Pub Date : 2024-03-03DOI: 10.26599/TST.2024.9010035
Guodong Xu;Hai Wang;Shuo Ji;Yuhui Ma;Yi Feng
Earthen ruins contain rich historical value. Affected by wind speed, temperature, and other factors, their survival conditions are not optimistic. Time series prediction provides more information for ruins protection. This work includes two challenges: (1) The ruin is located in an open environment, causing complex nonlinear temporal patterns. Furthermore, the usual wind speed monitoring requires the 10 meters observation height to reduce the influence of terrain. However, in order to monitor wind speed around the ruin, we have to set 4.5 meters observation height according to the ruin, resulting in a non-periodic and oscillating temporal pattern of wind speed; (2) The ruin is located in the arid and uninhabited region of northwest China, which results in accelerating aging of equipment and difficulty in maintenance. It significantly amplifies the device error rate, leading to duplication, missing, and outliers in datasets. To address these challenges, we designed a complete preprocessing and a Transformer-based multi-channel patch model. Experimental results on four datasets that we collected show that our model outperforms the others. Ruins climate prediction model can timely and effectively predict the abnormal state of the environment of the ruins. This provides effective data support and decision-making for ruins conservation, and exploring the relationship between the environmental conditions and the living state of the earthen ruins.
{"title":"MPformer: A Transformer-Based Model for Earthen Ruins Climate Prediction","authors":"Guodong Xu;Hai Wang;Shuo Ji;Yuhui Ma;Yi Feng","doi":"10.26599/TST.2024.9010035","DOIUrl":"https://doi.org/10.26599/TST.2024.9010035","url":null,"abstract":"Earthen ruins contain rich historical value. Affected by wind speed, temperature, and other factors, their survival conditions are not optimistic. Time series prediction provides more information for ruins protection. This work includes two challenges: (1) The ruin is located in an open environment, causing complex nonlinear temporal patterns. Furthermore, the usual wind speed monitoring requires the 10 meters observation height to reduce the influence of terrain. However, in order to monitor wind speed around the ruin, we have to set 4.5 meters observation height according to the ruin, resulting in a non-periodic and oscillating temporal pattern of wind speed; (2) The ruin is located in the arid and uninhabited region of northwest China, which results in accelerating aging of equipment and difficulty in maintenance. It significantly amplifies the device error rate, leading to duplication, missing, and outliers in datasets. To address these challenges, we designed a complete preprocessing and a Transformer-based multi-channel patch model. Experimental results on four datasets that we collected show that our model outperforms the others. Ruins climate prediction model can timely and effectively predict the abnormal state of the environment of the ruins. This provides effective data support and decision-making for ruins conservation, and exploring the relationship between the environmental conditions and the living state of the earthen ruins.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 6","pages":"1829-1838"},"PeriodicalIF":6.6,"publicationDate":"2024-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10520028","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141435226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the coal mining process, a large amount of Coal Mine-Associated energy (CMAE), such as coal mine methane and underground wastewater, is produced. Research on the modeling and optimization dispatching of a Coal Mine-Integrated Energy System (CMIES) with CMAE effectively saves energy and reduces carbon pollution. CMAE has great uncertainties owing to the affections of the hydrogeology conditions and mining schedules. In addition, thermal loads have high comfort requirements in mines, which brings great challenges to the optimization dispatching of CMIESs. Therefore, this paper studies the architecture and solution of CMIESs with a flexible thermal load and source-load uncertainty. First, to effectively improve the electric and thermal conversion efficiency, the architecture of CMIES, including a concentrating solar power station, is built. Second, for the scheduling model with bilateral uncertainty, the interval representation method with interval variables is proposed, and a multi-objective scheduling model based on the interval variables and flexible thermal load is constructed. Finally, we propose a solution method for the model with interval variables. A case study is conducted to demonstrate the performance of our model and method for lowering carbon emissions and cost.
{"title":"Dispatch of a Coal Mine-Integrated Energy System: Optimization Model with Interval Variables and Lower Carbon Emission","authors":"Hejuan Hu;Xiaoyan Sun;Bo Zeng;Dunwei Gong;Yong Zhang;Patrick Nyonganyi;Henerica Tazvinga","doi":"10.26599/TST.2023.9010110","DOIUrl":"https://doi.org/10.26599/TST.2023.9010110","url":null,"abstract":"In the coal mining process, a large amount of Coal Mine-Associated energy (CMAE), such as coal mine methane and underground wastewater, is produced. Research on the modeling and optimization dispatching of a Coal Mine-Integrated Energy System (CMIES) with CMAE effectively saves energy and reduces carbon pollution. CMAE has great uncertainties owing to the affections of the hydrogeology conditions and mining schedules. In addition, thermal loads have high comfort requirements in mines, which brings great challenges to the optimization dispatching of CMIESs. Therefore, this paper studies the architecture and solution of CMIESs with a flexible thermal load and source-load uncertainty. First, to effectively improve the electric and thermal conversion efficiency, the architecture of CMIES, including a concentrating solar power station, is built. Second, for the scheduling model with bilateral uncertainty, the interval representation method with interval variables is proposed, and a multi-objective scheduling model based on the interval variables and flexible thermal load is constructed. Finally, we propose a solution method for the model with interval variables. A case study is conducted to demonstrate the performance of our model and method for lowering carbon emissions and cost.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1441-1462"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517925","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.26599/TST.2023.9010087
Hongyan Sang;Zhongkai Li;M. Fatih Tasgetiren
Automated Guided Vehicle (AGV) scheduling problem is an emerging research topic in the recent literature. This paper studies an integrated scheduling problem comprising task assignment and path planning for AGVs. To reduce the transportation cost of AGVs, this work also proposes an optimization method consisting of the total running distance, total delay time, and machine loss cost of AGVs. A mathematical model is formulated for the problem at hand, along with an improved Discrete Invasive Weed Optimization algorithm (DIWO). In the proposed DIWO algorithm, an insertion-based local search operator is developed to improve the local search ability of the algorithm. A staggered time departure heuristic is also proposed to reduce the number of AGV collisions in path planning. Comprehensive experiments are conducted, and 100 instances from actual factories have proven the effectiveness of the optimization method.
{"title":"An Effective Optimization Method for Integrated Scheduling of Multiple Automated Guided Vehicle Problems","authors":"Hongyan Sang;Zhongkai Li;M. Fatih Tasgetiren","doi":"10.26599/TST.2023.9010087","DOIUrl":"https://doi.org/10.26599/TST.2023.9010087","url":null,"abstract":"Automated Guided Vehicle (AGV) scheduling problem is an emerging research topic in the recent literature. This paper studies an integrated scheduling problem comprising task assignment and path planning for AGVs. To reduce the transportation cost of AGVs, this work also proposes an optimization method consisting of the total running distance, total delay time, and machine loss cost of AGVs. A mathematical model is formulated for the problem at hand, along with an improved Discrete Invasive Weed Optimization algorithm (DIWO). In the proposed DIWO algorithm, an insertion-based local search operator is developed to improve the local search ability of the algorithm. A staggered time departure heuristic is also proposed to reduce the number of AGV collisions in path planning. Comprehensive experiments are conducted, and 100 instances from actual factories have proven the effectiveness of the optimization method.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1355-1367"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517928","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.26599/TST.2023.9010101
Tianyu Luo;Yong Heng;Lining Xing;Teng Ren;Qi Li;Hu Qin;Yizhi Hou;Kesheng Wang
An Electric Vehicle (EV) is an appropriate substitution for traditional transportation means for diminishing greenhouse gas emissions. However, decision-makers are beset by the limited driving range caused by the low battery capacity and the long recharging time. To resolve the former issue, several transportation companies increases the travel distance of the EV by establishing recharging stations in various locations. The proposed Electric Vehicle-Routing Problem with Time Windows (E-VRPTW) and recharging stations are constructed in this context; it augments the VRPTW by reinforcing battery capacity constraints. Meanwhile, super-recharging stations are gradually emerging in the surroundings. They can decrease the recharging time for an EV but consume more energy than regular stations. In this paper, we first extend the E-VRPRTW by adding the elements of super-recharging stations. We then apply a two-stage heuristic algorithm driven by a dynamic programming process to solve the new proposed problem to minimize the travel and total recharging costs. Subsequently, we compare the experimental results of this approach with other algorithms on several sets of benchmark instances. Furthermore, we analyze the impact of super-recharging stations on the total cost of the logistic plan.
{"title":"A Two-Stage Approach for Electric Vehicle Routing Problem with Time Windows and Heterogeneous Recharging Stations","authors":"Tianyu Luo;Yong Heng;Lining Xing;Teng Ren;Qi Li;Hu Qin;Yizhi Hou;Kesheng Wang","doi":"10.26599/TST.2023.9010101","DOIUrl":"https://doi.org/10.26599/TST.2023.9010101","url":null,"abstract":"An Electric Vehicle (EV) is an appropriate substitution for traditional transportation means for diminishing greenhouse gas emissions. However, decision-makers are beset by the limited driving range caused by the low battery capacity and the long recharging time. To resolve the former issue, several transportation companies increases the travel distance of the EV by establishing recharging stations in various locations. The proposed Electric Vehicle-Routing Problem with Time Windows (E-VRPTW) and recharging stations are constructed in this context; it augments the VRPTW by reinforcing battery capacity constraints. Meanwhile, super-recharging stations are gradually emerging in the surroundings. They can decrease the recharging time for an EV but consume more energy than regular stations. In this paper, we first extend the E-VRPRTW by adding the elements of super-recharging stations. We then apply a two-stage heuristic algorithm driven by a dynamic programming process to solve the new proposed problem to minimize the travel and total recharging costs. Subsequently, we compare the experimental results of this approach with other algorithms on several sets of benchmark instances. Furthermore, we analyze the impact of super-recharging stations on the total cost of the logistic plan.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1300-1322"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517978","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.26599/TST.2023.9010127
Zidong Wu;Linglong Dai
To meet the ever-increasing demand for the data rates of wireless communications, extremely largescale antenna array (ELAA) has emerged as one of the candidate technologies for future 6G communications. The significantly increased number of antennas in ELAA gives rise to near-field communications, necessitating tailored beamforming techniques within the near-field regions to accommodate the spherical-wave propagation characteristics. Among various array geometries of ELAA, uniform circular array (UCA) has gained much attention for its distinct capability of maintaining uniform beam pattern across different azimuth angles. However, existing analysis of near-field UCA beamforming indicates that the near-field region severely declines in the broadside of UCA, where the system fails to benefit from near-field communications. To tackle this problem, the near-field beamforming technique of uniform concentric circular arrays (UCCAs) is investigated in this paper, which has the potential to enlarge the near-field region in the broadside direction. First, the analysis of beamforming gain in the 3D space with UCA and UCCA is provided. Then, the distinct beamforming characteristics that set UCCA apart from UCA are delineated, revealing the superiority of UCCA in extending the near-field region in broadside at the cost of slightly reduced near-field region in the coplane. Simulation results are provided to verify the effectiveness of the theoretical analysis of beamforming gain with UCCA and the enhanced focusing ability of UCCA in the broadside direction.
为满足对无线通信数据传输速率日益增长的需求，超大规模天线阵列（ELAA）已成为未来 6G 通信的候选技术之一。ELAA 中天线数量的大幅增加带来了近场通信，因此需要在近场区域采用定制的波束成形技术，以适应球形波的传播特性。在 ELAA 的各种阵列几何结构中，均匀圆形阵列（UCA）因其在不同方位角保持均匀波束模式的独特能力而备受关注。然而，现有的近场 UCA 波束成形分析表明，UCA 宽边的近场区域严重衰减，系统无法从近场通信中获益。针对这一问题，本文研究了均匀同心圆阵列（UCCA）的近场波束成形技术，该技术有可能扩大宽边方向的近场区域。首先，分析了 UCA 和 UCCA 在三维空间中的波束成形增益。然后，分析了 UCCA 与 UCA 不同的波束成形特性，揭示了 UCCA 在扩大宽侧近场区域方面的优势，但代价是略微缩小了共面近场区域。仿真结果验证了 UCCA 波束成形增益理论分析的有效性，以及 UCCA 在宽边方向增强聚焦能力的效果。
{"title":"Focusing Ability Enhancement in Broadside Direction of Array: From UCA to UCCA","authors":"Zidong Wu;Linglong Dai","doi":"10.26599/TST.2023.9010127","DOIUrl":"https://doi.org/10.26599/TST.2023.9010127","url":null,"abstract":"To meet the ever-increasing demand for the data rates of wireless communications, extremely largescale antenna array (ELAA) has emerged as one of the candidate technologies for future 6G communications. The significantly increased number of antennas in ELAA gives rise to near-field communications, necessitating tailored beamforming techniques within the near-field regions to accommodate the spherical-wave propagation characteristics. Among various array geometries of ELAA, uniform circular array (UCA) has gained much attention for its distinct capability of maintaining uniform beam pattern across different azimuth angles. However, existing analysis of near-field UCA beamforming indicates that the near-field region severely declines in the broadside of UCA, where the system fails to benefit from near-field communications. To tackle this problem, the near-field beamforming technique of uniform concentric circular arrays (UCCAs) is investigated in this paper, which has the potential to enlarge the near-field region in the broadside direction. First, the analysis of beamforming gain in the 3D space with UCA and UCCA is provided. Then, the distinct beamforming characteristics that set UCCA apart from UCA are delineated, revealing the superiority of UCCA in extending the near-field region in broadside at the cost of slightly reduced near-field region in the coplane. Simulation results are provided to verify the effectiveness of the theoretical analysis of beamforming gain with UCCA and the enhanced focusing ability of UCCA in the broadside direction.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1593-1603"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517927","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Knowledge representation learning (KRL) aims to encode entities and relationships in various knowledge graphs into low-dimensional continuous vectors. It is popularly used in knowledge graph completion (or link prediction) tasks. Translation-based knowledge representation learning methods perform well in knowledge graph completion (KGC). However, the translation principles adopted by these methods are too strict and cannot model complex entities and relationships (i.e., N-1, 1-N, and N-N) well. Besides, these traditional translation principles are primarily used in static knowledge graphs and overlook the temporal properties of triplet facts. Therefore, we propose a temporal knowledge graph embedding model based on variable translation (TKGE-VT). The model proposes a new variable translation principle, which enables flexible transformation between entities and relationship embedding. Meanwhile, this paper considers the temporal properties of both entities and relationships and applies the proposed principle of variable translation to temporal knowledge graphs. We conduct link prediction and triplet classification experiments on four benchmark datasets: WN11, WN18, FB13, and FB15K. Our model outperforms baseline models on multiple evaluation metrics according to the experimental results.
{"title":"A Temporal Knowledge Graph Embedding Model Based on Variable Translation","authors":"Yadan Han;Guangquan Lu;Shichao Zhang;Liang Zhang;Cuifang Zou;Guoqiu Wen","doi":"10.26599/TST.2023.9010142","DOIUrl":"https://doi.org/10.26599/TST.2023.9010142","url":null,"abstract":"Knowledge representation learning (KRL) aims to encode entities and relationships in various knowledge graphs into low-dimensional continuous vectors. It is popularly used in knowledge graph completion (or link prediction) tasks. Translation-based knowledge representation learning methods perform well in knowledge graph completion (KGC). However, the translation principles adopted by these methods are too strict and cannot model complex entities and relationships (i.e., N-1, 1-N, and N-N) well. Besides, these traditional translation principles are primarily used in static knowledge graphs and overlook the temporal properties of triplet facts. Therefore, we propose a temporal knowledge graph embedding model based on variable translation (TKGE-VT). The model proposes a new variable translation principle, which enables flexible transformation between entities and relationship embedding. Meanwhile, this paper considers the temporal properties of both entities and relationships and applies the proposed principle of variable translation to temporal knowledge graphs. We conduct link prediction and triplet classification experiments on four benchmark datasets: WN11, WN18, FB13, and FB15K. Our model outperforms baseline models on multiple evaluation metrics according to the experimental results.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1554-1565"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517975","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.26599/TST.2023.9010141
Xixing Li;Qingqing Zhao;Hongtao Tang;Xing Guo;Mengzhen Zhuang;Yibing Li;Xi Vincent Wang
To obtain a suitable scheduling scheme in an effective time range, the minimum completion time is taken as the objective of Flexible Job Shop scheduling Problems (FJSP) with different scales, and Composite Dispatching Rules (CDRs) are applied to generate feasible solutions. Firstly, the binary tree coding method is adopted, and the constructed function set is normalized. Secondly, a CDR mining approach based on an Improved Genetic Programming Algorithm (IGPA) is designed. Two population initialization methods are introduced to enrich the initial population, and a superior and inferior population separation strategy is designed to improve the global search ability of the algorithm. At the same time, two individual mutation methods are introduced to improve the algorithm's local search ability, to achieve the balance between global search and local search. In addition, the effectiveness of the IGPA and the superiority of CDRs are verified through comparative analysis. Finally, Deep Reinforcement Learning (DRL) is employed to solve the FJSP by incorporating the CDRs as the action set, the selection times are counted to further verify the superiority of CDRs.
{"title":"Flexible Job Shop Composite Dispatching Rule Mining Approach Based on an Improved Genetic Programming Algorithm","authors":"Xixing Li;Qingqing Zhao;Hongtao Tang;Xing Guo;Mengzhen Zhuang;Yibing Li;Xi Vincent Wang","doi":"10.26599/TST.2023.9010141","DOIUrl":"https://doi.org/10.26599/TST.2023.9010141","url":null,"abstract":"To obtain a suitable scheduling scheme in an effective time range, the minimum completion time is taken as the objective of Flexible Job Shop scheduling Problems (FJSP) with different scales, and Composite Dispatching Rules (CDRs) are applied to generate feasible solutions. Firstly, the binary tree coding method is adopted, and the constructed function set is normalized. Secondly, a CDR mining approach based on an Improved Genetic Programming Algorithm (IGPA) is designed. Two population initialization methods are introduced to enrich the initial population, and a superior and inferior population separation strategy is designed to improve the global search ability of the algorithm. At the same time, two individual mutation methods are introduced to improve the algorithm's local search ability, to achieve the balance between global search and local search. In addition, the effectiveness of the IGPA and the superiority of CDRs are verified through comparative analysis. Finally, Deep Reinforcement Learning (DRL) is employed to solve the FJSP by incorporating the CDRs as the action set, the selection times are counted to further verify the superiority of CDRs.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1390-1408"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517917","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.26599/TST.2023.9010075
Nianbo Kang;Zhonghua Miao;Quan-Ke Pan;Weimin Li;M. Fatih Tasgetiren
With the emergence of the artificial intelligence era, all kinds of robots are traditionally used in agricultural production. However, studies concerning the robot task assignment problem in the agriculture field, which is closely related to the cost and efficiency of a smart farm, are limited. Therefore, a Multi-Weeding Robot Task Assignment (MWRTA) problem is addressed in this paper to minimize the maximum completion time and residual herbicide. A mathematical model is set up, and a Multi-Objective Teaching-Learning-Based Optimization (MOTLBO) algorithm is presented to solve the problem. In the MOTLBO algorithm, a heuristic-based initialization comprising an improved Nawaz Enscore, and Ham (NEH) heuristic and maximum load-based heuristic is used to generate an initial population with a high level of quality and diversity. An effective teaching-learning-based optimization process is designed with a dynamic grouping mechanism and a redefined individual updating rule. A multi-neighborhood-based local search strategy is provided to balance the exploitation and exploration of the algorithm. Finally, a comprehensive experiment is conducted to compare the proposed algorithm with several state-of-the-art algorithms in the literature. Experimental results demonstrate the significant superiority of the proposed algorithm for solving the problem under consideration.
随着人工智能时代的到来，各种机器人被广泛应用于农业生产。然而，农业领域的机器人任务分配问题与智能农场的成本和效率密切相关，相关研究却十分有限。因此，本文探讨了一个多播种机器人任务分配（MWRTA）问题，以最小化最大完成时间和残留除草剂。本文建立了一个数学模型，并提出了一种基于多目标教学学习的优化算法（MOTLBO）来解决该问题。在 MOTLBO 算法中，采用了一种基于启发式的初始化方法，包括改进的 Nawaz Enscore, and Ham (NEH) 启发式和基于最大负荷的启发式，以生成一个具有高质量和多样性的初始种群。通过动态分组机制和重新定义的个体更新规则，设计了一个有效的基于教学的优化过程。提供了一种基于多邻域的局部搜索策略，以平衡算法的开发和探索。最后，我们进行了一项综合实验，将所提出的算法与文献中几种最先进的算法进行了比较。实验结果表明，所提出的算法在解决所考虑的问题方面具有明显的优越性。
{"title":"Multi-Objective Teaching-Learning-Based Optimizer for a Multi-Weeding Robot Task Assignment Problem","authors":"Nianbo Kang;Zhonghua Miao;Quan-Ke Pan;Weimin Li;M. Fatih Tasgetiren","doi":"10.26599/TST.2023.9010075","DOIUrl":"https://doi.org/10.26599/TST.2023.9010075","url":null,"abstract":"With the emergence of the artificial intelligence era, all kinds of robots are traditionally used in agricultural production. However, studies concerning the robot task assignment problem in the agriculture field, which is closely related to the cost and efficiency of a smart farm, are limited. Therefore, a Multi-Weeding Robot Task Assignment (MWRTA) problem is addressed in this paper to minimize the maximum completion time and residual herbicide. A mathematical model is set up, and a Multi-Objective Teaching-Learning-Based Optimization (MOTLBO) algorithm is presented to solve the problem. In the MOTLBO algorithm, a heuristic-based initialization comprising an improved Nawaz Enscore, and Ham (NEH) heuristic and maximum load-based heuristic is used to generate an initial population with a high level of quality and diversity. An effective teaching-learning-based optimization process is designed with a dynamic grouping mechanism and a redefined individual updating rule. A multi-neighborhood-based local search strategy is provided to balance the exploitation and exploration of the algorithm. Finally, a comprehensive experiment is conducted to compare the proposed algorithm with several state-of-the-art algorithms in the literature. Experimental results demonstrate the significant superiority of the proposed algorithm for solving the problem under consideration.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1249-1265"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517982","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820355","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.26599/TST.2024.9010011
Shuangshuang Wu;Zhiming Li;Wenbai Chen;Fuchun Sun
Learning the accurate dynamics of robotic systems directly from the trajectory data is currently a prominent research focus. Recent physics-enforced networks, exemplified by Hamiltonian neural networks and Lagrangian neural networks, demonstrate proficiency in modeling ideal physical systems, but face limitations when applied to systems with uncertain non-conservative dynamics due to the inherent constraints of the conservation laws foundation. In this paper, we present a novel augmented deep Lagrangian network, which seamlessly integrates a deep Lagrangian network with a standard deep network. This fusion aims to effectively model uncertainties that surpass the limitations of conventional Lagrangian mechanics. The proposed network is applied to learn inverse dynamics model of two multi-degree manipulators including a 6-dof UR-5 robot and a 7-dof SARCOS manipulator under uncertainties. The experimental results clearly demonstrate that our approach exhibits superior modeling precision and enhanced physical credibility.
{"title":"Dynamic Modeling of Robotic Manipulator via an Augmented Deep Lagrangian Network","authors":"Shuangshuang Wu;Zhiming Li;Wenbai Chen;Fuchun Sun","doi":"10.26599/TST.2024.9010011","DOIUrl":"https://doi.org/10.26599/TST.2024.9010011","url":null,"abstract":"Learning the accurate dynamics of robotic systems directly from the trajectory data is currently a prominent research focus. Recent physics-enforced networks, exemplified by Hamiltonian neural networks and Lagrangian neural networks, demonstrate proficiency in modeling ideal physical systems, but face limitations when applied to systems with uncertain non-conservative dynamics due to the inherent constraints of the conservation laws foundation. In this paper, we present a novel augmented deep Lagrangian network, which seamlessly integrates a deep Lagrangian network with a standard deep network. This fusion aims to effectively model uncertainties that surpass the limitations of conventional Lagrangian mechanics. The proposed network is applied to learn inverse dynamics model of two multi-degree manipulators including a 6-dof UR-5 robot and a 7-dof SARCOS manipulator under uncertainties. The experimental results clearly demonstrate that our approach exhibits superior modeling precision and enhanced physical credibility.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"29 5","pages":"1604-1614"},"PeriodicalIF":6.6,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10517980","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140820287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","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：