检索结果-内蒙古大学图书馆

第34届中国控制与决策会议

作者： Jiexi Liu Chenyu Wang Lei Xu Beijing Institute of Technology Key Laboratory of Complex System Intelligent Control and Decision

Target Detection is one of the most important tasks in Computer Vision,which has broad application prospects in many *** the past few years,great progress has been made in this field with the rapid development of deep learning ***,it's still a big challenge at complex scenarios such as dim environment for traditional single-modal visible *** address this problem,researchers introduce thermal images as additional modal in consideration of that thermal cameras are less susceptible to interference and explore how to fuse the two modalities information ***,the lack of large labeled and high-quality visible-thermal datasets hampers the usage of convolutional neural networks for ***,we propose to use image-to-image translation model combined with a differentiable data augmentation method to generate fake thermal images from labeled visible images and use multi-modal target detection model to prove the validity of the *** experiment results show that our method can provide us a large labeled dataset of synthetic visible-thermal image pairs with better generalization and the introduction of thermal modality can obtain a better performance than single modality.

关键词： Image-to-image translation Multi-modal target detection Deep learning

来源：评论

学校读者我要写书评

暂无评论

Process Noise Estimator Based on Observation Sequence and its Application on Inertial Navigation system

Process Noise Estimator Based on Observation Sequence and it...

引用

2016 IEEE 13th International Conference on Signal Processing（ICSP2016）

作者： Xiao Xuan Huang Kun Yang Liming Liang Yuan Key Laboratory of Complex System Intelligent Control and Decision Beijing Institute of Technology

Kalman filter has been extensively applied in vast areas. However, it is widely acknowledged that the performance of Kalman filter depends on the accuracy of priori information such as model structure, statistics information of process and observation noise. Obtaining the covariance matrix of process noise is difficult in some application *** such background, this paper presents a process noise estimation algorithm based on the noise observation sequence. By constructing a transform matrix and removing the state variables from the observation, the noise observation sequence can be established, through which the covariance matrix of process noise can be estimated. Comparing to conventional adaptive filter, this algorithm needs less calculation. Moreover,the noise estimation process is separated from Kalman filter thus ensures Kalman Filters independence and optimality. The simulation results show that the new algorithm can effectively estimate the process noise covariance, and remain uninfluenced by the initial condition.

关键词： Adapt Kalman filter noise estimate

来源：评论

学校读者我要写书评

暂无评论

A Self-correcting Localization Approach for Automobile Robots Based on the Two Dimensional LADAR

A Self-correcting Localization Approach for Automobile Robot...

引用

第35届中国控制会议

作者： Jing Li Wenxue Liu Junzheng Wang Jianan Qiao Key Laboratory of Complex System Intelligent Control and Decision Beijing Institute of Technology

ISBN: (纸本)9781509009107

In this paper,a novel localization approach for autonomous mobile robots is *** can accomplish the all-terrain localization function with the laser radar(LADAR) and the inertial measurement unit(IMU).First of all,for the multiple-input nonlinear localization system,a fuzzy filter is built up to estimate the position of the mobile ***,an efficient evaluation index is raised to determine whether the landmarks extracted from the adjacent data frame are the same,which can help the mobile robot to realize the self-correction *** the final,the experiments in the real world are implemented to show the effectiveness of the proposed approach.

关键词： Localization SLAM Fuzzy filter Self-correcting Obstacle match

来源：评论

学校读者我要写书评

暂无评论

3D Semantic Mapping Based on Convolutional Neural Networks

3D Semantic Mapping Based on Convolutional Neural Networks

引用

第37届中国控制会议

作者： Jing Li Yanyu Liu Junzheng Wang Min Yan Yanzhi Yao Key Laboratory of Complex System Intelligent Control and Decision Beijing Institute of Technology

As an important part of environmental perception, maps guarantee the accuracy of intelligent robots in navigation,localization and path planning. The traditional 3D maps mainly focus on the spatial structure of the objects, which lacks the semantic information. A method is proposed in the paper, this method combines convolutional neural networks（CNNs） and Simultaneous Localization and Mapping（SLAM） to create global dense 3D semantic maps for indoor scenes. The deep neural network that includes convolution and deconvolution is designed to predict semantic category of every pixel. RGB-D camera is used to obtain scene information, accomplish localization and build 3D maps simultaneously. The semantic information is integrated into the 3D scene, we present an octree map method to replace traditional point clouds method, which can reduce the error from pose estimation and single frame labeling. By this method, the accuracy of semantic information is greatly improved.

关键词： CNNs SLAM Semantic Map Octree Map

来源：评论

学校读者我要写书评

暂无评论

An Improved Segmentation Method for Porous Transducer CT Images

An Improved Segmentation Method for Porous Transducer CT Ima...

引用

2019第十一届数字图像处理国际会议

作者： Meiling Wang Ruoyu Guo Ke Ning Li Ming Beijing Institute of Technology Key Laboratory of Complex System Intelligent Control and Decision

The paper presents an improved image segmentation method with a straightforward workflow for porous transducer CT images,which can be used to establish porous transducer three-dimensional model and further study its *** distribution of CT images is firstly analyzed and Gaussian filtering is conducted to reduce divergence of CT *** improved fully convolutional neural network model based on U-Net,for which multi-channel images are set as network input,is trained using training *** proposed method improves pore connectivity of the segmentation *** of porosity and permeability relative errors as well as MIOU on test set shows that the proposed method is an effective and generic two-phase segmentation method for porous transducer CT images without need of adjusting any parameters.

关键词： Porous transducer image segmentation permeability pore connectivity fully convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Target tracking algorithm with adaptive learning rate complementary filtering

Target tracking algorithm with adaptive learning rate comple...

引用

第三十九届中国控制会议

作者： Yulei Pan Yongqiang Bai Key Laboratory of Complex System Intelligent Control and Decision Beijing Institute of Technology

The Correlation filtering algorithm is not effective for fast deformation and fast movement. It is easy to lose when encountering problems such as occlusion. However, it has a good advantage of dealing with situations such as motion blur and lighting changes. A tracking algorithm based on color statistical features has a good effect on the rotation and translation of *** Staple algorithm combines the two algorithms to track using complementary fusion, but it also does not handle the occlusion and other issues well. In this paper, based on the Staple algorithm, the average peak correlation energy（APCE） and the maximum response are introduced. The value is used as the tracking confidence, and a detector using a support vector machine（SVM） is *** the tracking confidence is low, the target is blocked or moved violently. At this time, the detector works,and the search area is expanded around the original area for the target. At the same time, because the traditional tracking algorithm uses a fixed learning rate to update the template, this paper uses an adaptive tracking learning rate. When the tracking confidence is low, the update speed of the target model is reduced, which can effectively deal with the occlusion deformation in the tracking process. OTB100 benchmark experiments show that this method can solve the occlusion problem during target tracking. The degree of change is robust and stability.

关键词： Target Tracking Complementary Filtering Re-detection Adaptive Learning Rate

来源：评论

学校读者我要写书评

暂无评论

Research on the weak Signal Detection Method Based on Adaptive Vibrational Resonance 35

Research on the weak Signal Detection Method Based on Adapti...

引用

第35届中国控制会议

作者： JIAO Shang-bin SUN Di LIU Ding HUANG Wei Chao National & Local Joint Engineering Research Center of Crystal Growth Equipment and System Integration Xi'an University of Technology Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing

ISBN: (纸本)9781509009107

In this paper,the weak signal detection under a stable noise is investigated based on bistable vibrational resonance(VR) which is driven by a high frequency *** the one hand,the energy of the high frequency drive signal is transferred to the low frequency weak signal when VR occurs;on the other hand,the control of stochastic resonance(SR) is achieved based on VR,which transfers more noise energy into useful signal *** addition,considering the requirements of real-time detection,the amplitude and frequency of the high frequency drive signal are optimized by the knowledge-based particle swarm optimization(KPSO),which takes the mean signal-noise-ratio(MSNR) of output as the fitness function,and the property that VR system produces the best resonance effect just when the valid system parameter a(B,Ω) is greater than zero as ***,the parameter compensation is combined to achieve multi-high frequency weak signals detection with a stable ***,the method is applied to the vibration fault diagnosis of a mono-crystalline silicon furnace,and the experiment results show the effectiveness and practicability of the method.

关键词： Adaptive vibrational resonance weak signal detection knowledge-based particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

A new control method of quadruped robot walking on rough terrain based on linear inverted pendulum method

A new control method of quadruped robot walking on rough ter...

引用

IEEE International Conference on Robotics and Biomimetics

作者： Xin Li Junyao Gao Qiang Huang Haojian Lu Zhe Xu Yi Liu Intelligent Robotics Institute Key Laboratory of Intelligent Control and Decision of Complex System Beijing China

ISBN: (纸本)9781479973989

This paper proposes a new method of linear inverted pendulum control, which is used to control the quadruped robot walking on regular uneven terrain such as ramp and stair, furthermore, this method can be applied to a more rugged terrain. This paper analyzes the dynamic parameters of a quadruped robot walking on the complex terrain, which means the supporting points of the robot lie in different heights relative to the level ground. In order to ensure the same leg length, there will be energy dissipation caused by the raise of the center of mass. Therefore this paper proposes a dual length linear inverted pendulum method (DLLIPM), which not only effectively reduces the energy dissipation, but also promotes the workspace utilization. In addition, Newton-Raphson algorithm is used to optimize the linear inverted pendulum movement, which makes the movement symmetrical and smooth. Finally this paper presents simulation results with DLLIPM on a quadruped robot with 16 degrees of freedom.

关键词： Legged locomotion Foot Trajectory Mathematical model Robot sensing systems

来源：评论

学校读者我要写书评

暂无评论

The Optimal Duty Cycle Model-Predictive control of PMSM Used for Satellite Payload 1

引用

17th Annual Conference of China Electrotechnical Society, CES 2022

作者： Li, Jialin Chen, Zhen Hu, Hengzai Liu, Xiangdong Bao, Liang School of Automatic Beijing Institute of Technology Beijing100081 China Key Laboratory of Drive and Control of Servo Motion System Ministry of Industry and Information Technology Beijing100081 China State Key Laboratory of Intelligent Control and Decision for Complex System Beijing100081 China Beijing Aerospace Automatic Control Institute Beijing100854 China

ISBN: (数字)9789819904082

ISBN: (纸本)9789819904075

Model-predictive control is emerging as an advanced control strategy for its feature that the state variables can be predicted according to the math model of the control system. In the PMSM drive application, conventional model-predictive control strategy induces current ripple due to the fact that the optimal voltage vector always acts on the system for the whole control period. In order to restrain the current ripple, the idea of duty cycle control is taken into MPC by inserting a zero vector into the optimal non-zero vector and regulating the duty cycle of non-zero vector. In this paper, the optimal duty cycle FSC-MPC strategy based on two different principles is investigated. Principle A is that the estimation is equal to the reference at the end of control cycle, and principle B is that the area under trace of the estimation is equal to that of reference. The calculation of duty based on the two different principles and the flow chart of the strategy are introduced in detail. Finally, the effectiveness of the optimal duty cycle FSC-MPC is firstly verified by the simulation, and then verified by the experimental results, which is implemented on the PMSM driver of the DQ-1 satellite payload. © 2023, Beijing Paike Culture Commu. Co., Ltd.

关键词： Model predictive control

来源：评论

学校读者我要写书评

暂无评论

Observer-based output feedback control of networked control systems with non-uniform sampling and time-varying delay

引用

International Journal of systems Science 2017年第14期48卷 3118页

作者： Meng, Su Chen, Jie Sun, Jian Key Laboratory of Intelligent Control and Decision of Complex System Beijing Institute of Technology Beijing P.R. China

This paper investigates the problem of observer-based output feedback control for networked control systems with non-uniform sampling and time-varying transmission delay. The sampling intervals are assumed to vary within a given interval. The transmission delay belongs to a known interval. A discrete-time model is first established, which contains time-varying delay and norm-bounded uncertainties coming from non-uniform sampling intervals. It is then converted to an interconnection of two subsystems in which the forward channel is delay-free. The scaled small gain theorem is used to derive the stability condition for the closed-loop system. Moreover, the observer-based output feedback controller design method is proposed by utilising a modified cone complementary linearisation algorithm. Finally, numerical examples illustrate the validity and superiority of the proposed method.

关键词： Non-uniform sampling time-varying delay networked control systems output feedback control stability

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：