检索结果-内蒙古大学图书馆

A Virtual Quadrotor Simulation Platform for control Education

IFAC-PapersOnLine 2024年第16期58卷 101-106页

作者： Zhenhua Wang Ruihong Zhao Zhenwen Zhao Hongliang Zhou Yi Shen Department of Control Science and Engineering Harbin Institute of Technology Harbin 150001 China

This paper presents a web-based quadrotor simulation platform to facilitate automatic control education. The platform employs a client-server architecture. The server is developed using Python to implement numerical simulations, while the client leverages JavaScript libraries to achieve 3D visualization and an interactive user interface. Several built-in control modes based on PID controller are designed in this platform. Moreover, a programming interface is also provided to implement user-defined control algorithms. A key feature of this platform is that the controllers can be adjusted in real-time. The platform can be used to teach and practice automatic control theories such as model linearization, PID controller design, inner-outer loop control strategy, and LQR method. Through virtual control practice using the platform, students can understand automatic control theories better and improve their problem-solving capability.

关键词： Quadrotor Virtual Simulation Automatic control Education

来源：评论

学校读者我要写书评

暂无评论

Navigation line extraction based on image processing for weeding robot 48

Navigation line extraction based on image processing for wee...

引用

48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022

作者： Zheng, Hao Wang, Qiang Ji, Jinming Harbin Institute of Technology Department of Control Science and Engineering Harbin China

ISBN: (数字)9781665480253

ISBN: (纸本)9781665480253

For weeding, compared to satellite navigation, using the location information of crops in the field to directly navigate the weeding robot is more accurate, so navigation line extraction is a key point in weeding robot autonomous navigation in the field. Though deep learning and neural network has developed fast in recent years, it does not run fast enough to be used in real time scenarios, so an image processing based navigation line extraction algorithm is proposed in this paper. We used a camera to take RGB images, of which the top and the bottom portion would be cut out and the middle portion would be remained to reduce the impact of distance between the crops and the robot. The green in the original image would be found out by transforming the image into the HSV color and then the morphological operation would be used to remove the noises for further image processing. Generally there are more than one ridges in the image, so it is necessary to find out the ridge in the middle for navigation. We clustered the crops according to the distance between them and chose the cluster in the middle to extract the navigation line. At last, the least square algorithm was used to calculate the navigation line. We used the slope of the navigation line and the distance between it and the center point of the image to control the direction of the robot. Experiments of the weeding robot on a simulated farm field showed that the algorithm can achieve the desired navigation effect and further experiments in the field also showed its good performance. © 2022 IEEE.

关键词： Crops

来源：评论

学校读者我要写书评

暂无评论

SkeletonCLIP: Recognizing Skeleton-based Human Actions with Text Prompts 8

SkeletonCLIP: Recognizing Skeleton-based Human Actions with ...

引用

8th International Conference on Systems and Informatics, ICSAI 2022

作者： Yuan, Lin He, Zhen Wang, Qiang Xu, Leiyang Ma, Xiang Harbin Institute of Technology Department of Control Science and Engineering Harbin China

ISBN: (纸本)9781665474481

Human action recognition has been a hot research for decades, and mainst.eam supervised frameworks include a feature extraction backbone and a softmax classifier to predict daily human actions. When the number of classes applied to the dataset changes, we must retrain the classifier on the well-trained backbone. This pipeline restricts the generalization and transfer ability of the model due to an extra training period. Moreover, replacing action labels with simple number labels discards useful semantic information and can only receive a meaningless classifier at last. In this work, we present a model SkeletonCLIP for skeleton-based human action recognition. We add an alternative text encoder to extract semantic information from labels while keeping the original sequence encoder. We use dot production to measure the similarities of sequence-text pairs in place of traditional classifier head and cross-entropy loss. Experiments from three human action datasets show that our framework can reach a higher recognition accuracy with the help of semantic information when training the network from scratch. The code has been shown at eunseo-v/SkeletonCLIP. © 2022 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Sliding Mode control in Power Converters and Drives: A Review

引用

IEEE/CAA Journal of Automatica Sinica 2022年第3期9卷 392-406页

作者： Ligang Wu Jianxing Liu Sergio Vazquez Sudip K.Mazumder IEEE Department of Control Science and Engineering Harbin Institute of TechnologyHarbin 150001China Electronic Engineering Department Universidad de Sevilla41092 SevillaSpain Electrical and Computer Engineering Department University of Illinois ChicagoChicagoIL 60607 USA

Sliding mode control(SMC)has been studied since the 1950s and widely used in practical applications due to its insensitivity to matched *** aim of this paper is to present a review of SMC describing the key developments and examining the new trends and challenges for its application to power electronic *** fundamental theory of SMC is briefly reviewed and the key technical problems associated with the implementation of SMC to power converters and drives,such chattering phenomenon and variable switching frequency,are discussed and *** recent developments in SMC systems,future challenges and perspectives of SMC for power converters are discussed.

关键词： Motor drive power electronics systems sliding mode control

来源：评论

学校读者我要写书评

暂无评论

Design and Realization of Unmanned Vehicle Autonomous Navigation System Based on Binocular Camera 3

Design and Realization of Unmanned Vehicle Autonomous Naviga...

引用

3rd International Conference on Computer science, Electronic Information engineering and Intelligent control technology, CEI 2023

作者： Guo, Tianxing Lin, Lianlei Liu, Huanyu Li, Junbao School of Electronics and Information Engineering Harbin Institute of Technology Department of Automatic Test and Control Harbin China Information Countermeasure Technique Institute School of Computer Science and Technology Harbin Institute of Technology Harbin China

ISBN: (纸本)9798350372052

Aiming at the navigation problem of unmanned vehicles in extreme environments such as communication interference and limited GPS signals, this study proposes an autonomous navigation method based on binocular cameras. The method enables the unmanned vehicle to complete the task of localization and map building using only binocular images, without the need for other sensors or signal sources. Meanwhile, this study also proposes a local path planning obstacle avoidance method based on depth map, which, when combined with global localization information, can take into account both local obstacle avoidance and global path navigation, and can cope with random environmental changes. The system uses a single sensor for autonomous obstacle avoidance and navigation, which reduces the computational requirements and ensures the low cost of the navigation system. In order to verify the reliability and effectiveness of the system under signal-constrained conditions, the system is evaluated in a simulation environment and a real field scenario, respectively. The experimental results show that the system is able to achieve reliable localization and path planning under signal-constrained conditions. © 2023 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Design and Test of Electro-Mechanical Brake Experiment System 6

Design and Test of Electro-Mechanical Brake Experiment Syste...

引用

6th CAA International Conference on Vehicular control and Intelligence, CVCI 2022

作者： Houhua, Jing Ruixue, Feng Qinggan, Lin Harbin Institute of Technology Department of Control Science and Engineering Harbin China

ISBN: (数字)9781665453745

ISBN: (纸本)9781665453745

Brake by wire system is an important technology for the development of vehicle intelligence and electrification. EMB (Electro-Mechanical Braking System) will surely become a future automatic. The direction of the driving brake development because of its high degree of electrification, compact structure and strong expansibility, conforms to the development direction of electrification. In this paper, the EMB bench is built, and the corresponding simulation model is established. By testing the characteristics of the bench, it is proved that the established simulation model can fully reflect the characteristics of the EMB system. Finally, two clamping force control methods are designed for the EMB system, and the bench test is carried out to compare the control effects. Both braking strengths and fast braking are achieved. Tests show that the experimental system meets the requirements of the development of the electromechanical braking system, and will be used for control algorithm development. © 2022 IEEE.

关键词： Force control

来源：评论

学校读者我要写书评

暂无评论

Temporal-spatial Feature Fusion for Few-shot Skeleton-based Action Recognition 48

Temporal-spatial Feature Fusion for Few-shot Skeleton-based ...

引用

48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022

作者： Xu, Leiyang Wang, Qiang Lin, Xiaotian Yuan, Lin Ma, Xiang Harbin Institute of Technology Department of Control Science and Engineering Harbin China

ISBN: (数字)9781665480253

ISBN: (纸本)9781665480253

Recognizing new action categories from a few reference samples is an encouraging research field because the cost of labeling data is expensive. This work presents a method for few-shot (or one-shot) skeleton-based action recognition by fusing temporal and spatial features of actions. Trajectory primitives are proposed to characterize the temporal features, which can be obtained by segmenting and clustering the trajectories of joints. After that, we modify the original dynamic time warping (DTW) algorithm and use it to measure the similarity between trajectory primitive sequences. Besides, we compute the joint angles as spatial feature vectors. Support vector machines (SVM) are used to classify the joint angle vectors. In this way, the temporal distance matrix can be calculated by modified DTW, and the spatial distance matrix can be obtained by trained SVM. Finally, we fuse temporal and spatial distance matrices by adjusting a parameter to improve recognition accuracy. Furthermore, extensive experiments are conducted on three small-scale datasets to verify the effectiveness of our proposed method. © 2022 IEEE.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Spatial Transformer Network with Transfer Learning for Small-scale Fine-grained Skeleton-based Tai Chi Action Recognition 48

Spatial Transformer Network with Transfer Learning for Small...

引用

48th Annual Conference of the IEEE Industrial Electronics Society, IECON 2022

作者： Yuan, Lin He, Zhen Wang, Qiang Xu, Leiyang Ma, Xiang Harbin Institute of Technology Department of Control Science and Engineering Harbin China

ISBN: (数字)9781665480253

ISBN: (纸本)9781665480253

Human action recognition is a quite hugely investigated area where most remarkable action recognition networks usually use large-scale coarse-grained action datasets of daily human actions as inputs to state the superiority of their networks. We intend to recognize our small-scale fine-grained Tai Chi action dataset using neural networks and propose a transfer-learning method using NTU RGB+D dataset to pre-train our network. More specifically, the proposed method first uses a large-scale NTU RGB+D dataset to pre-train the Transformer-based network for action recognition to extract common features among human motion. Then we freeze the network weights except for the fully connected (FC) layer and take our Tai Chi actions as inputs only to train the initialized FC weights. Experimental results show that our general model pipeline can reach a high accuracy of small-scale fine-grained Tai Chi action recognition with even few inputs and demonstrate that our method achieves the state-of-the-art performance compared with previous Tai Chi action recognition methods. © 2022 IEEE.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Object Detection in Remote Sensing Scenes Using Anchor-free Methods and Multi-vector Loss Functions 41

Object Detection in Remote Sensing Scenes Using Anchor-free ...

引用

第41届中国控制会议

作者： Shuanghong Wu Qiaodi Wang Hongzhe Wang Haonan Guo Yanning Guo Department of Control Science and Engineering Harbin Institute of Technology STMicroelectronics

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

We propose a set of new loss functions applied to deep convolution neural network,which have a significant improvement on detecting remote-sensing *** detection is one of the most fundamental tasks of computer *** object detection in remote sensing scenes is difficult because these scenes have lots of small,oriented,and densely arranged objects,which are hard to be detected accurately and *** order to solve these difficulties,we use anchor-free method to design the detection algorithm,which can improve the detection *** our designed multi-vector bounding box and multi-vector loss functions are added to the baseline *** design can make detector regress the bounding box *** are carried out under two public *** experimental results show that our design effectively improves the detection accuracy and speed of objects in remote sensing ***,we make ablation study experiments on our multi-vector loss functions,which proves the rationality of our design.

关键词： Object Detection Anchor-free Remote Sensing Computer Vision

来源：评论

学校读者我要写书评

暂无评论

Research on Adaptive control of Tunneling Machine Based on PSO-BP Neural Network PID

Research on Adaptive Control of Tunneling Machine Based on P...

引用

2023 China Automation Congress, CAC 2023

作者： Liu, Ruohan Ren, Chunping Li, Degen Liu, Yongli School of Electrical and Control Engineering Safety Engineering College Heilongjiang University of Science and Technology Harbin China School of Mechanical Engineering Heilongjiang University of Science and Technology Harbin China Heilongjiang University of Science and Technology Academic Theory Research Department Harbin China School of Mining Engineering Heilongjiang University of Science and Technology Harbin China

ISBN: (纸本)9798350303759

In order to improve the response speed and control accuracy of the tunneling machine control system, and to address the limitations of traditional algorithms that are prone to falling into local extremum and weak generalization ability during the solving process, a PSO-BP neural network PID adaptive control method for tunneling machines is proposed. The weights of the BP neural network are optimized by the global search ability of the PSO algorithm. The neural network outputs the fitness values of the PID parameters, and the initial weights of the neural network are updated. The PID parameters are adjusted in realtime. The results show that the unit Step response curve of the system reaches the stable state of the system after 0.85 s in the traditional PID control, while the PSO-BP neural network PID controller only needs 0.4 s to be stable, and the overshoot of the PSO-BP neural network PID controller is far less than that of the traditional PID controller. Therefore, the PSO BP neural network PID controller has good dynamic and static characteristics, which can maximize the generalization ability of the traditional BP neural network and ensure the Rate of convergence and learning ability. © 2023 IEEE.

关键词： Three term control systems

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：