检索结果-内蒙古大学图书馆

4th International Conference on Machine Learning and intelligent Communications, MLICOM 2019

作者： Ahmad, Zohaib Yang, Cuili Qiao, Junfei Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing100124 China

ISBN: (纸本)9783030323875

In this paper, an accelerated particle swarm optimization (APSO) based radial basis function neural network (RBFNN) is designed for nonlinear system modeling. In APSO-RBFNN, the center, width of hidden neurons, weights of output layer and network size are optimized by using the APSO method. Two nonlinear system modeling experiments are used to illustrate the effectiveness of the proposed method. The simulation results show that the proposed method has obtained good performance in terms of network size and estimation accuracy. © 2019, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

关键词： Network layers

来源：评论

学校读者我要写书评

暂无评论

AGV multi-objective path planning method based on improved cuckoo algorithm 4

AGV multi-objective path planning method based on improved c...

引用

4th IEEE Advanced Information Technology, Electronic and Automation Control Conference, IAEAC 2019

作者： Yu, Jianjun Wang, Yang Ruan, Xiaogang Zuo, Guoyu Li, Chen Beijing University of Technology Faculty of Information Technology China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing China

ISBN: (纸本)9781728119076

In order to save time and reduce the energy loss of the AGV when transporting the objects, the discrete cuckoo algorithm is used to solve the path planning problem of multi targets points. Firstly, the cuckoo algorithm is applied to the path planning of the AGV single target point based on the grip map, and the optimal path of the AGV to each object and the optimal path of each object to the specified position are obtained. Secondly, discretizing the cuckoo algorithm, and in the process of discretization, multiple operators such as reverse sequence, insertion, substitution and double bridge are combined into different neighborhood structures, which will used to find the optimal solution through Levi flight mechanism. Finally, the discrete cuckoo algorithm is used to generate the global optimal path for AGV to transport multiple objects. Simulation results show that the proposed algorithm successfully solves the multi-target point path planning problem of AGV. Compared with other algorithms, the proposed algorithm can avoid partial optimality and achieve fast convergence speed. © 2019 IEEE.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Learning Dense Visual Object Descriptors to Fold Two-Dimensional Deformable Fabrics 13

Learning Dense Visual Object Descriptors to Fold Two-Dimensi...

引用

13th IEEE International Conference on CYBER Technology in Automation, Control, and intelligent systems, CYBER 2023

作者： Cao, Yang Gong, Daoxiong Yu, Jianjun Beijing University of Technology Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing100124 China

ISBN: (纸本)9798350315196

Manipulating two-dimensional fabrics is a significant research field in recent years. Fabric manipulation presents a formidable challenge due to the intricate dynamics and high-dimensional state space inherent in the process, prior research has predominantly relied on robot learning of task-specific strategies as a means to effectively accomplish the corresponding fabric manipulation tasks. In this work, we utilize dense visual object descriptors trained on synthetic RGB images to learn visual representations for two-dimensional fabrics. Based on the learned descriptors, the robot can learn the correspondences of similar fabrics in different configurations. We apply a novel Siamese network architecture to improve the quality of learned descriptors for three types of fabrics, including square fabrics, T-shirts and shorts. By utilizing the learned descriptors, the equivalent actions in an unknown configuration can be computed based on a fabric folding demonstration in an initial configuration. We perform a series of fabric folding tasks in different colors, sizes and shapes. The policy can achieve 87.7% average task success rate across 7 different folding tasks. © 2023 IEEE.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Research on Forecasting Method for Effluent Ammonia Nitrogen Concentration Based on GRA-TCN 3

Research on Forecasting Method for Effluent Ammonia Nitrogen...

引用

3rd International Conference on Industrial Artificial intelligence, IAI 2021

作者： Kang, Li Cui-Li, Yang Jun-Fei, Qiao Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Faculty of Information Technology Beijing100124 China

ISBN: (纸本)9781665435178

Aiming at the characteristics of high coupling degree, strong nonlinearity, and serious time delay in the measurement of ammonia nitrogen concentration in wastewater treatment process (WWTP), a prediction model of ammonia nitrogen concentration based on gray relational analysis (GRA) and time convolution network (TCN) was proposed. Firstly, based on the relevant water quality parameters collected in WWTP, the grey correlation analysis method was used to find out other characteristic variables highly related to the ammonia nitrogen concentration. Then, a new group of multivariate time series data was constructed by using the sliding window method. Finally, based on the advantages of the time convolution network in processing time series data, such as simple, flexible, and easy to parallel, the constructed time series data were modeled to predict the concentration of effluent ammonia-nitrogen. To verify the validity of the model, the predicted results were compared with the other four models. The experimental results show that the ammonia-nitrogen concentration prediction model based on GRA and TCN has good prediction performance, which is helpful to realize the accurate prediction of effluent ammonia-nitrogen concentration. At the same time, it can also provide timely and effective guidance for the control and optimization of the wastewater treatment process. © 2021 IEEE.

关键词： Effluents

来源：评论

学校读者我要写书评

暂无评论

A Geometric Motion Mapping Algorithm from Master Human Arm to Heterogeneous Slave Robot Arm with Offset Joints 11

A Geometric Motion Mapping Algorithm from Master Human Arm t...

引用

11th IEEE Annual International Conference on CYBER Technology in Automation, Control, and intelligent systems, CYBER 2021

作者： Gong, Daoxiong Wang, Ruihua Yu, Jianjun Beijing University of Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Faculty of Information Technology Beijing100124 China

ISBN: (纸本)9781665425278

In this paper, we presented a geometric motion mapping algorithm that translates the motor commands issued by the human master arm to a heterogeneous slave robot manipulator with offset joints for intuitive teleoperation. Firstly, we set up two spherical coordinate systems on the shoulder of the master tele-operator and the slave robotic arm, respectively. Then, we proposed a motion mapping algorithm comprising three phases of computations, i.e., to determine the positions of the wrist joints of the robot arm according to the target position and orientation of the slave end-effector (EE), to determine the configuration of the robotic arm for reaching the target position of the first wrist joint of the slave arm, and to determine the rotation angles of the wrist joints for realizing the expected position/orientation of the slave EE. Finally, we validate the proposed algorithm via simulations with the KIT human motion capture data. The proposed algorithm can achieve excellent motion mapping results, i.e., exactly attain the expected position/orientation of the slave EE and at the same time possess similar master/slave elbow motion. Therefore, the studied motion mapping algorithm is able to make the master human operator feeling natural and comfortable for intuitive and efficient teleoperation. © 2021 IEEE.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

Missing-Pattern-Aware-Based Multi-Model Imputation Method for Wastewater Treatment Process 13

Missing-Pattern-Aware-Based Multi-Model Imputation Method fo...

引用

13th IEEE International Conference on CYBER Technology in Automation, Control, and intelligent systems, CYBER 2023

作者： Sun, Meiting Han, Honggui Li, Fangyu Beijing University of Technology Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing100124 China

ISBN: (纸本)9798350315196

In wastewater treatment process (WWTP), data-based implement techniques ensure the stable operation and reduce energy consumption. Different missing patterns of missing values often occur due to some harsh environments and sensor failures. However, existing imputation methods are difficult to obtain satisfactory data for implementation techniques as they only consider a single missing pattern. To solve this problem, a missing-pattern-aware-based multi-model imputation method (MPAMI) is proposed. First, a missing pattern aware network combining the advantages of the encoder and the axial attention mechanism is designed to identify different missing patterns. The network employs the encoder to extract nonlinear feature representations and utilizes the axial attention mechanism to focus on the most informative differential features. Second, a hybrid multi-model imputation method based on the long short-term memory network and the denoising autoencoder is developed to recover missing values. The imputation method utilizes the spatiotemporal information of missing patterns to improve imputation performance. Finally, experiments are conducted on a real WWTP dataset to validate the imputation performance of the proposed MPAMI. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Sewage Treatment Control Method Based on Genetic-SOFNN 4th

Sewage Treatment Control Method Based on Genetic-SOFNN

引用

4th International Conference on Machine Learning and intelligent Communications, MLICOM 2019

作者： Yang, Zhuang Yang, Cuili Qiao, Junfei Faculty of Information Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing100124 China

ISBN: (纸本)9783030323875

In the sewage treatment process, the dissolved oxygen concentration is a very important control target, but it is difficult to be controlled. To solve this problem, a self-organizing fuzzy neural network controller based on genetic ideas (G-SOFNN) is proposed. In the controller structure reduction process, the deleted neuron information is merged with the remaining neurons to reduce the interference set. During the controller structure increasing phase, the information of new neurons is initialized to avoid overlapping of information. Then, the controller parameters are trained by the projection algorithm to improve the control precision. Experiments illustrate that the proposed method can accurately control the concentration of dissolved oxygen in the sewage treatment process. © 2019, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

关键词： Fuzzy neural networks

来源：评论

学校读者我要写书评

暂无评论

Vehicle Re-identification Based on Multi-Scale Attention Feature Fusion 11

Vehicle Re-identification Based on Multi-Scale Attention Fea...

引用

11th International Conference on computing and Pattern Recognition, ICCPR 2022

作者： Su, Geyan Sun, Zhonghua Jia, Kebin Feng, Jinchao Beijing Key Laboratory of Computational Intelligence and Intelligent System Faculty of Information Technology Beijing University of Technology Beijing100124 China

ISBN: (纸本)9781450397056

It is important to extract vehicle appearance features for vehicle re-identification. The appearance variation of the same vehicle from different viewpoints and the appearance similarity between vehicles from different classes bring challenges for capturing the descriptive features. Considering these, we propose a multi-scale attention feature fusion network (MSAF) for vehicle re-identification. It uses ResNet50 as the backbone, and introduces a scalable channel attention module for each feature channel. Then a multi-scale fusion module is designed to output the final extracted vehicle features. Experimental results on the VERI-Wild dataset indicate that the proposed MSAF achieves high Rank-1 index of 91.20% with mAP of 80.20%. © 2022 ACM.

关键词： Vehicles

来源：评论

学校读者我要写书评

暂无评论

Identification of Weather Phenomena Based on Lightweight Convolutional Neural Networks

引用

Computers, Materials & Continua 2020年第9期64卷 2043-2055页

作者： Congcong Wang Pengyu Liu Kebin Jia Xiaowei Jia Yaoyao Li Faculty of Information Technology Beijing University of TechnologyBeijing100124China Beijing Laboratory of Advanced Information Networks Beijing100124China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of TechnologyBeijing100124China Department of Computer Science and Engineering University of MinnesotaTwin CitiesUSA

Weather phenomenon recognition plays an important role in the field of ***,weather radars and weathers sensor have been widely used for weather ***,given the high cost in deploying and maintaining the devices,it is difficult to apply them to intensive weather phenomenon ***,advanced machine learning models such as Convolutional Neural Networks(CNNs)have shown a lot of promise in meteorology,but these models also require intensive computation and large memory,which make it difficult to use them in *** practice,lightweight models are often used to solve such ***,lightweight models often result in significant performance *** this end,after taking a deep dive into a large number of lightweight models and summarizing their shortcomings,we propose a novel lightweight CNNs model which is constructed based on new building *** experimental results show that the model proposed in this paper has comparable performance with the mainstream non-lightweight model while also saving 25 times of memory *** memory reduction is even better than that of existing lightweight models.

关键词： Deep learning convolution neural networks lightweight models weather identification

来源：评论

学校读者我要写书评

暂无评论

Design and Analysis of a Cross-Frame Wall-Climbing Robot

Design and Analysis of a Cross-Frame Wall-Climbing Robot

引用

第43届中国控制会议

作者： Naigong Yu Zheng Tian Beijing Key Laboratory of Computing Intelligence and Intelligent System

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

This paper analyzes the operating environment and actual use requirements of the photovoltaic glass curtain wall cavity in high-rise buildings, designs a cross-frame motion mechanism that combines the obstacle-crossing lifting and adsorption modules into an obstacle-crossing adsorption device and integrates the motion mechanism of the robot into one. In terms of the design of the robot's control system, all working conditions of the robot are defined according to the motion state of the robot in the cleaning test of photovoltaic curtain wall cavities and combined with indicators such as coverage and cleaning speed of the photovoltaic cleaning test of photovoltaic curtain wall cavities, a set of automatic control system suitable for cross-frame wall-climbing robots. In the experimental section, we tested and analyzed the robot's motion performance, obstacle-crossing ability, and control system, proving the rationality of the robot structure and control system.

关键词： Climbing robot Cross frame structure Photovoltaic curtain wall

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：