Please wait a minute...
Most Download
Most Cited
Total visitors:
Visitors of today:
Now online:

Current Issue

    15 September 2022, Volume 4 Issue 3
    Review Intelligence
    Parallel Yuan-Ming Yuan Imperial Garden: from digital twin garden to metaverse smart heritage park
    Mengzhen KANG, Wenzhong QIU, Zifu CHEN, Meng WANG, Shasha XU, Xiujuan WANG, Aidong NI, Yujie JIANG, Shichao CHEN, Philippe DEREFFYE, Fei-Yue WANG
    2022, 4(3):  301-307.  doi:10.11959/j.issn.2096-6652.202247
    Asbtract ( 544 )   HTML ( 61)   PDF (8115KB) ( 285 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Yuan-Ming Yuan Imperial Garden is a historically royal garden; it not only occupies an important position in the history of Chinese garden, but also enjoys a high reputation in the world.The historical and cultural values contained in Yuan-Ming Yuan Imperial Garden needs to be widely understood by the Chinese people and be remembered by the world through a new way.Focusing on the work policy for cultural relics of “protection first, strict management, mining value, rational utilization and good transmission”, a new solution of parallel Yuan-Ming Yuan Imperial Garden was proposed, which provided technical reference for the construction of the smart Yuan-Ming Yuan Imperial Garden.Parallel Yuan-Ming Yuan Imperial Garden is the application of ACP theory in the operation and management.Descriptive intelligence will be used to construct a virtual Yuan-Ming Yuan Imperial Garden, predictive intelligence will be used to conduct large-scale computational experiments in the virtual Yuan-Ming Yuan Imperial Garden, and prescriptive intelligence and parallel execution will be used to outbreak the geographical limit and lead to the smart management of Yuan-Ming Yuan Imperial Garden.It is expected that the development and enrichment of the Yuan-Ming Yuan Imperial Garden in the virtual world, and the increasingly parallel interaction and integration with the real world, will bring about a new operating mode.

    Surveys and Prospectives
    A review of continual learning for robotics
    Chao ZHAO, Jie XU, Xingyu CHEN, Kuizhi MEI, Xuguang LAN
    2022, 4(3):  308-323.  doi:10.11959/j.issn.2096-6652.202235
    Asbtract ( 542 )   HTML ( 116)   PDF (1356KB) ( 439 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    One of the limitations of robotics is that it is difficult for robots to adapt to fickle tasks.A robot will inevitably forget the knowledge from old environments or tasks when facing new environments or tasks.In order to summarize research in continual learning for robotics, firstly, the framework and evaluation protocols in continual learning were introduced.And then necessity and challenge of continual learning in the robotics were expounded.The research for continual learning was also summarized.Finally, the prospect of continual learning was predicted and some valuable research directions were put forward.

    Artificial intelligence technologies and applications in the metaverse
    Qiang WU, Xueting JI, Linyuan LYU
    2022, 4(3):  324-334.  doi:10.11959/j.issn.2096-6652.202241
    Asbtract ( 1572 )   HTML ( 293)   PDF (872KB) ( 1228 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Metaverse integrates and applies a variety of digital technologies, resulting in an Internet social form that integrates virtual and reality, digital and application.Artificial intelligence (AI) are systems and machine that imitate human intelligence to perform tasks and iteratively improve themselves based on the information gathered.In the process of constructing the metaverse, AI technologies not only vigorously promote the development of crucial metaverse technologies (human-computer interaction, communication, robotics, etc.) but also enables direct content creation in the metaverse, organically connecting the real and virtual worlds.By sorting out the concepts and representative technologies of the metaverse and AI, the optimization progress of AI technologies for constructing key technologies in the metaverse was introduced and the application process of AI in the metaverse was detailed.Finally, the AI technology’s development and application trends in the metaverse prospected.

    Overview of metro train driving technology development:from manual driving to intelligent unmanned driving
    Wenzhu LAI, Dewang CHEN, Zhenfeng HE, Xinguo DENG, CARLO Marano GIUSEPPE
    2022, 4(3):  335-343.  doi:10.11959/j.issn.2096-6652.202204
    Asbtract ( 608 )   HTML ( 121)   PDF (647KB) ( 485 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Based on the current development status of subway train driving technology in China and abroad, the four stages of subway train driving technology development were proposed and explained as manual driving, automatic driving, unmanned driving, and intelligent unmanned driving.After summarizing the construction situation of unmanned subway trains in China, the disadvantages of the current train control methods based on neural network-based machine learning methods were addressed.Then, the basic block diagram of metro intelligent unmanned driving based on man-machine hybrid intelligence was put forward and the deep fuzzy system was introduced.A promising solution for the combination of expert experience in dealing with emergency and interpretable AI algorithms for unmanned driving system to evolve into intelligent unmanned driving was provided.

    Papers and Reports
    Architecture and key techniques of parallel creation through the fusion of human-cyber-physical intelligence in CPSS
    Chao GUO, Yue LU, Xiao WANG, Da YI, Xiao WANG, Fei-Yue WANG
    2022, 4(3):  344-354.  doi:10.11959/j.issn.2096-6652.202246
    Asbtract ( 252 )   HTML ( 35)   PDF (3555KB) ( 291 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the expansion of the fields covered by AI, artistic creation will become the next hot spot for AI research and applications.Building a metaverse with diverse styles, realistic contents, flexible strokes and accurate descriptions based on parallel system theory and ACP approach will provide a feasible way to improve AI creation capability.The intelligence of human, AI, and robots were fused to develop a parallel creation architecture through the creation by AI, the evaluation by humans, and the execution by robots.The parallel creation with the key methods of style transfer, content combination, stroke generation and image captioning in computational experiments were explained.The parallel creation system was validated through the painting experiments.The parallel creation system will improve the creation capability of artificial intelligence in cyberspace and physical space, and promote the human-cyber-physical collaborative creation through the fusion of them.

    Knowledge-driven order allocation method for raw material supply chain in metallurgical enterprises
    Yishun LIU, Chunhua YANG, Keke HUANG
    2022, 4(3):  355-370.  doi:10.11959/j.issn.2096-6652.202238
    Asbtract ( 218 )   HTML ( 20)   PDF (765KB) ( 321 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The raw material supply chain is the premise of safe and stable production with an important position.As the front link of the supply chain, order allocation is the focus of enterprises.Due to the complex composition of raw materials, large qualitative differences in suppliers, and information coupling in the metallurgical industry, the current expert decision-making model is labor-intensive and difficult to deal with such complex order allocation problems, resulting in low decision-making efficiency, high procurement costs, and difficult to guarantee the quality of raw materials.Aiming at this problem, a knowledge-driven order allocation method for the raw material supply chain in metallurgical enterprises was proposed.First, on the basis of a multi-level supplier evaluation system, the entropy weight method and the fuzzy analytic hierarchy process were adopted to make full use of data knowledge and experience knowledge, and compatibility degree and different degree were introduced to reasonably allocate the importance of each evaluation index.Then, a multi-attribute decision-making evaluation model was built based on the technique for order preference by similarity to an ideal solution (TOPSIS) to automatically obtain the comprehensive performance and ranking of suppliers, so as to realize the efficient evaluation and management of suppliers.Finally, a multi-objective order quantity allocation model was established by comprehensively considering the supplier characteristics knowledge, ore blending mechanism knowledge, business status knowledge, etc., and the optimal order quantity of the each supplier under complex resource constraints was automatically obtained.Taking a domestic zinc smelting enterprise as an example, the validity and applicability of the proposed method are verified by the relevant data of the raw material supply chain.The results show that the proposed method can automatically complete knowledge-based work such as supplier evaluation and order quantity allocation, which will greatly liberate manual labor, improve decision-making efficiency, effectively reduce procurement cost and improve the quality of raw materials.

    A constructive approach to ultrastable systems based on the self-awareness and self-expression architecture
    Junyi WEN, Xiang YU, Bifei MAO, Jinghui LI, Xin YAO
    2022, 4(3):  371-379.  doi:10.11959/j.issn.2096-6652.202245
    Asbtract ( 169 )   HTML ( 17)   PDF (924KB) ( 84 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Abstarct: An ultrastable system can adaptively return to the stable state when the complex environment changes.W.Ross.Ashby has made the homeostat to prove the existence of the ultrastable system.However, there is no methodological guidance on how to construct anultrastable system in engineering practice.A constructive approach to ultrastable systemswas presented.Based on the principles of self-awareness and self-expression, a constructive framework of the ultrastable system was proposed, and a feasible theoretical framework and methodology for designing the ultrastable system were established.The framework makes the ultrastable system easier to understand and optimize.The simulation results on the homeostat, an ultrastable system, show the effectiveness of the proposed framework.

    Short-term traffic state reasoning and precise prediction in urban networks
    Yuanqi QIN, Qingyuan JI, Jun GE, Xingyuan DAI, Yuanyuan CHEN, Xiao WANG
    2022, 4(3):  380-395.  doi:10.11959/j.issn.2096-6652.202233
    Asbtract ( 302 )   HTML ( 38)   PDF (1718KB) ( 380 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    The structure of urban traffic network has a significant impact on the formation and spatio-temporal pattern propagation of traffic congestions.However, in studies based on traditional traffic models or deep learning models, the generation of traffic mode can only be described indirectly by traffic indicators, without considering the traffic network feature.This makes it very difficult to accurately describe the propagation dynamics both in temporal and spatial dimensions and lacks specificity.To tackle the above-mentioned problems, a novel traffic state prediction approach based on traffic pattern reasoning (TP2) framework was proposed.The framework modeled congestion propagation as a dynamically evolving temporal knowledge graph (TKG), and applied an inferencing framework (TPP-TKG) that was based on a novel aggregator called RGraAN.TPP-TKG captured the spatial-temporal propagation pattern of traffic congestion, and combined related road links to a given link, and constructed correlated sub region of the traffic network.Then a traffic state predicting based on graph neural network was employed to predict short-term speed evolution of road links in this sub region.Comparing to the state-of-the-art benchmark models, TP2 achieves 1% ~ 2% higher accuracy.

    A metaverse-oriented digital citizen authentication scheme
    Qiuyun LYU, Shaopeng CHENG, Manzhi YANG, Xiaoguang CHEN, Zhen WANG
    2022, 4(3):  396-409.  doi:10.11959/j.issn.2096-6652.202239
    Asbtract ( 396 )   HTML ( 33)   PDF (1172KB) ( 255 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    As a surreal ternary world, the metaverse provides people with massive and diverse network services.However, due to the lack of face-to-face communication in reality and the broken geographical and time constraints, the existing schemes cannot effectively authenticate digital citizens while realizing privacy protection and accountability.Therefore, a metaverse-oriented digital citizen authentication scheme was proposed.Firstly, a layered authentication scheme based on the time-sensitive identity was constructed.Secondly, for protecting citizens’ privacy, the Paillier homomorphic encryption algorithm and structure-preserving signature scheme were introduced to realize the double unlink ability of decentralized identifiers and verifiable credentials, in addition, the citizens’ right to be forgotten was guaranteed by the code embedding technology.Thirdly, the regulatory factor during the authentication phase for accountability was designed.Finally, the security analysis proves that the proposed scheme satisfies the unlinkability and traceability.And, the performance analysis shows that the proposed scheme has lower computational cost and gas overhead, which can meet the needs of practical use.

    Dynamic configuration of distribution network based on improved hierarchical clustering and GL-APSO algorithm
    Yun WANG, Meiyun WANG, Jian ZHOU, Yuanyuan ZOU, Shaoyuan LI
    2022, 4(3):  410-417.  doi:10.11959/j.issn.2096-6652.202243
    Asbtract ( 161 )   HTML ( 15)   PDF (773KB) ( 64 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the problem of dynamic reconfiguration of distribution network with distributed generation (DG), a dynamic distribution networks reconfiguration scheme considering the time-varying property of DG and distribution network load was proposed.Firstly, according to the comprehensive similarity between different periods based on both load characteristics and optimal network structure, an improved hierarchical clustering method was used to divide the reconstruction interval into segments.On this basis, the genetic learning adaptive particle swarm optimization algorithm was proposed to realize the dynamic reconstruction with minimum network loss.To tackle the shortcomings such as the lack of speed dynamic adjustment strategy and ease to fall into local optimum in basic particle swarm optimization algorithm, a genetic learning scheme based on the optimal position of individual particles was proposed to enhance diversity and improve global search ability.Adaptive inertia weight and acceleration coefficients were introduced to meet the optimization requirements of different periods.Finally, a simulation was carried out through the IEEE 33-bus distribution system as an example to verify the effectiveness and superiority of the proposed method.

    A path planning method for complex naval battle field based on an improved DQN algorithm
    Zhou YU, Jing BI, Haitao YUAN
    2022, 4(3):  418-425.  doi:10.11959/j.issn.2096-6652.202244
    Asbtract ( 350 )   HTML ( 79)   PDF (1972KB) ( 183 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    To solve a target tracking problem of multiple warships in a sea battlefield environment, the multiple agents (warships) were focused on, and an improved deep Q-network (DQN) algorithm was proposed.It considers the characteristics of a multi-agent reinforcement learning environment based on a traditional DQN algorithm.It adds a network with the same structure and different parameters and updates a Q actual value and a Q estimated one, respectively to realize convergence of a value function.Besides, it adopts a mechanism of experience playback and an update of one of two parameters for a target network to effectively solve problems of high training errors of neural networks, poor generalization ability, and unstable training.Experimental results demonstrate that compared with the traditional DQN algorithm, the improved DQN one adapts to different complex and dynamical sea battlefield environments in a faster manner, and the ability to avoid obstacles has more than doubled and larger training reward.

    HVAC model-free optimal control method based on double-pools DQN
    Shuai MA, Qiming FU, Jianping CHEN, Fan FENG, You LU, Zhengwei LI, Shunian QIU
    2022, 4(3):  426-444.  doi:10.11959/j.issn.2096-6652.202208
    Asbtract ( 231 )   HTML ( 43)   PDF (2690KB) ( 278 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    In the field of HVAC (heating, ventilation and air conditioning) control, the model-based optimal control method has been extensively studied and verified by scholars, but this method highly depends on the accuracy of the model, the collection of a large amount of historical data, and the deployment of sensors.In response to the above problems,combined with EnergyPlus, actual system parameters and historical data, the HVAC optimized control model was constructed, and an improved double pools-based DQN (DPs-DQN) algorithm was proposed.Finally, it was applied to the load distribution of different types of chillers, the combined optimal control of cooling tower fan frequency and cooling water pump frequency in HVAC system.Based on the constructed problem model, aiming at the problem of sample imbalance in the decision-making optimization process, the algorithm established two independent experience pools on the basis of DQN to store load distribution and non load distribution samples respectively.During the training process, followed a certain ratio to sample from the experience pool to speed up the algorithm convergence.The proposed method was compared with the model-based control method and the baseline method.The experimental results show that compared with the baseline method, the model-based HVAC controller can save 11.5% (optimal energy-saving efficiency), while the DPs-DQN can save energy by 7.5% in the first year.At the same time, as the system runs, the controller can obtain results close to the optimal energy saving efficiency in the eighth year.In addition, compared with the model-based HVAC controller, the controller does not depend on the system model, and requires less prior knowledge and sensors in the online control process, which is more valuable in actual engineering applications.

    Basic framework and key technologies of parallel tires
    Xiangwen ZHANG, Fei-Yue WANG
    2022, 4(3):  445-457.  doi:10.11959/j.issn.2096-6652.202242
    Asbtract ( 246 )   HTML ( 31)   PDF (11711KB) ( 100 )   Knowledge map   
    Figures and Tables | References | Related Articles | Metrics

    With the development of intelligent vehicles and connected vehicles, intelligent tires are required to provide more tire state information.In order to make comprehensive use of the state information of the tires, a framework of parallel tires was established based on the theory of parallel systems to control and manage the tired state intelligently, which improved the safety, economy, and comfort of the tires correspondingly.In view of the established framework of parallel tires, the realization process of the key technologies was described in detail, including artificial tire modeling, artificial tires computational experiments, and parallel execution between artificial tires and real tires.In addition, the application fields of parallel tires prospected, and the knowledge automation development direction of tire technology was discussed accordingly.