Pub Date : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053860
Yixiang Shen, Song Xiao, Zheng Chen, Yujun Guo, Xueqin Zhang, Jiancheng Liu, Changlei Ju
The grounding current of the train is an important reason for the electrical corrosion of the carbon brushes and axles of the train wheels. The distribution and variation of train grounding current are closely related to the train integrated grounding system. The integrated grounding system of high-speed train includes two subsystems: on-board mobile grounding subsystem and traction network fixed grounding subsystem. AT power supply mode has many technical advantages, such as long power supply distance and strong power transmission capacity, etc., so it has become the main power supply solution for traction power supply of modern high-speed railways. The traction network fixed grounding subsystem under the AT power supply mode is very different from other traction power supply schemes. Therefore, in view of the grounding current problem of modern high-speed trains, this paper establishes a simulation model of the integrated grounding system of high-speed trains under the AT power supply mode, and analyzes the distribution and variation of grounding currents of modern high-speed trains.
{"title":"The Analysis of Grounding Current of Modern High-Speed Trains","authors":"Yixiang Shen, Song Xiao, Zheng Chen, Yujun Guo, Xueqin Zhang, Jiancheng Liu, Changlei Ju","doi":"10.1109/ICCR55715.2022.10053860","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053860","url":null,"abstract":"The grounding current of the train is an important reason for the electrical corrosion of the carbon brushes and axles of the train wheels. The distribution and variation of train grounding current are closely related to the train integrated grounding system. The integrated grounding system of high-speed train includes two subsystems: on-board mobile grounding subsystem and traction network fixed grounding subsystem. AT power supply mode has many technical advantages, such as long power supply distance and strong power transmission capacity, etc., so it has become the main power supply solution for traction power supply of modern high-speed railways. The traction network fixed grounding subsystem under the AT power supply mode is very different from other traction power supply schemes. Therefore, in view of the grounding current problem of modern high-speed trains, this paper establishes a simulation model of the integrated grounding system of high-speed trains under the AT power supply mode, and analyzes the distribution and variation of grounding currents of modern high-speed trains.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126145046","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 : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053858
Xu Rongjing, L. Bin, Li Na, Feng Yangshu, Chen Fangdong
The bottom sewage raking and suction obstacle avoidance robot for aquaculture pond is a sewage cleaning device that uses raking and suction technology and path planning as technical means to build a multi particle hybrid identification system, which can realize the effective cleaning of aquaculture pond. The team took turbot breeding and seedling raising as the research object, and carried out simulation tests on the decontamination robot. The test found that the path planning of GBNN algorithm in obstacle environment can achieve high-efficiency decontamination effect, and the total energy consumption and distance are small, which proved that the robot can use cbnn model to carry out adaptive cruise in static obstacle environment. At the same time, the raking and suction performance of the robot is tested, and through the range analysis method of orthogonal experiment, it is determined that the raking and suction efficiency and raking and suction coverage of the cleaning robot can reach 90%.
{"title":"Research on Methods of Rake Suction and Obstacle Avoidance of Sewage Cleaning Robot for Aquaculture Pond","authors":"Xu Rongjing, L. Bin, Li Na, Feng Yangshu, Chen Fangdong","doi":"10.1109/ICCR55715.2022.10053858","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053858","url":null,"abstract":"The bottom sewage raking and suction obstacle avoidance robot for aquaculture pond is a sewage cleaning device that uses raking and suction technology and path planning as technical means to build a multi particle hybrid identification system, which can realize the effective cleaning of aquaculture pond. The team took turbot breeding and seedling raising as the research object, and carried out simulation tests on the decontamination robot. The test found that the path planning of GBNN algorithm in obstacle environment can achieve high-efficiency decontamination effect, and the total energy consumption and distance are small, which proved that the robot can use cbnn model to carry out adaptive cruise in static obstacle environment. At the same time, the raking and suction performance of the robot is tested, and through the range analysis method of orthogonal experiment, it is determined that the raking and suction efficiency and raking and suction coverage of the cleaning robot can reach 90%.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129134624","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 : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053927
Yun‐Shan Wei, Jiaxuan Wang, Jin‐Fan Wang
This article presents an open-closed-loop iterative learning control (ILC) scheme for non-linear discrete-time multiple-input multiple-output (MIMO) systems under iterative varying duration. The improved P-type ILC law with feedback control is presented to compensate the missing tracking information of the previous iterations due to the iterative varying duration. It is proved that when the initial state expectation is identical to the reference sate, ILC tracking error can be driven to zero in mathematical expectation sense. Finally, a numerical example of simulation is provided to verify the validity of the proposed ILC law.
{"title":"Open-Closed-Loop Iterative Learning Control for Non-linear Discrete-time Systems under Iterative Varying Duration","authors":"Yun‐Shan Wei, Jiaxuan Wang, Jin‐Fan Wang","doi":"10.1109/ICCR55715.2022.10053927","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053927","url":null,"abstract":"This article presents an open-closed-loop iterative learning control (ILC) scheme for non-linear discrete-time multiple-input multiple-output (MIMO) systems under iterative varying duration. The improved P-type ILC law with feedback control is presented to compensate the missing tracking information of the previous iterations due to the iterative varying duration. It is proved that when the initial state expectation is identical to the reference sate, ILC tracking error can be driven to zero in mathematical expectation sense. Finally, a numerical example of simulation is provided to verify the validity of the proposed ILC law.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130678127","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 : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053904
Xuan Rao, Songyi Xiao, Jiaxin Li, Qiuye Wu, Bo Zhao, Derong Liu
From the perspective of data stream, neural architecture search (NAS) can be formulated as a graph optimization problem. However, many state-of-the-art black-box optimization algorithms, such as Bayesian optimization and simulated annealing, operate in continuous space primarily, which does not match the NAS optimization due to the discreteness of graph structures. To tackle this problem, the latent space Bayesian optimization NAS (LSBO-NAS) algorithm is developed in this paper. In LSBO-NAS, the neural architectures are represented as sequences, and a variational auto-encoder (VAE) is trained to convert the discrete search space of NAS into a continuous latent space by learning the continuous representation of neural architectures. Hereafter, a Bayesian optimization (BO) algorithm, i.e., the tree-structure parzen estimator (TPE) algorithm, is developed to obtain admirable neural architectures. The optimization loop of LSBO-NAS consists of two stages. In the first stage, the BO algorithm generates a preferable architecture representation according to its search strategy. In the second stage, the decoder of VAE decodes the representation into a discrete neural architecture, whose performance evaluation is regarded as the feedback signal for the BO algorithm. The effectiveness of the developed LSBO-NAS is demonstrated on the NAS-Bench-301 benchmark, where the LSBO-NAS achieves a better performance than several NAS baselines.
{"title":"LSBO-NAS: Latent Space Bayesian Optimization for Neural Architecture Search","authors":"Xuan Rao, Songyi Xiao, Jiaxin Li, Qiuye Wu, Bo Zhao, Derong Liu","doi":"10.1109/ICCR55715.2022.10053904","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053904","url":null,"abstract":"From the perspective of data stream, neural architecture search (NAS) can be formulated as a graph optimization problem. However, many state-of-the-art black-box optimization algorithms, such as Bayesian optimization and simulated annealing, operate in continuous space primarily, which does not match the NAS optimization due to the discreteness of graph structures. To tackle this problem, the latent space Bayesian optimization NAS (LSBO-NAS) algorithm is developed in this paper. In LSBO-NAS, the neural architectures are represented as sequences, and a variational auto-encoder (VAE) is trained to convert the discrete search space of NAS into a continuous latent space by learning the continuous representation of neural architectures. Hereafter, a Bayesian optimization (BO) algorithm, i.e., the tree-structure parzen estimator (TPE) algorithm, is developed to obtain admirable neural architectures. The optimization loop of LSBO-NAS consists of two stages. In the first stage, the BO algorithm generates a preferable architecture representation according to its search strategy. In the second stage, the decoder of VAE decodes the representation into a discrete neural architecture, whose performance evaluation is regarded as the feedback signal for the BO algorithm. The effectiveness of the developed LSBO-NAS is demonstrated on the NAS-Bench-301 benchmark, where the LSBO-NAS achieves a better performance than several NAS baselines.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129233480","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 : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053919
Peng Li, Mao Wang, Jinyu Fu, Yankun Wang
Aiming at the problem of simple and fast recognition of non-cooperative targets in 3D space, a simple recognition algorithm for point cloud targets is proposed. First, the point cloud data was divided into $n$ categories with the first K-means clustering. Second, the target class was identified with a coarse sieve, and the speed of the algorithm was improved with sparse processing. The more accurate target class was obtained with secondary clustering. The two types of point cloud data are processed by principal component analysis (PCA), which obtains the feature root matrices. Then cosine distance matching was applied to the feature root matrices and target library (trained by 12 groups of point cloud data). This type of data was retained when the similarity was greater than the upper threshold. Therefore, the center point coordinates, distances, and similarity of the target were outputted. The experimental test results of the 13th and 14th groups indicated that the target segmentation similarity of this algorithm could reach 95.75% and 96.98% respectively, and the accuracy reached 100%.
{"title":"A Simple Algorithm for Non-cooperative Target Recognition Based on Lidar","authors":"Peng Li, Mao Wang, Jinyu Fu, Yankun Wang","doi":"10.1109/ICCR55715.2022.10053919","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053919","url":null,"abstract":"Aiming at the problem of simple and fast recognition of non-cooperative targets in 3D space, a simple recognition algorithm for point cloud targets is proposed. First, the point cloud data was divided into $n$ categories with the first K-means clustering. Second, the target class was identified with a coarse sieve, and the speed of the algorithm was improved with sparse processing. The more accurate target class was obtained with secondary clustering. The two types of point cloud data are processed by principal component analysis (PCA), which obtains the feature root matrices. Then cosine distance matching was applied to the feature root matrices and target library (trained by 12 groups of point cloud data). This type of data was retained when the similarity was greater than the upper threshold. Therefore, the center point coordinates, distances, and similarity of the target were outputted. The experimental test results of the 13th and 14th groups indicated that the target segmentation similarity of this algorithm could reach 95.75% and 96.98% respectively, and the accuracy reached 100%.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126510434","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 : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053686
Jing Li, Zhe Wang, Shengzhi Yuan, Haidi Dong
The problem of how to approximate unknown time-varying nonlinear functions is researched in this paper. Firstly, a new RBF NN with time-varying weight is proposed to approximate the unknown time-varying nonlinear function. Secondly, the approximate theorem of the proposed time-varying RBF NN is obtained. Accordingly, a conclusion can be drawn that a continuous time-varying nonlinear function defined on finite time interval [0, T] can be approximated by at least a piecewise continuous time-varying weight vector and a finite number of RBF neurons. Finally, simulation examples are given to validate the effectiveness of proposed time-varying RBF NN.
{"title":"Research on Time-varying RBF NN and Its Application","authors":"Jing Li, Zhe Wang, Shengzhi Yuan, Haidi Dong","doi":"10.1109/ICCR55715.2022.10053686","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053686","url":null,"abstract":"The problem of how to approximate unknown time-varying nonlinear functions is researched in this paper. Firstly, a new RBF NN with time-varying weight is proposed to approximate the unknown time-varying nonlinear function. Secondly, the approximate theorem of the proposed time-varying RBF NN is obtained. Accordingly, a conclusion can be drawn that a continuous time-varying nonlinear function defined on finite time interval [0, T] can be approximated by at least a piecewise continuous time-varying weight vector and a finite number of RBF neurons. Finally, simulation examples are given to validate the effectiveness of proposed time-varying RBF NN.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"160 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123104536","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 : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053912
Yaxian Zhang, Sen Zhang
The reliability prediction of time series of blast furnace gas flow is beneficial to the stable running of blast furnace condition. Aiming at the problem of gas flow time series prediction, this paper proposes a single-step prediction and multi-step prediction based on LSTM algorithm. Firstly, the original data is preprocessed, such as outlier processing and denoising processing of Fourier Transform, so as to reduce the prediction error. Secondly, it will finish single-step prediction and multi-step prediction by adopting LSTM algorithm. Finally, it evaluates the performance of LSTM prediction model. The experiments show that the accuracy of LSTM prediction is high, but the single-step prediction takes a long time; however, in the process of blast furnace gas flow prediction, the time parameter is an indispensable characteristic. Considering comprehensively, the LSTM multi-step prediction shows a better prediction effect, which provides a reliable reference for the stable operation of blast furnace.
{"title":"Research on Blast Furnace Gas Flow Prediction Method Based on LSTM","authors":"Yaxian Zhang, Sen Zhang","doi":"10.1109/ICCR55715.2022.10053912","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053912","url":null,"abstract":"The reliability prediction of time series of blast furnace gas flow is beneficial to the stable running of blast furnace condition. Aiming at the problem of gas flow time series prediction, this paper proposes a single-step prediction and multi-step prediction based on LSTM algorithm. Firstly, the original data is preprocessed, such as outlier processing and denoising processing of Fourier Transform, so as to reduce the prediction error. Secondly, it will finish single-step prediction and multi-step prediction by adopting LSTM algorithm. Finally, it evaluates the performance of LSTM prediction model. The experiments show that the accuracy of LSTM prediction is high, but the single-step prediction takes a long time; however, in the process of blast furnace gas flow prediction, the time parameter is an indispensable characteristic. Considering comprehensively, the LSTM multi-step prediction shows a better prediction effect, which provides a reliable reference for the stable operation of blast furnace.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"90 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121444079","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 : 2022-12-02DOI: 10.1109/ICCR55715.2022.10053895
Rong Yuan, Shengli Xie, Feng Gao, Zhenni Li
Classic multi-frequency combination is mainly to build a long wavelength combination and eliminate the ionospheric delay combination, and the combined noise factor is greater than 1. On the basis of un-differenced and un-combining observations, this paper proposes a general real coefficient combination based on the principle of minimum noise coefficients. The combined noise coefficient is less than 1, which can be applied to any multi-frequency observation combination. According to the frequency of each signal frequency of Galileo, we give the optimal combination of Galileo multi-frequency noise coefficients and the effect of theoretical improving accuracy. Compared with single frequency observation, the optimal combination of dual frequency triple frequency, four frequency and five frequency improve the observation accuracy by 40%, 49%,54% and 59% respectively. It is verified by the observation of the actual Galileo three frequency E1, E5A and E5b signals, the optimal combination of dual frequency and triple frequency improves the observation accuracy by 52% and 62% respectively compared with single frequency observation. According to the actual measurements, the observation accuracy improvement of minimum noise coefficients combination is basically consistent with the theoretical analysis. Finally, we analyze the equivalence between the minimum noise combination and the un-differenced un-combining observations, the results show that the minimum noise combination is an optimal weight model of un-differenced un-combining observations, which the optimization criterion is the minimum observation noise.
{"title":"Galileo Multi-frequency Observation Combination Method Based on Minimum Noise Coefficients","authors":"Rong Yuan, Shengli Xie, Feng Gao, Zhenni Li","doi":"10.1109/ICCR55715.2022.10053895","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053895","url":null,"abstract":"Classic multi-frequency combination is mainly to build a long wavelength combination and eliminate the ionospheric delay combination, and the combined noise factor is greater than 1. On the basis of un-differenced and un-combining observations, this paper proposes a general real coefficient combination based on the principle of minimum noise coefficients. The combined noise coefficient is less than 1, which can be applied to any multi-frequency observation combination. According to the frequency of each signal frequency of Galileo, we give the optimal combination of Galileo multi-frequency noise coefficients and the effect of theoretical improving accuracy. Compared with single frequency observation, the optimal combination of dual frequency triple frequency, four frequency and five frequency improve the observation accuracy by 40%, 49%,54% and 59% respectively. It is verified by the observation of the actual Galileo three frequency E1, E5A and E5b signals, the optimal combination of dual frequency and triple frequency improves the observation accuracy by 52% and 62% respectively compared with single frequency observation. According to the actual measurements, the observation accuracy improvement of minimum noise coefficients combination is basically consistent with the theoretical analysis. Finally, we analyze the equivalence between the minimum noise combination and the un-differenced un-combining observations, the results show that the minimum noise combination is an optimal weight model of un-differenced un-combining observations, which the optimization criterion is the minimum observation noise.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121578889","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, we propose a co-saliency detection algorithm based on multi-scale feature extraction and feature fusion. The algorithm extracts multi-scale features of images based on image information and combines these multiscale features with single image saliency maps (SISMs) generated by the edge guidance network (EGNet) to obtain single image vectors (SIVs). Based on these features, self-correlated features (SCFs) and rearranged self-correlated features (RSCFs) are calculated, and co-saliency attention (CSA) maps are created by weighting. Finally, the decoder receives the rearranged self-correlation and co-saliency maps in order to generate the final prediction maps. It can effectively solve the problem of poor performance of current feature extraction and saliency detection algorithms in complex scenes with multiple saliency targets. The simulation results show that the proposed algorithm not only improves the accuracy of co-saliency detection of RGB images in complex scenes but also reduces the error, and its performance is better than other algorithms.
{"title":"Co-Saliency Detection Based on Multi-Scale Feature Extraction and Feature Fusion","authors":"Kuang Zuo, Huiqing Liang, De-Cheng Wang, Dehua Zhang","doi":"10.1109/ICCR55715.2022.10053903","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053903","url":null,"abstract":"In this paper, we propose a co-saliency detection algorithm based on multi-scale feature extraction and feature fusion. The algorithm extracts multi-scale features of images based on image information and combines these multiscale features with single image saliency maps (SISMs) generated by the edge guidance network (EGNet) to obtain single image vectors (SIVs). Based on these features, self-correlated features (SCFs) and rearranged self-correlated features (RSCFs) are calculated, and co-saliency attention (CSA) maps are created by weighting. Finally, the decoder receives the rearranged self-correlation and co-saliency maps in order to generate the final prediction maps. It can effectively solve the problem of poor performance of current feature extraction and saliency detection algorithms in complex scenes with multiple saliency targets. The simulation results show that the proposed algorithm not only improves the accuracy of co-saliency detection of RGB images in complex scenes but also reduces the error, and its performance is better than other algorithms.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133480606","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}
Cooperative Unmanned aerial vehicles (UAVs) have been widely employed as effective tools for various information-gathering tasks in complex environments with increased efficiency and resiliency. The mission-level guidance and control of UAVs often depend on an accurate map and inaccurate maps may lead to the UAV's inappropriate accommodation to the environment. In this paper, we propose a new framework to generate and utilize semantic map information, which we defined as risk factors for cooperative UAVs. First, we generate a high-precision panorama as a global map by mosaicking a bird's-eye atlas. Afterward, we build a semantic map based on a neural network. Finally, we utilize the semantic information-enhanced map to guide the path-planning functions. Experiments show that our proposed method can improve the success rate of planning in the outdoor scene, and demonstrate its efficiency.
{"title":"Semantic Information Based Path Planning for Cooperative UAV Systems","authors":"Zhiwei Wang, Chunhui Zhao, Yang Lyu, Huixia Liu, Jin-wen Hu, X. Hou","doi":"10.1109/ICCR55715.2022.10053900","DOIUrl":"https://doi.org/10.1109/ICCR55715.2022.10053900","url":null,"abstract":"Cooperative Unmanned aerial vehicles (UAVs) have been widely employed as effective tools for various information-gathering tasks in complex environments with increased efficiency and resiliency. The mission-level guidance and control of UAVs often depend on an accurate map and inaccurate maps may lead to the UAV's inappropriate accommodation to the environment. In this paper, we propose a new framework to generate and utilize semantic map information, which we defined as risk factors for cooperative UAVs. First, we generate a high-precision panorama as a global map by mosaicking a bird's-eye atlas. Afterward, we build a semantic map based on a neural network. Finally, we utilize the semantic information-enhanced map to guide the path-planning functions. Experiments show that our proposed method can improve the success rate of planning in the outdoor scene, and demonstrate its efficiency.","PeriodicalId":441511,"journal":{"name":"2022 4th International Conference on Control and Robotics (ICCR)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133540864","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：