加载中…【当期目录】IEEE/CAAJAS第11卷第6期
(2024-10-08 15:48:11)
标签：
it |
「主题」
故障诊断、神经网络、深度学习、进化算法、滑模控制、多智能体系统、自适应控制、非线性系统、缺陷检测、无人机、强化学习、多目标优化、路径跟踪...
「全球科研机构」
美国University of Texas at Arlington、Missouri University of Science and Technology、Florida State University、New Jersey Institute of Technology；英国Brunel University London、The University of Nottingham；澳大利亚Swinburne University of Technology；加拿大University of Victoria；新加坡Nanyang Technology University；日本University of Toyama；清华大学、北京大学、中国科学院大学、中国人民大学、同济大学、东北大学、北京航空航天大学、西安交通大学、中南大学、中国石油大学、厦门大学、南京邮电大学、北京化工大学、中国移动研究院...
T. Shen, J. Sun, S. Kong, Y. Wang, J. Li, X. Li, and F.-Y.
Wang,
> A novel kernel framework of embodied intelligence.
> Correspondence between embodied intelligence and parallel intelligence.
> Journey/DAO/TAO and prospects of embodied intelligence.
J. Ren, J. Wen, Z. Zhao, R. Yan, X. Chen, and
> Multivariate Gaussian distribution is employed into the deep architecture and some methods of representing diversity are combined to quantify both aleatoric and epistemic uncertainties simultaneously.
> On the basis of uncertainty decomposition graph given by our simplified algorithm, we proposed a unified trustworthy fault diagnosis framework, named as UU-DLF.
> Through the powerful performance and promising prospects shown by experiments, the uncertainty of deep learning proposed by our paper may provide a promising way for deep learning models to gain industrial users’ trust.
Z. Zhang, S. Gao, M. C. Zhou, M. Yan, and
> Proposes a novel deep learning method for solving turbulence prediction.
> Proposed method can extract the spatial-temporal feature in the turbulence well.
> Proposed method outperforms other state-of-the-art methods.
S. Qi, R. Wang, T. Zhang, X. Yang, R. Sun, and
> Discussed the limitations of the prevailing two-layer encoding SLMOEAs.
> Propose an innovative masking learning (ML) strategy based on the frequent itemsets.
> Propose a dynamic mutation strategy for binary variables suitable for ML.
Z. Wu, Y. Zhao, F. Li, T. Yang, Y. Shi, and
> Arbitrary distribution of sojourn time and asynchronous strategy are applied to controller design.
> Chattering of traditional sliding mode control is alleviated by a novel iterative learning term.
> Sliding surface synthesis under incompletely available state information is presented.
J. Li, R. Qin, S. Guan, W. Ding, F. Lin, and
> Design attention markets based on the Haberger Tax for DAOs.
> Formulate the Stackelberg game model in DAOs’ attention markets to capture the dynamics of attention trading.
> Discuss the equilibrium trading strategies to explore the attention pricing mechanisms among DAO members.
Z. Zuo, J. Tang, R. Ke, and
> New hyperbolic tangent function-based protocols for achieving finite-time consensus in multi-agent systems, offering a user-prescribed bounded control level with an explicit estimate of the settling time bound.
> Simplified design and stability analysis surpassing traditional saturation function and homogeneity theory approaches.
> Proven effectiveness and practicability via illustrative examples and applications in multi-BLDC motor systems.
L. Feng, B. Huang, J. Sun, Q. Sun, and
> Studies a multilevel multigroup problem where agents are categorised into tracking leaders, formation leaders, and followers.
> Investigates the time-varying output group formation containment problem for general linear heterogeneous multi-agent systems under a directed topology, where the agents have different characteristics and dynamic models.
> Control protocol proposed is fully distributed and adopts new dynamic event-triggered mechanisms to efficiently save communication resources, and the Zeno behavior is ruled out by giving positive lower bounds of the inter-event intervals of two subsequences.
X. Tan, C. Hu, G. Cao, Q. Wei, W. Li, and
> For MASs with dynamics in the form of strict feedback affine
nonlinearity, a DFTO is developed for each follower to obtain an
estimate of the leader's output by utilizing the topology of the
communication network.
> A FTDO is given to estimate the lumped disturbances for feedforward compensation.
> A nonsmooth antidisturbance backstepping tracking controller with feedforward compensation for lumped disturbances is designed base on the proposed DFTO and FTDO.
Q. Ge, Y. Cheng, H. Li, Z. Ye, Y. Zhu, and
> On the condition of unknown data distribution, a Parzen window
method on the sliding window technology is adopted for the data PDF
rough estimation.
> With the shape difference between a Gaussian-like PDF curve and a Gaussian PDF curve, the intelligent optimization algorithm adjusts mean and variance to analyse the difference between two curve shapes (slope, kullback-leibler divergence).
> Multi-index evaluation based on the method of weighted mean is proposed to improve the correct recognition rate between Gaussian-like PDF curve and standard Gaussian PDF curve.
Y. Li, X. Wang, Z. He, Z. Wang, K. Cheng, S.
> Proposes a novel defect detection method for complex PCBAs which is geared towards actual production lines.
> A semantic segmentation model combined with computer vision
methods is proposed to build up a defect detection
framework.
> Proposed method can significantly improve the throughput of the PCBA production line with a lower false-call rate and missing detection rate.
Q. Ma, P. Jin, and
> Safety, robustness, and optimality are important, but
sometimes they are conflicting criteria in controller
design.
> A simple control barrier function and a special disturbance attenuation function are designed to characterize the violations of safety constraints and tolerance of uncertain disturbances.
> A new weight update law for critic neural networks is designed by combining the concurrent learning technique with gradient descent method, which enhances the utilization rate of system data and relaxes the requirement for persistent excitation condition.
J. Liang, H. Lin, C. Yue, P. Suganthan, and
> A new framework of MMO test functions is proposed, incorporating various components with different functions to control convergence difficulty, multimodal characteristics, and the extensibility of decision variables.
> Fifteen test instances are generated, allowing for a more comprehensive evaluation of the performance of MMOEAs.
> A new MMOEA is developed, called HDMMODE.
Z. Chen, J. Tang, and
> A novel paradigm termed the PPPF problem is introduced.
> A guiding vector field adhering to prescribed-performance criteria is advanced to facilitate adaptable path following tasks.
> Leveraging the prescribed-time sliding mode control approach, a novel path-following control framework is posited to propel a controlled vehicle towards specific segments of a designated trajectory.
Y. Zhao, X. He, M. Zhou, and
> Proposed APDPNA-S with fast convergence rates.
> APDPNA-S can handle convex optimization problems with linear and set constraints, providing wider applicability.
> Extended APDPNA-S to differential inclusion dynamical approach (APDPNA-NS) with same results.
S. Shi and
> An ASE framework for motion planning which first focuses on exploring in a small promising region and then explores the outward expanded space.
> A particular hyper-ellipsoid ring expansion strategy such that
the algorithm can also avoid repeatedly exploring the same
space.
> An adaptive division strategy divides a global hyper-ellipsoid into several small local hyper-ellipsoids.
C. Liu, W. Liu, T. Wei, and X. Li,
L. Xue, J. Ye, Y. Wu, J. Liu, and D. C. Wunsch, “Prescribed-time
Nash equilibrium seeking for pursuit-evasion
game,”
M. Yang,
S. Wu,
J. Liu, S. Li, and R. Liu,
M. Deng and C. Wang,
F. Yan, X. Liu, and T. Feng,
P. Wu, H. Li, L. Hu, J. Ge, and N.
Zeng,
喜欢
0
赠金笔