检索结果-内蒙古大学图书馆

Optimal consensus control for partially unknown multi-agent systems with switching communication topology*

IFAC-PapersOnLine 2023年第2期56卷 1210-1216页

作者： Wenpeng He Xin Chen Menglin Zhang Yipu Sun Akinori Sekiguchi Jinhua She School of Automation China University of Geosciences Wuhan 430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan 430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan 430074 China School of Engineering Tokyo University of Technology Hachioji Tokyo 192-0982 Japan

Since the local neighborhood tracking error system for switching multi-agent systems (MASs) is time-varying, the existing optimal consensus method for fixed MASs cannot be used in switching MASs. This paper proposes a new approach to solve the optimal consensus problems of linear switching MASs. Firstly, a new augment local neighborhood tracking error (ALNTE) system, which is based on the joint communication topology, is developed to construct the control object, and the open-loop estimation method is used to estimate the ALNTE for individuals. Secondly, based on the ALNTE system, the performance index is designed, and the value iteration method is used to solve the coupled Hamilton-Jacobi-Bellman equations and obtain the optimal control. Finally, theoretical analysis is given to prove the learning stability and control stability of the value iteration method, and a numerical simulation is used to verify the effectiveness of the proposed method.

关键词： Switching topology multi-agent systems (MASs) optimal control value iteration augment local neighborhood tracking error (ALNTE)

来源：评论

学校读者我要写书评

暂无评论

An Attention Detection system Based on Gaze Estimation Using Self-supervised Learning 8th

An Attention Detection System Based on Gaze Estimation Usin...

引用

8th International Workshop on advanced Computational Intelligence and intelligent Informatics, IWACIII 2023

作者： Zeng, Xiang-Yu Zhang, Bo-Yang Liu, Zhen-Tao School of Future Technology China University of Geosciences Wuhan430074 China School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

ISBN: (纸本)9789819975891

With the rapid development of online education around world, it becomes difficult for teachers to evaluate the effect of learning during the online class. To solve this problem, an attention detection system based on gaze estimation using self-supervised learning is proposed. Self-supervised learning is used to train the gaze estimation network and it achieved an average gaze angle error of 6. 5∘ on the MPIIFaceGaze dataset. Cross dataset testing was performed by 5 test subjects in the RT-GENE dataset and all samples in the Columbia dataset, in which achieved average gaze angle error 13. 8∘ and 5. 5∘. It is found that the network has good cross dataset generalization performance, but it is susceptible to factors such as feature occlusion and low image clarity. Then, we developed an attention detection system which has functions of unmanned detection, eye state detection, and gaze detection. Online classroom experiments were conducted to verify the effectiveness of the proposed method, in which the average accuracy of the detection of distraction is around 60%. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Supervised learning

来源：评论

学校读者我要写书评

暂无评论

Cascade conditional generative adversarial nets for spatial-spectral hyperspectral sample generation

引用

Science China(Information Sciences) 2020年第4期63卷 81-96页

作者： Xiaobo LIU Yulin QIAO Yonghua XIONG Zhihua CAI Peng LIU School of Automation China University of Geosciences (Wuhan) Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems School of Computer Science China University of Geosciences (Wuhan) The Beibu Gulf Big Data Resources Utilization Laboratory Beibu Gulf University

Sample generation is an effective way to solve the problem of the insufficiency of training data for hyperspectral image classification. The generative adversarial network（GAN） is one of the popular deep learning methods, which utilizes adversarial training to generate the region of samples based on the required class label. In this paper, we propose cascade conditional generative adversarial nets for hyperspectral image complete spatial-spectral sample generation, named C;GAN. The C;GAN includes two stages. The stageone model consists of the spatial information generation with a window size that entails feeding random noise and the required class label. The second stage is the spatial-spectral information generation that generates spectral information of all bands in the spatial region by feeding the label regions. The visualization and verification of generated samples based on the Pavia University and Salinas datasets show superior performance, which demonstrates that our method is useful for hyperspectral image classification.

关键词： hyperspectral image sample generation generative adversarial nets(GAN) deep learning spatial-spectral information

来源：评论

学校读者我要写书评

暂无评论

Disturbance Observer-Based control of Weight on Bit in Drilling Process with Formation Variation

Disturbance Observer-Based Control of Weight on Bit in Drill...

引用

Asian control Conference

作者： Shipeng Chen Wangnian Li Chengda Lu Yang Zhou Weihua Cao Min Wu School of Automation China University of Geosciences Wuhan China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan China CCTEG Xi'an Research Institute (Group) Co. Ltd. Xi'an China

ISBN: (数字)9789887581598

ISBN: (纸本)9798331540845

In deep geological drilling, encountering various formations is inevitable. Formation changes can lead to fluc-tuations in the weight on bit due to the bit-rock interaction. This paper proposes a disturbance observer-based weight on bit control system to mitigate the adverse effects of formation changes on system performance. First, an axis-torsion coupled dynamic model is established, and the relationship between the two dimensions is analyzed. The validity of the model is confirmed through comparison with field data collected from a geothermal well drilling operation. Then, a disturbance observer is designed to estimate and compensate for the disturbance at the control input, thereby reducing weight on bit fluctuations caused by formation variations. Simulation results considering formation variation during drilling verify the effectiveness of the proposed method.

关键词： Drilling Simulation system performance Geology Process control control systems Disturbance observers

来源：评论

学校读者我要写书评

暂无评论

A NOVEL ARTIFICIAL POTENTIAL FIELD APPROACH CONSIDERING ANGULAR ACCELERATION CONSTRAINT

引用

International Journal of Robotics and automation 2024年第1期39卷 24-30页

作者： Sun, Mingxiao Yuan, Zhangjie Zhang, Yongde Zhen, Liqiang Luan, Tiantian Yuan, Xiaoliang Li, Xiaogang School of Automation Harbin University of Science and Technology No. 52 Xuefu Road Nangang District Heilongjiang Province Harbin City150080 China Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration Harbin University of Science and Technology No. 52 Xuefu Road Nangang District Heilongjiang Province Harbin City150080 China Key Laboratory of Advanced Manufacturing and Intelligent Technology Ministry of Education Harbin University of Science and Technology No. 52 Xuefu Road Nangang District Heilongjiang Province Harbin City150080 China Wuxi CREG Urban Rail Transit Equipment Co. Ltd. No. 1061 Jiaoshan Road Xishan Development Zone Jiangsu Province Wuxi City214000 China

An improved artificial potential field approach (APFA) is proposed for unmanned surface vehicle (USV) in path planning. The proposed approach can effectively avoid easily falling into local minima, also, plan the optimal path. Firstly, a map of the Marine environment is established, and the obstacle is treated with circular expansion. Especially, the expansion coefficient is introduced to ensure the safety of navigation. Secondly, aiming at the local minimum problem, a rotation angle formula satisfying its own maximum rotation angle and angular acceleration constraints is constructed, so that a path to avoid local minimum is planned. Moreover, the local minimum of the complex environment is considered, the virtual repulsive force potential field is constructed to generate virtual repulsive force on USV to avoid local minimum. Numerical simulation results show that the proposed approach can complete path planning under different complex and special environments. © 2024 Acta Press. All rights reserved.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Vision-Based Autonomous Landing of a Quadrotor UAV on a Double-Layered Nested Landing Marker

Vision-Based Autonomous Landing of a Quadrotor UAV on a Doub...

引用

International Conference on Guidance, Navigation and control, ICGNC 2022

作者： Mu, Lingxia Zhang, Yuan Xin, Jing Zhang, Youmin School of Automation and Information Engineering Xi’an University of Technology Xi’an710048 China Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi’an University of Technology Xi’an710048 China Department of Mechanical and Industrial Engineering Concordia University MontrealH3G 1M8 Canada

ISBN: (纸本)9789811966125

This paper focuses on the vision-based autonomous landing mission of a quadrotor unmanned aerial vehicle (UAV). A double-layered nested Aruco landing marker is designed which can adapt to the situation that the field of sight of airborne camera changes as the UAV’s altitude decreases during the landing. Then, considering that the landing marker might be obstructed in certain situation, traditional Aruco detection method cannot effectively recognize marker. A YOLOv4 deep learning algorithm is used for marker detection. In addition, to address the problem of false detection in some test results, normalized Wasserstein distance is used to replace the non-maximum suppression to deal with redundant frames. The UAV’s relative location to the marker is then calculated and landing mission is finally achieved by the image-based visual servoing strategy. The simulation on Gazebo demonstrates that the designed marker is effective for the UAV landing mission and the revised YOLOv4 detection algorithm can increase detection results accuracy. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

A Trajectory Generation Method Based on control Points and Improved Whale Algorithm Optimization Algorithm for Solving Unmanned Aerial Vehicle Path Planning Problems

SSRN

引用

SSRN 2023年

作者： Li, Bingyu Gan, Chao Zhu, Yifeng Cao, Wei-Hua Wuhan430074 China Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

Three-dimensional trajectory planning is crucial for improving the safety and operational efficiency of unmanned aerial vehicle(UAV). However, it is hard to plan a suitable path in complex real-world environments. To overcome this challenge, this paper introduces a novel trajectory planning method based on control point modelling and an advanced whale optimisation algorithm (WOA). Firstly, a trajectory modeling method based on control point interpolation algorithm was proposed, which constrains the one-dimensional dimensions of 3D points, reducing the complexity of the traditional modelling methods and increasing the efficiency of trajectory generation. After that, an advanced adaptive multi-population cooperative whale optimisation algorithm (AMCWOA) has been proposed to enhance the problem-solving capabilities of the algorithm in complex environments. Through a population evolution stage evaluation strategy (ESE), this algorithm dynamically divides the population size, thereby balancing the problem-solving breadth and precision of the algorithm. It increases algorithm accuracy and convergence speed in solving trajectory planning problems. It is imperative to emphasis that the evaluation strategy in the evolution stage purposefully addresses the blindness of multi-population evolution algorithms when dividing populations. Finally, multiple function sets are utilized to test the solving performance of the algorithm and the simulations in multiple environments are designed to demonstrate the performance in UAV path planning. The experimental comparison results demonstrate that the proposed population evolution stage evaluation strategy (ESE) and population information sharing strategy can effectively increase the search range and convergence accuracy of the algorithm in generating high-quality paths and optimising test functions. © 2023, The Authors. All rights reserved.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

A Reinforcement Learning Based Dynamic Multi-Objective Constrained Evolutionary Algorithm for Open-Pit Mine Truck Scheduling

A Reinforcement Learning Based Dynamic Multi-Objective Const...

引用

Chinese automation Congress (CAC)

作者： Junxiang Qiu Changhe Li Shengxiang Yang Engineering Research Center of Intelligent Technology for Geo- Exploration Ministry of Education School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China School of Computer Science and Informatics De Montfort University Leicester UK

Aiming at the truck scheduling problem in the open-pit mine scenario, a truck scheduling model based on real-time ore blending is established, and an adaptive evolution algorithm for truck scheduling based on DCNSGA-III is proposed. In the established scheduling model, the real-time grade variance of the crushing plant is minimized as one of the optimization objectives, and the Q-learning algorithm is introduced to adaptively select one of the most effective operators during the search process. Experiments show that the proposed method can effectively control the grade fluctuation of the ore flow and better scheduling schemes are obtained in comparison with algorithms equipped with the traditional search operator selection methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Multi-modal Panoramic Speaker Localization Method

A Multi-modal Panoramic Speaker Localization Method

引用

2021 China automation Congress, CAC 2021

作者： Long, Y.Z. Jian, Y.W. Liu, Z.T. School of Automation China University of Geosciences WuHan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems WuHan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

ISBN: (纸本)9781665426473

As a supporting technology in the field of human-computer interaction, speaker localization method has been a research hotspot in recent years. However, the existing single-mode speaker location methods cannot meet the requirements of stability and rapidity. For improve the accuracy and effective positioning range of speaker localization system, a multi-modal panoramic speaker localization method is proposed in this paper. Firstly, the coarse position of the speaker was preliminarily obtained by sound source localization. Then, according to the initial positioning of the speaker, whether the image used for face recognition need to be stitched is decided. What's more, the position of speaker is calculated by face detection and calibration. Finally, the results of voice and image localization are fused by coordinate transformation. The proposal is tested on self-built experimental system, and the experimental results prove the effectiveness of the proposed method. © 2021 IEEE

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Research on Path Planning of Mobile Robot Based on DWA-IMP-A* Algorithm

Research on Path Planning of Mobile Robot Based on DWA-IMP-A...

引用

Chinese control Conference (CCC)

作者： Zhi Liu Lin Zhang Shoukun Wang Tianwei Niu Yongkang Xu Jin Cao Key Laboratory of Intelligent Control and Decision of Complex Systems School of Automation Beijing Institute of Technology Beijing China Key Laboratory of Servo Motion System Drive and Control Ministry of Industry and Information Technology School of Automation Beijing Institute of Technology Beijing China

Path planning is one of the most critical links in mobile robots. Its timeliness, security and accessibility are crucial to the development and wide application of mobile robots. However, in solving the problem of path planning, the most popular A* algorithm has some problems, such as heuristic function cannot be estimated accurately, node redundancy, path is not smooth, and obstacle avoidance cannot be achieved in real time. To solve these problems, A fusion algorithm of improved A* combined with reverse path and dynamic window method (DWA-IMP-A*) was proposed. The algorithm refines the heuristic function by incorporating the reverse path. The node optimization algorithm is used to further reduce the path length. The generated trajectories are smoothed by cubic spline interpolation. At the same time, it is integrated with the improved DWA algorithm to improve the efficiency and safety of robot path planning. The algorithm takes ROS mobile robot as the carrier and is tested under typical road conditions. Compared with A* algorithm, the planning time is reduced by 54.6% and the path length is reduced by 6.37%. Experimental results verify the effectiveness and robustness of the algorithm. The research results have certain reference significance for the path planning of various types of mobile robots and the research of driverless vehicles.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：