popular for its ability of computing collision-free and asymptotically optimal path without explicit formulation of the configuration space. SMPs use sampling to generate a discrete representation of the problem and then run graph searching algorithm on this representation. Which means the representation itself is at least as important as graph searching algorithm. In general this is enabled by uniformly sampling the configuration space. This paper proposes a machine learning based biased sampling approach for autonomous driving. The sampling distribution was learned from previous demonstrations using conditional variational encoder(CVAE) with attention mechanism. Combined with a sampling-based algorithm called rapidly-exploring random tree*(RRT*), we proposed Attention-RRT*. This approach was proved to be effective in several driving scenarios.
{"title":"Attention-based Sampling Distribution for Motion Planning in Autonomous Driving","authors":"Jikun Rong, S. Arrigoni, N. Luan, F. Braghin","doi":"10.23919/CCC50068.2020.9189020","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9189020","url":null,"abstract":"Sampling-based motion planning(SMPs) approach has been very popular for its ability of computing collision-free and asymptotically optimal path without explicit formulation of the configuration space. SMPs use sampling to generate a discrete representation of the problem and then run graph searching algorithm on this representation. Which means the representation itself is at least as important as graph searching algorithm. In general this is enabled by uniformly sampling the configuration space. This paper proposes a machine learning based biased sampling approach for autonomous driving. The sampling distribution was learned from previous demonstrations using conditional variational encoder(CVAE) with attention mechanism. Combined with a sampling-based algorithm called rapidly-exploring random tree*(RRT*), we proposed Attention-RRT*. This approach was proved to be effective in several driving scenarios.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129604885","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 : 2020-07-01DOI: 10.23919/CCC50068.2020.9188625
Wanzhen Quan, Yanhong Zhang, Xia Liu, Xiaogang Yang, Jianxiang Xi
The problem of energy-limited formation for second multi-agent systems in leaderless case is investigated in this paper. First, considering that the energy is given in advance and is limited, an energy-limited formation protocol is proposed by local relative velocities and positions of each agent and its neighbors. Second, according to a special property of the Laplace matrix, the formation consensus problem is converted into a stabilization problem. Then, by the Lyapunov theory, sufficient conditions for second multi-agent systems in the form of analytic solutions of control gains are presented to achieve the energy-limited formation. Besides, the formation center function is determined, which is to show the macroscopic motion of formations. Finally, an example is given to demonstrate the validity of the the proposed theoretical results.
{"title":"Energy-limited formation design for second-order multi-agent systems","authors":"Wanzhen Quan, Yanhong Zhang, Xia Liu, Xiaogang Yang, Jianxiang Xi","doi":"10.23919/CCC50068.2020.9188625","DOIUrl":"https://doi.org/10.23919/CCC50068.2020.9188625","url":null,"abstract":"The problem of energy-limited formation for second multi-agent systems in leaderless case is investigated in this paper. First, considering that the energy is given in advance and is limited, an energy-limited formation protocol is proposed by local relative velocities and positions of each agent and its neighbors. Second, according to a special property of the Laplace matrix, the formation consensus problem is converted into a stabilization problem. Then, by the Lyapunov theory, sufficient conditions for second multi-agent systems in the form of analytic solutions of control gains are presented to achieve the energy-limited formation. Besides, the formation center function is determined, which is to show the macroscopic motion of formations. Finally, an example is given to demonstrate the validity of the the proposed theoretical results.","PeriodicalId":255872,"journal":{"name":"2020 39th Chinese Control Conference (CCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129941908","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：