Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000125
Le Xinlong;Cao Xibin;Dai Yu;Wu Fan
Satellites with altitudes below 400 km are called super low altitude satellites (SLAS), often used to achieve responsive imaging tasks. Therefore, it is important for the manipulation of its ground track. Aiming at the problem of ground track manipulation of SLAS, a control method based on tangential impulse thrust is proposed. First, the equation of the longitude difference between SLAS and the target point on the target latitude is derived based on Gauss's variational equations. On this basis, the influence of the tangential impulse thrust on the ground track's longitude is derived. Finally, the method for ground track manipulation of SLAS under the tangential impulse thrust is proposed. The simulation results verify the effectiveness of the method, after manipulation, the satellite can visit the target point and revisit it for multiple days.
{"title":"Method of SLAS's ground track manipulation based on tangential impulse thrust","authors":"Le Xinlong;Cao Xibin;Dai Yu;Wu Fan","doi":"10.23919/JSEE.2023.000125","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000125","url":null,"abstract":"Satellites with altitudes below 400 km are called super low altitude satellites (SLAS), often used to achieve responsive imaging tasks. Therefore, it is important for the manipulation of its ground track. Aiming at the problem of ground track manipulation of SLAS, a control method based on tangential impulse thrust is proposed. First, the equation of the longitude difference between SLAS and the target point on the target latitude is derived based on Gauss's variational equations. On this basis, the influence of the tangential impulse thrust on the ground track's longitude is derived. Finally, the method for ground track manipulation of SLAS under the tangential impulse thrust is proposed. The simulation results verify the effectiveness of the method, after manipulation, the satellite can visit the target point and revisit it for multiple days.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1285-1293"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000131
Pan Li;Wu Zhonghong;Yu Minggang;Liu Jintao;Mei Dan
The key advantage of unmanned swarm operation is its autonomous cooperation. How to improve the proportion of cooperators is one of the key issues of autonomous collaboration in unmanned swarm operations. This work proposes a strategy dominance mechanism of autonomous collaboration in unmanned swarm within the framework of public goods game. It starts with the requirement analysis of autonomous collaboration in unmanned swarm; and an aspiration-driven multiplayer evolutionary game model is established based on the requirement. Then the average abundance function and strategy dominance condition of the model are constructed by theoretical derivation. Furthermore, the evolutionary mechanism of parameter adjustment in swarm cooperation is revealed via simulation, and the influences of the multiplication factor r, aspiration level a, threshold $m$ and other parameters on the strategy dominance conditions were simulated for both linear and threshold public goods games (PGGs) to determine the strategy dominance characteristics; Finally, deliberate proposals are suggested to provide a meaningful exploration in the actual control of unmanned swarm cooperation.
{"title":"Strategy dominance mechanism of autonomous collaboration in unmanned swarm within the framework of public goods game","authors":"Pan Li;Wu Zhonghong;Yu Minggang;Liu Jintao;Mei Dan","doi":"10.23919/JSEE.2023.000131","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000131","url":null,"abstract":"The key advantage of unmanned swarm operation is its autonomous cooperation. How to improve the proportion of cooperators is one of the key issues of autonomous collaboration in unmanned swarm operations. This work proposes a strategy dominance mechanism of autonomous collaboration in unmanned swarm within the framework of public goods game. It starts with the requirement analysis of autonomous collaboration in unmanned swarm; and an aspiration-driven multiplayer evolutionary game model is established based on the requirement. Then the average abundance function and strategy dominance condition of the model are constructed by theoretical derivation. Furthermore, the evolutionary mechanism of parameter adjustment in swarm cooperation is revealed via simulation, and the influences of the multiplication factor r, aspiration level a, threshold \u0000<tex>$m$</tex>\u0000 and other parameters on the strategy dominance conditions were simulated for both linear and threshold public goods games (PGGs) to determine the strategy dominance characteristics; Finally, deliberate proposals are suggested to provide a meaningful exploration in the actual control of unmanned swarm cooperation.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1252-1266"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000126
Jin Tao;Zhang Xindong
Rare labeled data are difficult to recognize by using conventional methods in the process of radar emitter recognition. To solve this problem, an optimized cooperative semisupervised learning radar emitter recognition method based on a small amount of labeled data is developed. First, a small amount of labeled data are randomly sampled by using the bootstrap method, loss functions for three common deep learning networks are improved, the uniform distribution and cross-entropy function are combined to reduce the overconfidence of softmax classification. Subsequently, the dataset obtained after sampling is adopted to train three improved networks so as to build the initial model. In addition, the unlabeled data are preliminarily screened through dynamic time warping (DTW) and then input into the initial model trained previously for judgment. If the judgment results of two or more networks are consistent, the unlabeled data are labeled and put into the labeled data set. Lastly, the three network models are input into the labeled dataset for training, and the final model is built. As revealed by the simulation results, the semi-supervised learning method adopted in this paper is capable of exploiting a small amount of labeled data and basically achieving the accuracy of labeled data recognition.
{"title":"Radar emitter signal recognition method based on improved collaborative semi-supervised learning","authors":"Jin Tao;Zhang Xindong","doi":"10.23919/JSEE.2023.000126","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000126","url":null,"abstract":"Rare labeled data are difficult to recognize by using conventional methods in the process of radar emitter recognition. To solve this problem, an optimized cooperative semisupervised learning radar emitter recognition method based on a small amount of labeled data is developed. First, a small amount of labeled data are randomly sampled by using the bootstrap method, loss functions for three common deep learning networks are improved, the uniform distribution and cross-entropy function are combined to reduce the overconfidence of softmax classification. Subsequently, the dataset obtained after sampling is adopted to train three improved networks so as to build the initial model. In addition, the unlabeled data are preliminarily screened through dynamic time warping (DTW) and then input into the initial model trained previously for judgment. If the judgment results of two or more networks are consistent, the unlabeled data are labeled and put into the labeled data set. Lastly, the three network models are input into the labeled dataset for training, and the final model is built. As revealed by the simulation results, the semi-supervised learning method adopted in this paper is capable of exploiting a small amount of labeled data and basically achieving the accuracy of labeled data recognition.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1182-1190"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000124
Wang Xinfeng;Wang Tao;Zhou Xin;Wang Yanfeng
Strategic management of equipment system development must attach importance to effective strategic risk management. Aiming at the identification of strategic risk of equipment system development, firstly, the source of strategic risk of equip-ment system development is analyzed and classified. Based on this, a causal loop diagram of strategic risk of equipment system development based on system dynamics is established. The system dynamics analysis software Vensim PLE is used to carry out the risk influencing factors analysis, risk consequences analysis, risk feedback loop identification and corresponding pre-control measures, and achieves a good risk identification effect.
{"title":"Research on strategic risk identification method of equipment system development based on system dynamics","authors":"Wang Xinfeng;Wang Tao;Zhou Xin;Wang Yanfeng","doi":"10.23919/JSEE.2023.000124","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000124","url":null,"abstract":"Strategic management of equipment system development must attach importance to effective strategic risk management. Aiming at the identification of strategic risk of equipment system development, firstly, the source of strategic risk of equip-ment system development is analyzed and classified. Based on this, a causal loop diagram of strategic risk of equipment system development based on system dynamics is established. The system dynamics analysis software Vensim PLE is used to carry out the risk influencing factors analysis, risk consequences analysis, risk feedback loop identification and corresponding pre-control measures, and achieves a good risk identification effect.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1225-1234"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000129
Lei Xue;Lu Ningyun;Chen Chuang;Hu Tianzhen;Jiang Bin
Effective bearing fault diagnosis is vital for the safe and reliable operation of rotating machinery. In practical applications, bearings often work at various rotational speeds as well as load conditions. Yet, the bearing fault diagnosis under multiple conditions is a new subject, which needs to be further explored. Therefore, a multi-scale deep belief network (DBN) method integrated with attention mechanism is proposed for the purpose of extracting the multi-scale core features from vibration signals, containing four primary steps: preprocessing of multi-scale data, feature extraction, feature fusion, and fault classification. The key novelties include multi-scale feature extraction using multiscale DBN algorithm, and feature fusion using attention mechanism. The benchmark dataset from University of Ottawa is applied to validate the effectiveness as well as advantages of this method. Furthermore, the aforementioned method is compared with four classical fault diagnosis methods reported in the literature, and the comparison results show that our proposed method has higher diagnostic accuracy and better robustness.
{"title":"Attention mechanism based multi-scale feature extraction of bearing fault diagnosis","authors":"Lei Xue;Lu Ningyun;Chen Chuang;Hu Tianzhen;Jiang Bin","doi":"10.23919/JSEE.2023.000129","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000129","url":null,"abstract":"Effective bearing fault diagnosis is vital for the safe and reliable operation of rotating machinery. In practical applications, bearings often work at various rotational speeds as well as load conditions. Yet, the bearing fault diagnosis under multiple conditions is a new subject, which needs to be further explored. Therefore, a multi-scale deep belief network (DBN) method integrated with attention mechanism is proposed for the purpose of extracting the multi-scale core features from vibration signals, containing four primary steps: preprocessing of multi-scale data, feature extraction, feature fusion, and fault classification. The key novelties include multi-scale feature extraction using multiscale DBN algorithm, and feature fusion using attention mechanism. The benchmark dataset from University of Ottawa is applied to validate the effectiveness as well as advantages of this method. Furthermore, the aforementioned method is compared with four classical fault diagnosis methods reported in the literature, and the comparison results show that our proposed method has higher diagnostic accuracy and better robustness.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1359-1367"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000130
Yang Shunqi;Zeng Ying;Li Xiang;Li Yanfeng;Huang Hongzhong
The dynamic wireless communication network is a complex network that needs to consider various influence factors including communication devices, radio propagation, network topology, and dynamic behaviors. Existing works focus on suggesting simplified reliability analysis methods for these dynamic networks. As one of the most popular modeling methodologies, the dynamic Bayesian network (DBN) is proposed. However, it is insufficient for the wireless communication network which contains temporal and non-temporal events. To this end, we present a modeling methodology for a generalized continuous time Bayesian network (CTBN) with a 2-state conditional probability table (CPT). Moreover, a comprehensive reliability analysis method for communication devices and radio propagation is suggested. The proposed methodology is verified by a reliability analysis of a real wireless communication network.
{"title":"Reliability analysis for wireless communication networks via dynamic Bayesian network","authors":"Yang Shunqi;Zeng Ying;Li Xiang;Li Yanfeng;Huang Hongzhong","doi":"10.23919/JSEE.2023.000130","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000130","url":null,"abstract":"The dynamic wireless communication network is a complex network that needs to consider various influence factors including communication devices, radio propagation, network topology, and dynamic behaviors. Existing works focus on suggesting simplified reliability analysis methods for these dynamic networks. As one of the most popular modeling methodologies, the dynamic Bayesian network (DBN) is proposed. However, it is insufficient for the wireless communication network which contains temporal and non-temporal events. To this end, we present a modeling methodology for a generalized continuous time Bayesian network (CTBN) with a 2-state conditional probability table (CPT). Moreover, a comprehensive reliability analysis method for communication devices and radio propagation is suggested. The proposed methodology is verified by a reliability analysis of a real wireless communication network.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1368-1374"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903688","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000138
Dai Yan;Liu Dan;Li Chuanming;Wei Shaopeng;Hu Qingrong
In the scene of wideband radar, due to the spread of target scattering points, the attitude and angle of view of the target constantly change in the process of moving. It is difficult to predict, and the actual echo of multiple scattered points is not fully matched with the transmitted signal. Therefore, it is challenging for the traditional matching filter method to achieve a complete matching effect in wideband echo detection. In addition, the energy dispersion of complex target echoes is still a problem in radar target detection under broadband conditions. Therefore, this paper proposes a wideband target detection method based on dual-channel correlation processing of rangeextended targets. This method fully uses the spatial distribution characteristics of target scattering points of echo signal and the matching characteristics of the dual-channel point extension function itself. The radial accumulation of wideband target echo signal in the complex domain is realized through the adaptive correlation processing of a dual-channel echo signal. The accu-mulation effect of high matching degree is achieved to improve the detection probability and the performance of wideband detection. Finally, electromagnetic simulation experiments and measured data verify that the proposed method has the advan-tages of high signal to noise ratio (SNR) gain and high detection probability under low SNR conditions.
{"title":"Robust dual-channel correlation algorithm for complex weak target detection with wideband radar","authors":"Dai Yan;Liu Dan;Li Chuanming;Wei Shaopeng;Hu Qingrong","doi":"10.23919/JSEE.2023.000138","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000138","url":null,"abstract":"In the scene of wideband radar, due to the spread of target scattering points, the attitude and angle of view of the target constantly change in the process of moving. It is difficult to predict, and the actual echo of multiple scattered points is not fully matched with the transmitted signal. Therefore, it is challenging for the traditional matching filter method to achieve a complete matching effect in wideband echo detection. In addition, the energy dispersion of complex target echoes is still a problem in radar target detection under broadband conditions. Therefore, this paper proposes a wideband target detection method based on dual-channel correlation processing of rangeextended targets. This method fully uses the spatial distribution characteristics of target scattering points of echo signal and the matching characteristics of the dual-channel point extension function itself. The radial accumulation of wideband target echo signal in the complex domain is realized through the adaptive correlation processing of a dual-channel echo signal. The accu-mulation effect of high matching degree is achieved to improve the detection probability and the performance of wideband detection. Finally, electromagnetic simulation experiments and measured data verify that the proposed method has the advan-tages of high signal to noise ratio (SNR) gain and high detection probability under low SNR conditions.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1130-1146"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000113
Xia Jiawei;Zhu Xufang;Liu Zhong;Xia Qingtao
To solve the path following control problem for unmanned surface vehicles (USVs), a control method based on deep reinforcement learning (DRL) with long short-term memory (LSTM) networks is proposed. A distributed proximal policy optimization (DPPO) algorithm, which is a modified actorcritic-based type of reinforcement learning algorithm, is adapted to improve the controller performance in repeated trials. The LSTM network structure is introduced to solve the strong temporal correlation USV control problem. In addition, a specially designed path dataset, including straight and curved paths, is established to simulate various sailing scenarios so that the reinforcement learning controller can obtain as much handling experience as possible. Extensive numerical simulation results demonstrate that the proposed method has better control performance under missions involving complex maneuvers than trained with limited scenarios and can potentially be applied in practice.
{"title":"LSTM-DPPO based deep reinforcement learning controller for path following optimization of unmanned surface vehicle","authors":"Xia Jiawei;Zhu Xufang;Liu Zhong;Xia Qingtao","doi":"10.23919/JSEE.2023.000113","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000113","url":null,"abstract":"To solve the path following control problem for unmanned surface vehicles (USVs), a control method based on deep reinforcement learning (DRL) with long short-term memory (LSTM) networks is proposed. A distributed proximal policy optimization (DPPO) algorithm, which is a modified actorcritic-based type of reinforcement learning algorithm, is adapted to improve the controller performance in repeated trials. The LSTM network structure is introduced to solve the strong temporal correlation USV control problem. In addition, a specially designed path dataset, including straight and curved paths, is established to simulate various sailing scenarios so that the reinforcement learning controller can obtain as much handling experience as possible. Extensive numerical simulation results demonstrate that the proposed method has better control performance under missions involving complex maneuvers than trained with limited scenarios and can potentially be applied in practice.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1343-1358"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000067
Wu Feng;Liu Xiuluo;Wang Jia;Li Chao;Liu Ying;Su Jianbin;Zhang Ailiang;Wang Min
Space emergency launching is to send a satellite into space by using a rapid responsive solid rocket in the bounded time to implement the emergency Earth observation mission. The key and difficult points mainly include the business process construction of launching mission planning, validation of the effectiveness of the launching scheme, etc. This paper proposes the agile space emergency launching mission planning simulation and verification method, which systematically constructs the overall technical framework of space emergency launching mission planning with multi-field area, multi-platform and multi-task parallel under the constraint of resource scheduling for the first time. It supports flexible reconstruction of mission planning processes such as launching target planning, trajectory planning, path planning, action planning and launching time analysis, and can realize on-demand assembly of operation links under different mission scenarios and different plan conditions, so as to quickly modify and generate launching schemes. It supports the fast solution of rocket trajectory data and the accurate analysis of multi-point salvo time window recheck and can realize the fast conflict resolution of launching missions in the dimensions of launching position and launching window sequence. It supports lightweight scenario design, modular flexible simulation, based on launching style, launching platform, launching rules, etc., can realize the independent mapping of mission planning results to two-dimensional and three-dimensional visual simulation models, so as to achieve a smooth connection between mission planning and simulation.
{"title":"Research on agile space emergency launching mission planning simulation and verification method","authors":"Wu Feng;Liu Xiuluo;Wang Jia;Li Chao;Liu Ying;Su Jianbin;Zhang Ailiang;Wang Min","doi":"10.23919/JSEE.2023.000067","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000067","url":null,"abstract":"Space emergency launching is to send a satellite into space by using a rapid responsive solid rocket in the bounded time to implement the emergency Earth observation mission. The key and difficult points mainly include the business process construction of launching mission planning, validation of the effectiveness of the launching scheme, etc. This paper proposes the agile space emergency launching mission planning simulation and verification method, which systematically constructs the overall technical framework of space emergency launching mission planning with multi-field area, multi-platform and multi-task parallel under the constraint of resource scheduling for the first time. It supports flexible reconstruction of mission planning processes such as launching target planning, trajectory planning, path planning, action planning and launching time analysis, and can realize on-demand assembly of operation links under different mission scenarios and different plan conditions, so as to quickly modify and generate launching schemes. It supports the fast solution of rocket trajectory data and the accurate analysis of multi-point salvo time window recheck and can realize the fast conflict resolution of launching missions in the dimensions of launching position and launching window sequence. It supports lightweight scenario design, modular flexible simulation, based on launching style, launching platform, launching rules, etc., can realize the independent mapping of mission planning results to two-dimensional and three-dimensional visual simulation models, so as to achieve a smooth connection between mission planning and simulation.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1267-1284"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-01DOI: 10.23919/JSEE.2023.000096
Du Hao;Wang Wei;Wang Xuerao;Zuo Jingqiu;Wang Yuanda
In this paper, we study scene image recognition with knowledge transfer for drone navigation. We divide navigation scenes into three macro-classes, namely outdoor special scenes (OSSs), the space from indoors to outdoors or from outdoors to indoors transitional scenes (TSs), and others. However, there are difficulties in how to recognize the TSs, to this end, we employ deep convolutional neural network (CNN) based on knowledge transfer, techniques for image augmentation, and fine tuning to solve the issue. Moreover, there is still a novelty detection problem in the classifier, and we use global navigation satellite systems (GNSS) to solve it in the prediction stage. Experiment results show our method, with a pre-trained model and fine tuning, can achieve 91.3196% top-1 accuracy on Scenes21 dataset, paving the way for drones to learn to understand the scenes around them autonomously.
{"title":"Scene image recognition with knowledge transfer for drone navigation","authors":"Du Hao;Wang Wei;Wang Xuerao;Zuo Jingqiu;Wang Yuanda","doi":"10.23919/JSEE.2023.000096","DOIUrl":"https://doi.org/10.23919/JSEE.2023.000096","url":null,"abstract":"In this paper, we study scene image recognition with knowledge transfer for drone navigation. We divide navigation scenes into three macro-classes, namely outdoor special scenes (OSSs), the space from indoors to outdoors or from outdoors to indoors transitional scenes (TSs), and others. However, there are difficulties in how to recognize the TSs, to this end, we employ deep convolutional neural network (CNN) based on knowledge transfer, techniques for image augmentation, and fine tuning to solve the issue. Moreover, there is still a novelty detection problem in the classifier, and we use global navigation satellite systems (GNSS) to solve it in the prediction stage. Experiment results show our method, with a pre-trained model and fine tuning, can achieve 91.3196% top-1 accuracy on Scenes21 dataset, paving the way for drones to learn to understand the scenes around them autonomously.","PeriodicalId":50030,"journal":{"name":"Journal of Systems Engineering and Electronics","volume":"34 5","pages":"1309-1318"},"PeriodicalIF":0.0,"publicationDate":"2023-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"71903321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","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：