Pub Date : 2024-09-04DOI: 10.1109/TSMC.2024.3445117
Xuan Li;Xiao Wang;Fang Deng;Fei-Yue Wang
Reidentification (Re-ID) is a crucial computer vision application with a variety of potential uses in many maritime scenarios, including search, rescue, and surveillance. However, the development of advanced boat reidentification (Boat Re-ID) algorithms necessitates the availability of large-scale Re-ID datasets for model training and evaluation. Inspired by scenarios engineering, this study proposes a new framework for automatically generating a realistic synthetic dataset for boat Re-ID investigation. The synthetic dataset contains 107 boat models and various visual conditions in 36 real backgrounds. The use of synthetic datasets enables the learning-based Re-ID algorithm’s performance to be quantitatively verificated under varying imaging conditions. Nonetheless, our experiments prove that synthetic datasets are inadequate to handle real-world challenges. Therefore, we present a domain adaptation approach that integrates both real and synthetic data to create trustworthy models. This approach employs a multistep training strategy, gradient reversal layer and novel loss functions to preserve the features from two distribution dataset domains. The results of the experiments demonstrate that 1) synthetic datasets can be employed to train boat Re-ID algorithms and quantitatively test the performance of these algorithms under diverse imaging conditions and 2) our approach utilizes the attributes of the two data domains (real and synthetic) to achieve exceptional performance in real-world applications.
{"title":"Scenarios Engineering for Trustworthy AI: Domain Adaptation Approach for Reidentification With Synthetic Data","authors":"Xuan Li;Xiao Wang;Fang Deng;Fei-Yue Wang","doi":"10.1109/TSMC.2024.3445117","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3445117","url":null,"abstract":"Reidentification (Re-ID) is a crucial computer vision application with a variety of potential uses in many maritime scenarios, including search, rescue, and surveillance. However, the development of advanced boat reidentification (Boat Re-ID) algorithms necessitates the availability of large-scale Re-ID datasets for model training and evaluation. Inspired by scenarios engineering, this study proposes a new framework for automatically generating a realistic synthetic dataset for boat Re-ID investigation. The synthetic dataset contains 107 boat models and various visual conditions in 36 real backgrounds. The use of synthetic datasets enables the learning-based Re-ID algorithm’s performance to be quantitatively verificated under varying imaging conditions. Nonetheless, our experiments prove that synthetic datasets are inadequate to handle real-world challenges. Therefore, we present a domain adaptation approach that integrates both real and synthetic data to create trustworthy models. This approach employs a multistep training strategy, gradient reversal layer and novel loss functions to preserve the features from two distribution dataset domains. The results of the experiments demonstrate that 1) synthetic datasets can be employed to train boat Re-ID algorithms and quantitatively test the performance of these algorithms under diverse imaging conditions and 2) our approach utilizes the attributes of the two data domains (real and synthetic) to achieve exceptional performance in real-world applications.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2024-09-04DOI: 10.1109/TSMC.2024.3448453
Wei Song;Zhi Liu;Shaocong Liu;Xiaofeng Ding;Yinan Guo;Shengxiang Yang
In dynamic optimization problems (DOPs), environmental changes can be characterized as various dynamics. Faced with different dynamics, existing dynamic optimization algorithms (DOAs) are difficult to tackle, because they are incapable of learning in each environment to control the search. Besides, diversity loss is a critical issue in solving DOPs. Maintaining a high-diversity over dynamic environments is reasonable as it can address such an issue automatically. In this article, we propose a particle search control network (PSCN) to maintain a high-diversity over time and control two key search actions of each input individual, i.e., locating the local learning target and adjusting the local acceleration coefficient. Specifically, PSCN adequately considers the diversity to generate subpopulations located by hidden node centers, where each center is assessed by significance-based criteria and distance-based criteria. The former enable a small intrasubpopulation distance and a big search scope (subpopulation width) for each subpopulation, while the latter make each center distant from other existing centers. In each subpopulation, the best-found position is selected as the local learning target. In the output layer, PSCN determines the action of adjusting the local acceleration coefficient of each individual. Reinforcement learning is introduced to obtain the desired output of PSCN, enabling the network to control the search by learning in different iterations of each environment. The experimental results especially performance comparisons with eight state-of-the-art DOAs demonstrate that PSCN brings significant improvements in performance of solving DOPs.
{"title":"Particle Search Control Network for Dynamic Optimization","authors":"Wei Song;Zhi Liu;Shaocong Liu;Xiaofeng Ding;Yinan Guo;Shengxiang Yang","doi":"10.1109/TSMC.2024.3448453","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3448453","url":null,"abstract":"In dynamic optimization problems (DOPs), environmental changes can be characterized as various dynamics. Faced with different dynamics, existing dynamic optimization algorithms (DOAs) are difficult to tackle, because they are incapable of learning in each environment to control the search. Besides, diversity loss is a critical issue in solving DOPs. Maintaining a high-diversity over dynamic environments is reasonable as it can address such an issue automatically. In this article, we propose a particle search control network (PSCN) to maintain a high-diversity over time and control two key search actions of each input individual, i.e., locating the local learning target and adjusting the local acceleration coefficient. Specifically, PSCN adequately considers the diversity to generate subpopulations located by hidden node centers, where each center is assessed by significance-based criteria and distance-based criteria. The former enable a small intrasubpopulation distance and a big search scope (subpopulation width) for each subpopulation, while the latter make each center distant from other existing centers. In each subpopulation, the best-found position is selected as the local learning target. In the output layer, PSCN determines the action of adjusting the local acceleration coefficient of each individual. Reinforcement learning is introduced to obtain the desired output of PSCN, enabling the network to control the search by learning in different iterations of each environment. The experimental results especially performance comparisons with eight state-of-the-art DOAs demonstrate that PSCN brings significant improvements in performance of solving DOPs.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2024-09-02DOI: 10.1109/TSMC.2024.3443860
Miao Guo;Bin Xin;Yipeng Wang;Jie Chen
This article focuses on the coalition formation (CF) problem in urgent missions, e.g., disaster rescue, where coalition members should reach mission locations quickly. A mathematical model is first constructed to minimize the latest arrival time of coalition members, considering the capability requirements of missions, nonredundant agents in coalitions, etc. Then, incorporating the benefits in both the diversity of random search and the effectiveness of utilizing problem knowledge, a local-search-based heuristic is put forward to solve the CF problem. An initial solution is incrementally constructed by prioritizing agents with shorter movement times for missions with higher-remaining capability requirements. Additionally, two types of neighborhood search operators, namely, the tabu-based one-to-one swap and the destroy and repair operators, are proposed to search the solution space from two perspectives, i.e., “adjustment” and “reconstruction.” To solve the problem effectively and efficiently, the former excludes certain agent-exchange combinations that do not improve the current solution, while the latter consists of multiple heuristic rules extracted from the correlation among different model elements. Experimental results have demonstrated that the proposed method surpasses several advanced methods across various scenarios regarding multiple factors, such as the number of agents, the number of missions, and the demand-supply ratio on capabilities.
{"title":"A Local-Search-Based Heuristic for Coalition Formation in Urgent Missions","authors":"Miao Guo;Bin Xin;Yipeng Wang;Jie Chen","doi":"10.1109/TSMC.2024.3443860","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3443860","url":null,"abstract":"This article focuses on the coalition formation (CF) problem in urgent missions, e.g., disaster rescue, where coalition members should reach mission locations quickly. A mathematical model is first constructed to minimize the latest arrival time of coalition members, considering the capability requirements of missions, nonredundant agents in coalitions, etc. Then, incorporating the benefits in both the diversity of random search and the effectiveness of utilizing problem knowledge, a local-search-based heuristic is put forward to solve the CF problem. An initial solution is incrementally constructed by prioritizing agents with shorter movement times for missions with higher-remaining capability requirements. Additionally, two types of neighborhood search operators, namely, the tabu-based one-to-one swap and the destroy and repair operators, are proposed to search the solution space from two perspectives, i.e., “adjustment” and “reconstruction.” To solve the problem effectively and efficiently, the former excludes certain agent-exchange combinations that do not improve the current solution, while the latter consists of multiple heuristic rules extracted from the correlation among different model elements. Experimental results have demonstrated that the proposed method surpasses several advanced methods across various scenarios regarding multiple factors, such as the number of agents, the number of missions, and the demand-supply ratio on capabilities.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2024-09-02DOI: 10.1109/TSMC.2024.3412954
Guang-Ren Duan
This article first studies the problem of replacing one single real open-loop eigenvalue in a multivariable linear system by state feedback, simultaneously achieving minimization of the sensitivity of this assigned eigenvalue. Two types of sensitivity indices of the assigned closed-loop eigenvalue corresponding, respectively, to the cases of structured and unstructured parameter perturbations are considered. It is shown that for this problem simple and neat analytical globally optimal solutions exist. By using the derived globally optimal solutions repeatedly, this article second proposes a circulation design for eigenstructure assignment in a stabilizable linear system via state feedback with low closed-loop eigenvalue sensitivities. As a consequence, in those rounds of replacing a real open-loop eigenvalue of order 1, the sensitivity index of the assigned closed-loop eigenvalue can be globally minimized. In the case that all the open-loop eigenvalues to be replaced are real ones of order 1, the proposed circulation design not only turns out to be extremely simple and efficient, but also possesses good numerical reliability because it removes completely matrix inverse operations. Two illustrative examples demonstrate the simplicity and effect of the proposed approach.
{"title":"Circulation Design for Eigenvalue Replacement: Minimizing Eigenvalue Sensitivities","authors":"Guang-Ren Duan","doi":"10.1109/TSMC.2024.3412954","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3412954","url":null,"abstract":"This article first studies the problem of replacing one single real open-loop eigenvalue in a multivariable linear system by state feedback, simultaneously achieving minimization of the sensitivity of this assigned eigenvalue. Two types of sensitivity indices of the assigned closed-loop eigenvalue corresponding, respectively, to the cases of structured and unstructured parameter perturbations are considered. It is shown that for this problem simple and neat analytical globally optimal solutions exist. By using the derived globally optimal solutions repeatedly, this article second proposes a circulation design for eigenstructure assignment in a stabilizable linear system via state feedback with low closed-loop eigenvalue sensitivities. As a consequence, in those rounds of replacing a real open-loop eigenvalue of order 1, the sensitivity index of the assigned closed-loop eigenvalue can be globally minimized. In the case that all the open-loop eigenvalues to be replaced are real ones of order 1, the proposed circulation design not only turns out to be extremely simple and efficient, but also possesses good numerical reliability because it removes completely matrix inverse operations. Two illustrative examples demonstrate the simplicity and effect of the proposed approach.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2024-08-30DOI: 10.1109/TSMC.2024.3446671
Jiewu Leng;Guolei Ruan;Caiyu Xu;Xueliang Zhou;Kailin Xu;Yan Qiao;Zhihong Liu;Qiang Liu
Mass individualized prototyping (MIP) is a kind of advanced and high-value-added manufacturing service. In the MIP context, the service providers usually receive massive individualized prototyping orders, and they should keep a stable state in the presence of continuous significant stresses or disruptions to maximize profit. This article proposed a graph convolutional neural network-based deep reinforcement learning (GCNN-DRL) method to achieve the resilient production control of MIP (RPC-MIP). The proposed method combines the excellent feature extraction ability of graph convolutional neural networks with the autonomous decision-making ability of deep reinforcement learning. First, a three-dimensional disjunctive graph is defined to model the RPC-MIP, and two dimensionality-reduction rules are proposed to reduce the dimensionality of the disjunctive graph. By extracting the features of the reduced-dimensional disjunctive graph through a graph isomorphic network, the convergence of the model is improved. Second, a two-stage control decision strategy is proposed in the DRL process to avoid poor solution quality in the large-scale searching space of the RPC-MIP. As a result, the high generalization capability and efficiency of the proposed GCNN-DRL method are obtained, which is verified by experiments. It could withstand system performance in the presence of continuous significant stresses of workpiece replenishment and also make fast rearrangement of dispatching decisions to achieve rapid recovery after disruptions happen in different production scenarios and system scales, thereby improving the system’s resilience.
{"title":"Deep Reinforcement Learning of Graph Convolutional Neural Network for Resilient Production Control of Mass Individualized Prototyping Toward Industry 5.0","authors":"Jiewu Leng;Guolei Ruan;Caiyu Xu;Xueliang Zhou;Kailin Xu;Yan Qiao;Zhihong Liu;Qiang Liu","doi":"10.1109/TSMC.2024.3446671","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3446671","url":null,"abstract":"Mass individualized prototyping (MIP) is a kind of advanced and high-value-added manufacturing service. In the MIP context, the service providers usually receive massive individualized prototyping orders, and they should keep a stable state in the presence of continuous significant stresses or disruptions to maximize profit. This article proposed a graph convolutional neural network-based deep reinforcement learning (GCNN-DRL) method to achieve the resilient production control of MIP (RPC-MIP). The proposed method combines the excellent feature extraction ability of graph convolutional neural networks with the autonomous decision-making ability of deep reinforcement learning. First, a three-dimensional disjunctive graph is defined to model the RPC-MIP, and two dimensionality-reduction rules are proposed to reduce the dimensionality of the disjunctive graph. By extracting the features of the reduced-dimensional disjunctive graph through a graph isomorphic network, the convergence of the model is improved. Second, a two-stage control decision strategy is proposed in the DRL process to avoid poor solution quality in the large-scale searching space of the RPC-MIP. As a result, the high generalization capability and efficiency of the proposed GCNN-DRL method are obtained, which is verified by experiments. It could withstand system performance in the presence of continuous significant stresses of workpiece replenishment and also make fast rearrangement of dispatching decisions to achieve rapid recovery after disruptions happen in different production scenarios and system scales, thereby improving the system’s resilience.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142442924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2024-08-30DOI: 10.1109/TSMC.2024.3446841
Lu Chen;Mingdi Niu;Jing Yang;Yuhua Qian;Zhuomao Li;Keqi Wang;Tao Yan;Panfeng Huang
Most available grasp detection methods tend to directly predict grasp configurations with deep neural networks, where all features are equally extracted and utilized, leading to the relative restriction of truly useful grasping features. Inspired by the observed three-section structure pattern revealed by human-labeled graspable rectangles, we first design a structure prior attention (SPA) module which uses two-dimensional encoding to enhance the local patterns and utilizes self-attention mechanism to reallocate distribution of grasping-specific features. Then, the proposed SPA module is integrated with fundamental feature extraction modules and residual connection to achieve the implicit and explicit feature fusion, which further serves as the building block of our proposed Unet-like grasp detection network. It takes RGBD images as input and outputs image-size feature maps, from which the grasp configurations can be determined. Extensive comparative experiments on the five public datasets prove our method’s superiority to other approaches in detection accuracy, achieving 99.2%, 96.1%, 98.0%, 86.7%, and 92.6% on the Cornell, Jacquard, Clutter, VMRD, and GraspNet datasets. With visual evaluation metrics and user study, the quality maps generated by our method possess more concentrative distribution of high-confidence grasps and clearer discrimination with backgrounds. In addition, its effectiveness is also verified by robotic grasping under real-world scenario, leading to higher success rate.
{"title":"Robotic Grasp Detection Using Structure Prior Attention and Multiscale Features","authors":"Lu Chen;Mingdi Niu;Jing Yang;Yuhua Qian;Zhuomao Li;Keqi Wang;Tao Yan;Panfeng Huang","doi":"10.1109/TSMC.2024.3446841","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3446841","url":null,"abstract":"Most available grasp detection methods tend to directly predict grasp configurations with deep neural networks, where all features are equally extracted and utilized, leading to the relative restriction of truly useful grasping features. Inspired by the observed three-section structure pattern revealed by human-labeled graspable rectangles, we first design a structure prior attention (SPA) module which uses two-dimensional encoding to enhance the local patterns and utilizes self-attention mechanism to reallocate distribution of grasping-specific features. Then, the proposed SPA module is integrated with fundamental feature extraction modules and residual connection to achieve the implicit and explicit feature fusion, which further serves as the building block of our proposed Unet-like grasp detection network. It takes RGBD images as input and outputs image-size feature maps, from which the grasp configurations can be determined. Extensive comparative experiments on the five public datasets prove our method’s superiority to other approaches in detection accuracy, achieving 99.2%, 96.1%, 98.0%, 86.7%, and 92.6% on the Cornell, Jacquard, Clutter, VMRD, and GraspNet datasets. With visual evaluation metrics and user study, the quality maps generated by our method possess more concentrative distribution of high-confidence grasps and clearer discrimination with backgrounds. In addition, its effectiveness is also verified by robotic grasping under real-world scenario, leading to higher success rate.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2024-08-29DOI: 10.1109/TSMC.2024.3444811
Xinyang Deng;Siyu Xue;Wen Jiang;Xiaoge Zhang
Measuring the uncertainty of information is a crucial problem in many fields. Recent studies have found a new uncertainty measure for probabilities called “extropy” as a complementary dual function of classical Shannon entropy. In this article, the extropy measure of randomness is generalized to the case of information with epistemic uncertainty by means of a framework of Dempster-Shafer evidence theory. Specifically, a novel measure called plausibility extropy is proposed, which inherits the intriguing properties of original extropy. Moreover, the duality and complementarity between the proposed plausibility extropy and existing plausibility entropy are proved strictly, which constitutes an entropy-extropy combination for mass functions to measure the epistemic uncertainty. In addition, the maximum plausibility extropy is also studied in this article. Through comparing with existing extropy-like measures in Dempster-Shafer evidence theory, the rationality of proposed plausibility extropy is further demonstrated.
{"title":"Plausibility Extropy: The Complementary Dual of Plausibility Entropy","authors":"Xinyang Deng;Siyu Xue;Wen Jiang;Xiaoge Zhang","doi":"10.1109/TSMC.2024.3444811","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3444811","url":null,"abstract":"Measuring the uncertainty of information is a crucial problem in many fields. Recent studies have found a new uncertainty measure for probabilities called “extropy” as a complementary dual function of classical Shannon entropy. In this article, the extropy measure of randomness is generalized to the case of information with epistemic uncertainty by means of a framework of Dempster-Shafer evidence theory. Specifically, a novel measure called plausibility extropy is proposed, which inherits the intriguing properties of original extropy. Moreover, the duality and complementarity between the proposed plausibility extropy and existing plausibility entropy are proved strictly, which constitutes an entropy-extropy combination for mass functions to measure the epistemic uncertainty. In addition, the maximum plausibility extropy is also studied in this article. Through comparing with existing extropy-like measures in Dempster-Shafer evidence theory, the rationality of proposed plausibility extropy is further demonstrated.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In a realistic two-sided assembly line, a preventive maintenance (PM) activity may cause a stoppage of the whole line and a waste of capacity in most stations. To promote production continuity, multiple interchangeable task assignment schemes are required, each targeting one of the regular and PM scenarios. Yet previous studies have not solved the resulting two-sided assembly line balancing problem considering PM scenarios (TALBP-PM), and the domain knowledge deserves extraction. Hence, a multiobjective mixed-integer linear programming model is formulated to minimize cycle times and total task adjustment simultaneously, and a knowledge-assisted variable neighborhood search (KVNS) is customized. Specifically, a decoding mechanism with idle time reduction is proposed to achieve schemes with the shortest cycle times. A rule-based initialization relying on the externalization of implicit relations among unique attributes is designed to derive a high-quality initial solution. Supported by the critical station and task knowledge, objective-oriented neighborhood structures are developed to generate neighbor solutions with increasingly better objectives. Besides, a restart operator adaptive to multidomain knowledge is refined to escape from local optima. Computational results show that the knowledge assistance is effective, and KVNS is superior to other state-of-the-art meta-heuristics in achieving well-converged and -distributed Pareto fronts of TALBP-PM.
{"title":"A Knowledge-Assisted Variable Neighborhood Search for Two-Sided Assembly Line Balancing Considering Preventive Maintenance Scenarios","authors":"Lianpeng Zhao;Qiuhua Tang;Zikai Zhang;Yingying Zhu","doi":"10.1109/TSMC.2024.3407724","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3407724","url":null,"abstract":"In a realistic two-sided assembly line, a preventive maintenance (PM) activity may cause a stoppage of the whole line and a waste of capacity in most stations. To promote production continuity, multiple interchangeable task assignment schemes are required, each targeting one of the regular and PM scenarios. Yet previous studies have not solved the resulting two-sided assembly line balancing problem considering PM scenarios (TALBP-PM), and the domain knowledge deserves extraction. Hence, a multiobjective mixed-integer linear programming model is formulated to minimize cycle times and total task adjustment simultaneously, and a knowledge-assisted variable neighborhood search (KVNS) is customized. Specifically, a decoding mechanism with idle time reduction is proposed to achieve schemes with the shortest cycle times. A rule-based initialization relying on the externalization of implicit relations among unique attributes is designed to derive a high-quality initial solution. Supported by the critical station and task knowledge, objective-oriented neighborhood structures are developed to generate neighbor solutions with increasingly better objectives. Besides, a restart operator adaptive to multidomain knowledge is refined to escape from local optima. Computational results show that the knowledge assistance is effective, and KVNS is superior to other state-of-the-art meta-heuristics in achieving well-converged and -distributed Pareto fronts of TALBP-PM.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2024-08-29DOI: 10.1109/TSMC.2024.3445109
Weiwei Zhan;Zhiqiang Miao;Hui Zhang;Yanjie Chen;Zheng-Guang Wu;Wei He;Yaonan Wang
This article presents a resilient formation control framework for networked nonholonomic mobile robots (NMRs) that enables long-time recovery abilities subject to denial-of-service (DoS) attacks by taking advantage of the Koopman operator. Due to the intermittent interruption of communication under DoS, the transmitted signals among the networked NMRs are incomplete. In the lifted space, the infinite-dimensional Koopman operator is employed to capture a linear characteristic of the missed signals from the available signals. Specifically, a data-driven cost function is developed to approximate the infinite-dimensional Koopman operator, allowing long-term recovery capabilities for the missed signals, where the useful historical data is identified by an event-triggered mechanism (ETM). Then, the least-squares method is implemented to calculate a finite-dimensional approximation of the Koopman operator. Once DoS attacks are active, the missed signals are recovered forward from the latest received signals through the approximation Koopman operator. Furthermore, according to the recovered and transmitted signals, the resilient formation controller with a variable gain takes into account the convergence rate and the steady state formation error. The Lyapunov theorem is introduced to prove that the formation error quickly converges to the minor compact set. A distributed DoS attack example is conducted to validate the efficiency and superiority in numerical simulation, and the proposed method is implemented on the real networked NMRs.
本文提出了一种用于网络化非全局移动机器人(NMR)的弹性编队控制框架,该框架利用库普曼算子(Koopman operator)的优势,可在受到拒绝服务(DoS)攻击时实现长时间恢复能力。由于在 DoS 攻击下通信会间歇性中断,联网的 NMR 之间传输的信号是不完整的。在提升空间中,采用无穷维 Koopman 算子从可用信号中捕捉遗漏信号的线性特征。具体来说,我们开发了一个数据驱动的成本函数来近似无穷维 Koopman 算子,从而实现对遗漏信号的长期恢复能力,其中有用的历史数据由事件触发机制 (ETM) 识别。然后,采用最小二乘法计算库普曼算子的有限维近似值。一旦 DoS 攻击激活,就会通过近似库普曼算子从最新接收到的信号中向前恢复错过的信号。此外,根据恢复和传输的信号,具有可变增益的弹性编队控制器会考虑收敛速率和稳态编队误差。通过引入 Lyapunov 定理,证明了编队误差会快速收敛到次要紧凑集。通过一个分布式 DoS 攻击实例验证了数值模拟的效率和优越性,并在实际网络化 NMR 上实现了所提出的方法。
{"title":"Resilient Formation Control With Koopman Operator for Networked NMRs Under Denial-of-Service Attacks","authors":"Weiwei Zhan;Zhiqiang Miao;Hui Zhang;Yanjie Chen;Zheng-Guang Wu;Wei He;Yaonan Wang","doi":"10.1109/TSMC.2024.3445109","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3445109","url":null,"abstract":"This article presents a resilient formation control framework for networked nonholonomic mobile robots (NMRs) that enables long-time recovery abilities subject to denial-of-service (DoS) attacks by taking advantage of the Koopman operator. Due to the intermittent interruption of communication under DoS, the transmitted signals among the networked NMRs are incomplete. In the lifted space, the infinite-dimensional Koopman operator is employed to capture a linear characteristic of the missed signals from the available signals. Specifically, a data-driven cost function is developed to approximate the infinite-dimensional Koopman operator, allowing long-term recovery capabilities for the missed signals, where the useful historical data is identified by an event-triggered mechanism (ETM). Then, the least-squares method is implemented to calculate a finite-dimensional approximation of the Koopman operator. Once DoS attacks are active, the missed signals are recovered forward from the latest received signals through the approximation Koopman operator. Furthermore, according to the recovered and transmitted signals, the resilient formation controller with a variable gain takes into account the convergence rate and the steady state formation error. The Lyapunov theorem is introduced to prove that the formation error quickly converges to the minor compact set. A distributed DoS attack example is conducted to validate the efficiency and superiority in numerical simulation, and the proposed method is implemented on the real networked NMRs.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To address responsiveness, time-dependence, and limited emergency supply issues, we introduce a new flexible districting policy, aiming to improve satisfaction in multiperiod emergency resource allocation (MPERA), and set demand priorities to guarantee allocation balance in resource-limited scenarios. The modeling and solution process involves the following: 1) formulating a mixed-integer programming (MILP) model for MPERA with demand priority (MPERA-DP), aiming to maximize utility considering the transportation cost, districting change, and penalty for unsatisfied demand and 2) incorporating the justifiable granularity principle (JGP) and particle swarm optimization (PSO) into the brand-and-price (B&P) algorithm for initial districting and allocating decisions to improve the solution quality and calculation speed. The results of the experiments show that 1) the JGP-PSO-B&P algorithm achieves superior efficiency in terms of optimality and convergence for large-scale cases. This algorithm could improve the optimality by 13.42% compared with that of the JGP-PSO algorithm, 13.15% compared with that of the B&P algorithm, and 28.18% compared with that of the PSO algorithm, on average; 2) the MPERA-DP model with flexible districting policy outperforms flexible MPERA without demand priority, emergency resource allocation with rescheduling (ERAR) and fixed emergency resource allocation with demand priority (FERA-DP), improving the utility by 20.56%, 5.14% and 41.84%, respectively; and 3) the scheme efficiency is influenced by the desirable satisfaction deviation, and when set to 0.6, it allows for the optimization of both demand satisfaction and utility.
{"title":"Flexible Districting Policy for the Multiperiod Emergency Resource Allocation Problem With Demand Priority","authors":"Xiaofeng Xu;Ziru Lin;Xiang Li;Wanli Yi;Witold Pedrycz","doi":"10.1109/TSMC.2024.3443116","DOIUrl":"https://doi.org/10.1109/TSMC.2024.3443116","url":null,"abstract":"To address responsiveness, time-dependence, and limited emergency supply issues, we introduce a new flexible districting policy, aiming to improve satisfaction in multiperiod emergency resource allocation (MPERA), and set demand priorities to guarantee allocation balance in resource-limited scenarios. The modeling and solution process involves the following: 1) formulating a mixed-integer programming (MILP) model for MPERA with demand priority (MPERA-DP), aiming to maximize utility considering the transportation cost, districting change, and penalty for unsatisfied demand and 2) incorporating the justifiable granularity principle (JGP) and particle swarm optimization (PSO) into the brand-and-price (B&P) algorithm for initial districting and allocating decisions to improve the solution quality and calculation speed. The results of the experiments show that 1) the JGP-PSO-B&P algorithm achieves superior efficiency in terms of optimality and convergence for large-scale cases. This algorithm could improve the optimality by 13.42% compared with that of the JGP-PSO algorithm, 13.15% compared with that of the B&P algorithm, and 28.18% compared with that of the PSO algorithm, on average; 2) the MPERA-DP model with flexible districting policy outperforms flexible MPERA without demand priority, emergency resource allocation with rescheduling (ERAR) and fixed emergency resource allocation with demand priority (FERA-DP), improving the utility by 20.56%, 5.14% and 41.84%, respectively; and 3) the scheme efficiency is influenced by the desirable satisfaction deviation, and when set to 0.6, it allows for the optimization of both demand satisfaction and utility.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":null,"pages":null},"PeriodicalIF":8.6,"publicationDate":"2024-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142443010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}