检索结果-内蒙古大学图书馆

43rd Chinese Control Conference, CCC 2024

作者： Qin, Xiaolin Cheng, Lei Tao, Xiang Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education Wuhan430081 China Wuhan University of Science and Technology Institute of Robotics and Intelligent Systems Wuhan430081 China Wuhan University of Science and Technology School of Information Science and Engineering Wuhan430081 China

ISBN: (纸本)9789887581581

Accurately and promptly identifying harmful gases in the environment can effectively prevent many accidents. Most existing pattern recognition methods are focused on addressing gas classification problems in static scenarios, but the real world is a dynamic and changing environment. Therefore, researching gas classification in continuous learning scenarios will better meet the needs of suddenly emerging gas classification tasks in real-world environments. To address this issue, an incremental learning model was introduced into gas classification, and an improved deep convolutional network was proposed for gas image feature extraction. The proposed model and other baseline methods experimented on an open-source gas dataset. The proposed model reduces knowledge loss due to catastrophic forgetting and effectively classifies the 10 gas samples in the dataset in continuous learning scenarios compared to non-incremental methods. It outperforms other baseline incremental methods by optimizing the network's shortcut connections, making the feature extraction network easier to train and improving gas classification performance. In dynamic learning scenarios, the proposed model achieved a classification accuracy of 86.30% for 10 gas samples. The experimental results demonstrate that the proposed method effectively mitigates catastrophic forgetting interference and completes gas classification tasks in continuous learning scenarios. © 2024 Technical Committee on Control Theory, Chinese Association of automation.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

AsynEVO: Asynchronous Event-Driven Visual Odometry for Pure Event Streams

arXiv

引用

arXiv 2024年

作者： Wang, Zhixiang Li, Xudong Zhang, Yizhai Huang, Panfeng The Research Center for Intelligent Robotics Shaanxi Province Innovation Team of Intelligent Robotic Technology School of Automation Northwestern Polytechnical University Xi’an710072 China The Research Center for Intelligent Robotics Shaanxi Province Innovation Team of Intelligent Robotic Technology School of Astronautics Northwestern Polytechnical University Xi’an710072 China

Event cameras are bio-inspired vision sensors that asynchronously measure per-pixel brightness changes. The high-temporal resolution and asynchronicity of event cameras offer great potential for estimating robot motion states. Recent works have adopted the continuous-time estimation methods to exploit the inherent nature of event cameras. However, existing methods either have poor runtime performance or neglect the high-temporal resolution of event cameras. To alleviate it, an Asynchronous Event-driven Visual Odometry (AsynEVO) based on sparse Gaussian Process (GP) regression is proposed to efficiently infer the motion trajectory from pure event streams. Concretely, an asynchronous frontend pipeline is designed to adapt event-driven feature tracking and manage feature trajectories;a parallel dynamic sliding-window backend is presented within the framework of sparse GP regression on SE(3). Notably, a dynamic marginalization strategy is employed to ensure the consistency and sparsity of this GP regression. Experiments conducted on public datasets and real-world scenarios demonstrate that AsynEVO achieves competitive precision and superior robustness compared to the state-of-the-art. The experiment in the repeated-texture scenario indicates that the high-temporal resolution of AsynEVO plays a vital role in the estimation of high-speed movement. Furthermore, we show that the computational efficiency of AsynEVO significantly outperforms the incremental method. Copyright © 2024, The Authors. All rights reserved.

关键词： Continuous time systems

来源：评论

学校读者我要写书评

暂无评论

Continuous Gaussian Process Pre-Optimization for Asynchronous Event-Inertial Odometry

arXiv

引用

arXiv 2024年

作者： Wang, Zhixiang Li, Xudong Zhang, Yizhai Zhang, Fan Huang, Panfeng The Research Center for Intelligent Robotics Shaanxi Province Innovation Team of Intelligent Robotic Technology School of Automation Northwestern Polytechnical University Xi’an710072 China The Research Center for Intelligent Robotics Shaanxi Province Innovation Team of Intelligent Robotic Technology School of Astronautics Northwestern Polytechnical University Xi’an710072 China

Event cameras, as bio-inspired sensors, are asynchronously triggered with high-temporal resolution compared to intensity cameras. Recent work has focused on fusing the event measurements with inertial measurements to enable ego-motion estimation in high-speed and HDR environments. However, existing methods predominantly rely on IMU preintegration designed mainly for synchronous sensors and discrete-time frameworks. In this paper, we propose a continuous-time preintegration method based on the Temporal Gaussian Process (TGP) called GPO. Concretely, we model the preintegration as a time-indexed motion trajectory and leverage an efficient two-step optimization to initialize the precision preintegration pseudo-measurements. Our method realizes a linear and constant time cost for initialization and query, respectively. To further validate the proposal, we leverage the GPO to design an asynchronous event-inertial odometry and compare with other asynchronous fusion schemes within the same odometry system. Experiments conducted on both public and own-collected datasets demonstrate that the proposed GPO offers significant advantages in terms of precision and efficiency, outperforming existing approaches in handling asynchronous sensor fusion. Copyright © 2024, The Authors. All rights reserved.

关键词： Continuous time systems

来源：评论

学校读者我要写书评

暂无评论

UAV Autonomous Landing Technology Based on April Tags Vision Positioning Algorithm

UAV Autonomous Landing Technology Based on April Tags Vision...

引用

第三十八届中国控制会议

作者： Zhou Li Yang Chen Hao Lu Huaiyu Wu Lei Cheng Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology

The issue of the autonomous landing of the quadrotor unmanned aerial vehicle(UAV) has become a hot topic in recent years. How to improve the positioning accuracy of autonomous landing, reduce the calculation time, and improve the dynamic tracking effect are all problems that need to be solved. In this paper, an onboard solution to the tracking and landing problems of UAV and ground vehicle is presented. The April Tags visual positioning algorithm based on monocular camera is used to locate and estimate the UAV's states. The platform used for UAV landing is labeled with different size of tags so that the drone can get good positioning accuracy at different heights. Combined with the PID control algorithm, the performance of dynamic tracking is improved. The experimental results show that the proposed method has good positioning accuracy and fast calculation speed.

关键词： Quadrotor autonomous landing April Tags PID

来源：评论

学校读者我要写书评

暂无评论

Quadrotor Control based on Self-Tuning LQR

Quadrotor Control based on Self-Tuning LQR

引用

第37届中国控制会议

作者： Kaiwen Pan Yang Chen Zhengxi Wang Huaiyu Wu Lei Cheng Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology

Recently, quadrotor has become one of the hotspots in robot research. To design an efficient controller for a quadrotor UAV is always challenging to for its under-actuated, strong coupling and nonlinearity. In this paper, a double closed-loop cascaded linear quadratic regulator(LQR) controller is proposed to track the referred position and attitude respectively. In order to solve the problem of tedious parameter setting, a parameter self-tuning control structure based on Simultaneous Perturbation Stochastic Approximation(SPSA) is designed and the parameters in the position and attitude controller are optimized respectively. By comparing with the original LQR controller, the simulation results show that the parameter-optimized controller improves the dynamic performance and control effect both in position control and attitude control, which verifies the correctness and feasibility of the proposed method.

关键词： Quadrotor UAVs LQR SPSA Parameter Self-Tuning

来源：评论

学校读者我要写书评

暂无评论

The generalized synchronization for double-layer delayed coupling complex networks by dynamic soft variable structure control method

The generalized synchronization for double-layer delayed cou...

引用

2022 Chinese automation Congress, CAC 2022

作者： Qian, Kang Chen, Zhihuan Sun, Yanqin Engineering Research Center for Metallurgical Automation and Measurement Technology Ministry of Education Wuhan University of Science and Technology Wuhan China Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology Wuhan China School of Information Science and Engineering Wuhan University of Science and Technology Wuhan China

ISBN: (纸本)9781665465335

This study focuses on the synchronization problem of double-layer complex networks with time delayed coupling. Distinguish from most of previous works, a more general network model is introduced, where the number and the dimension of the nodes can be different between the drive network and the response network. This network model is more close to the practice. In order to obtain the synchronization of double-layer complex networks, an ingenious dynamic soft variable structure controller is proposed, where three sub-controllers and one adaptive observers are designed, and the stability of the controller is demonstrated by a strict mathematical proof. Numerical simulations verify the effectiveness of proposed strategy. © 2022 IEEE.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Multi-UAV Cooperative 3D Coverage Path Planning Based on Asynchronous Ant Colony Optimization

Multi-UAV Cooperative 3D Coverage Path Planning Based on Asy...

引用

第40届中国控制会议

作者： Hui Li Yang Chen Zhihuan Chen Huaiyu Wu Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology

UAVs carrying visual sensors to capture the defects on the building surface has the advantages of high efficiency,low cost,flexibility and *** kind of inspection task is called coverage path planning problem in 3 D *** the target to be detected is a large complex building,multi-UAV collaboration is usually ***,how to obtain the optimal path of multi-UAV becomes a great *** order to solve the problem,Asynchronous Ant Colony Optimization（AACO）,which makes multiple ant colonies move forward asynchronously,is proposed here to conquer the ***,the random sampling method is used to get the potential UAV waypoints in the 3 D free environment,based on which a Primitive Coverage Graph（PCG） is ***,the visibility matrix and coverage rate are defined to quantify the coverage performance of UAV path ***,Asynchronous Ant Colony Optimization combined with reward strategy is proposed to solve the problem by selecting jump cities in ***,several simulations are provided to verify the feasibility and effectiveness of the algorithm.

关键词： 3D Coverage Path Planning Asynchronous Ant Colony Optimization Path Primitive

来源：评论

学校读者我要写书评

暂无评论

Multi-Domain Image Super-Resolution Generative Adversarial Network for Low-Resolution Person Re-Identification

Multi-Domain Image Super-Resolution Generative Adversarial N...

引用

第40届中国控制会议

作者： Siwen Chen Huaiyu Wu Yang Chen Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology

Person re-identification（ReID） is an important task in video surveillance *** address the issue that various low-resolutions and scale mismatching always exist in the real world,a multi-domain image-to-image translation network,termed Multi-Domain image Super-Resolution Generative Adversarial Network（MSRGAN）,is proposed to learn the mapping relationship between the various low-resolution domains and the high-resolution *** can ensure that the transferred image has a similar resolution as in the target *** is also able to keep the identity information of images from low-resolution domain during the *** addition,a novel ReID model,termed CSA-ReID in which channel attention and spatial attention module are introduced,is designed to learn resolution-invariant deep *** proposed method achieves 90.7% rank-1 accuracy and 96.4% rank-5 accuracy on multiple low-resolutions Market-1501 *** experimental results prove that the proposed method achieves promising generalization ability and accuracy compared with the state-of-theart methods.

关键词： person re-identification cross-resolution super-resolution attention

来源：评论

学校读者我要写书评

暂无评论

Distributed Non-Fragile State Estimation for discrete nonlinear systems with parameter uncertainties

Distributed Non-Fragile State Estimation for discrete nonlin...

引用

第40届中国控制会议

作者： Ping Xu Huaiyu Wu Xiujuan Zheng Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology

The distributed non-fragile state estimation problem is proposed for a class of discrete nonlinear systems with parameter uncertainties. Considering the parameter uncertainty of the system and the stochastic disturbance of the corresponding filter gain, a distributed non-fragile recursive state estimator is designed. Based on the strategy of linear consensus, a distributed non-fragile recursive filter is derived for each sensor node. Due to some uncertain parameters of the system, the upper bound of the filter error covariance is further derived. Then, the filter gain is obtained by minimizing the upper bound of the error covariance. Finally, an indoor localization system for mobile robots is provided to verify the proposed state estimator.

关键词： Non-fragile state estimation distributed filtering parameter uncertainties wireless sensor network

来源：评论

学校读者我要写书评

暂无评论

Intuitive maneuver of multi-DoFs laparoscopic system for minimally invasive surgery

Intuitive maneuver of multi-DoFs laparoscopic system for min...

引用

IEEE International Conference on Industrial Technology (ICIT)

作者： Ren C. Luo Jung-Yu Tsai Keng-Ming Lee Chen Hsuan Tsung Center for Intelligent Robotics and Automation Research (iCeiRA) National Taiwan University Taipei Taiwan Hiwin Technologies Corp Taichung Taichung TW

Robot-assisted minimally invasive surgery (RMIS) has become widespread because of its accuracy, stability and intuition. In the traditional laparoscopic surgery, an assistant is essential to help the surgeon in controlling the laparoscope. However, surgeons need to communicate with the assistant, and it is also difficult for an assistant manipulate the laparoscope to an anatomical position that the surgeon desired. Accordingly, the objective of this paper is to develop the Intuitively Robotic Laparoscope System (IRLS) with a more intuitive means to conquer these problems. With the gyroscope mounted on the head, the surgeon can dominate the laparoscope maneuver directly. To maneuver multi degree of freedoms and increase safety, a pedal is used as a power and mode switch. Furthermore, an algorithm is also adopted to adjust the angle collected by the gyroscope to make this system more ergonomic. Finally, an experimental result demonstrates that the data from the gyroscope being used is feasible and the error is acceptable.

关键词： Gyroscopes Laparoscopes Endoscopes Magnetic heads Switches Minimally invasive surgery

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：