检索结果-内蒙古大学图书馆

Scribble-Supervised Video Object Segmentation

IEEE/CAA Journal of Automatica Sinica 2022年第2期9卷 339-353页

作者： Peiliang Huang Junwei Han Nian Liu Jun Ren Dingwen Zhang Brain and Artificial Intelligence Laboratory School of AutomationNorthwestern Polytechnical UniversityXi’an 710072China IEEE Department of Engagement Services Mohamed Bin Zayed University of Artificial IntelligenceAbuDhabiUnited Arab Emirate Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory BeijingChina

Recently,video object segmentation has received great attention in the computer vision *** of the existing methods heavily rely on the pixel-wise human annotations,which are expensive and time-consuming to *** tackle this problem,we make an early attempt to achieve video object segmentation with scribble-level supervision,which can alleviate large amounts of human labor for collecting the manual ***,using conventional network architectures and learning objective functions under this scenario cannot work well as the supervision information is highly sparse and *** address this issue,this paper introduces two novel elements to learn the video object segmentation *** first one is the scribble attention module,which captures more accurate context information and learns an effective attention map to enhance the contrast between foreground and *** other one is the scribble-supervised loss,which can optimize the unlabeled pixels and dynamically correct inaccurate segmented areas during the training *** evaluate the proposed method,we implement experiments on two video object segmentation benchmark datasets,You Tube-video object segmentation(VOS),and densely annotated video segmentation(DAVIS)-*** first generate the scribble annotations from the original per-pixel ***,we train our model and compare its test performance with the baseline models and other existing *** experiments demonstrate that the proposed method can work effectively and approach to the methods requiring the dense per-pixel annotations.

关键词： Convolutional neural networks(CNNs) scribble self-attention video object segmentation weakly supervised

来源：评论

学校读者我要写书评

暂无评论

Leader-follower Formation control for a Multi-missile system via Deep Reinforcement Learning

Leader-follower Formation Control for a Multi-missile System...

引用

Chinese control and Decision Conference, CCDC

作者： Jianfeng Li Keyuan Yu Shenmin Song Center for Control Theory and Guidance Technology School of Astronautics Harbin Institute of Technology Harbin China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (数字)9798350387780

ISBN: (纸本)9798350387797

The methods for the formation tracking control problem are mainly based on nonlinear control theories. Usually, the control effect of such methods heavily depends on an accurate control model and a lot of controller parameters need to be selected properly. To cope with these shortcomings, a novel reinforcement learning-based formation controller is proposed for a leader-follower multi-missile system. The deep deterministic policy gradient (DDPG) algorithm is implemented to generate the formation control signals. An interaction environment containing a leader and a follower is formulated, in which a model-free optimal control policy is obtained by a training process. The convergent actor-network is tested for a multi-missile system. Simulation results show that the formation control objective can be rapidly achieved in spite of the change of leader maneuvers. There is no need to change the hyperparameters of the policy for different kinds of formation scenarios, which shows great advantages over the conventional nonlinear control theory-based methods in terms of parameter selection.

关键词： Training Three-dimensional displays Target tracking Simulation Heuristic algorithms Process control Optimal control

来源：评论

学校读者我要写书评

暂无评论

Visual Explanation of Object Detectors via Saliency Maps 18

Visual Explanation of Object Detectors via Saliency Maps

引用

18th IEEE Conference on Industrial Electronics and Applications, ICIEA 2023

作者： Xiao, Jin Liu, Wenrui Hu, Xiaoguang Jiang, Hao Wang, Weipeng Yahang, Li Beihang University State Key Laboratory of Virtual Reality Technology and Systems Beijing China Beihang University Beijing China The Third Academy of Casic Beijing China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (纸本)9798350312201

In recent years, rapid progress of deep learning has resulted in object detection algorithms based on deep neural networks nearly completely replacing traditional methods. This shift has led to significant improvements in both accuracy and speed of object detection. However, deep neural networks that support decision-making inobject detection are complex black boxes with hidden internal logic and operations. This lack of transparency results in deep learning models being unexplainable, which limits their practical deployment. In this paper, we analyze the current state and future trends of interpretable technology in object detection and propose an effective object detection interpretable algorithm. This algorithm aims to explain the underlying logic of AI decision-making and provide support for the high performance and credibility of object detection algorithms based on deep neural networks through quantitative evaluation indicators. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Collaborative Mission Planning for UAV Swarms Launched from Aerial Carriers

Collaborative Mission Planning for UAV Swarms Launched from...

引用

International Conference on Guidance, Navigation and control, ICGNC 2024

作者： Yuan, Mingzhou Dong, Bo Yao, Weiran School of Astronautics Harbin Institute of Technology Harbin 150001 China National Key Laboratory of Complex System Control and Intelligent Agent Cooperation Beijing 100074 China Intelligent Science and Technology Academy Limited of China Aerospace Science and Industry Corporation (CASIC) Beijing 100043 China

ISBN: (纸本)9789819622115

This paper takes the aerial carriers to launch UAV swarms as the background, aims to minimize the total flight distance of the UAV swarms, and proposes a clustering-launching-planning solution framework. Aiming at the scenario where UAVs adopt Dubins curve trajectories, a Dubins-KMeans clustering algorithm based on the Dubins curve length as the similarity measure is proposed. To ensure the shortest distance for air carriers to traverse each cluster, a two-part genetic algorithm is used to solve this MTSP problem and find the optimal order for air carriers to traverse each cluster. Ultimately, the energy efficiency of the entire unmanned system is optimized. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Dubins-KMeans clustering algorithm Launching UAVs two-part genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Unmanned Aerial Vehicles Cooperative Search Approach for complex Environments

arXiv

引用

arXiv 2023年

作者： Liu, Zhenchang Hao, Mingrui Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing100074 China

This paper studies a heterogeneous Unmanned Aerial Vehicles (UAVs) cooperative search approach suitable for complex environments. In the application, a fixed-wing UAV drops rotor UAVs to deploy the cluster rapidly. Meanwhile, the fixed-wing UAV works as a communication relay node to improve the search performance of the cluster further. The distributed model predictive control and genetic algorithms are adopted to make online intelligent decisions on UAVs’ search directions. On this basis, a jump grid decision method is proposed to satisfy the maneuverability constraints of UAVs, a parameter dynamic selection method is developed to make search decisions more responsive to task requirements, and a search information transmission method with low bandwidth is designed. This approach can enable UAVs to discover targets quickly, cope with various constraints and unexpected situations, and make adaptive decisions, significantly improving the robustness of search tasks in complex, dynamic, and unknown environments. The proposed approach is tested with several search scenarios, and simulation results show that the cooperative search performance of heterogeneous UAVs is significantly improved compared to homogeneous UAVs. © 2023, CC BY.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Minimal Curvature and Range Path Connecting Two Oriented Points

Minimal Curvature and Range Path Connecting Two Oriented Poi...

引用

International Conference on Autonomous Unmanned systems, ICAUS 2021

作者： Guo, Kun Xu, Cheng Wei, Zhenyan Xu, Heng Beijing Institute of Mechanical and Electrical Engineering Beijing100074 China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Bejing100074 China

ISBN: (纸本)9789811694912

Time and acceleration constraints have been extensively considered in flight vehicle path planning area. However, the limits of the required acceleration guaranteeing the existence of a path has not been fully studied yet. This paper presents an algebraic proof of the minimax curvature path connecting two oriented points with one inflection point, and then the minimal required acceleration and flight time are analyzed for all planar scenarios. It is revealed that for some geometry conditions, there exists a specific combination of curvature and range limits, while for other conditions, this combination can change continuously by increasing one and decreasing another. For all cases, the optimal paths are concatenations of circular arcs of equivalent radii in different forms, which means that to reduce the acceleration and flight time to the utmost, the overload should be evenly distributed along the whole path. The results in this paper can support fast on-line path planning in the presence of incident direction constraint with limited maneuverability and fuel. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Research on Multi-target Association Tracking Algorithm Based on IMM

Research on Multi-target Association Tracking Algorithm Base...

引用

International Conference on Autonomous Unmanned systems, ICAUS 2021

作者： Siting, Peng Lei, Li Yichao, Ma Jun, Ren Beijing Electro-Mechanical Engineering Institute Beijing100074 China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing100074 China

ISBN: (纸本)9789811694912

Radar sensors are important detection and tracking equipment in national defense work. There are multiple targets in actual combat scenarios. It is necessary to quickly and accurately capture suspicious targets in the reconnaissance area and track them in real time. complex electromagnetic interference conditions and multi-target tracking requirements pose a great challenge to track correlation. This paper establishes a multi-target maneuver scenario and uses a joint probabilistic data interconnection algorithm based on interactive multi-model (IMM) to perform multi-target tracking. Correlation tracking, simulation results show that the algorithm is used to correlate the target maneuver and measurement information with good accuracy. Interactive multi-model filtering can mix the estimation results of each filter to generate a fused state estimation result. The description of the model is more accurate, which improves the tracking accuracy and tracking stability of the algorithm. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Kalman filters

来源：评论

学校读者我要写书评

暂无评论

Remote Sensing Image Object Detection Algorithm Based on Receptive Field Enhancement

Remote Sensing Image Object Detection Algorithm Based on Rec...

引用

International Conference on Autonomous Unmanned systems, ICAUS 2021

作者： MingJun, Liu Yun, Zhang Rui, Zhang Cheng, Xu Beijing Electro-Mechanical Engineering Institute Beijing10074 China Science and Technology on Complex System Control and Intelligent Agent Corporation Laboratory Beijing10074 China

ISBN: (纸本)9789811694912

Small object detection algorithm is always a difficult research topic in remote sensing images. A remote sensing image object detection algorithm which is based on receptive field enhancement is proposed in this paper. Firstly, the backbone network is the SSD net. In order to form the feature map pyramid, the receptive field enhanced convolution module is added in feature extraction. Then, the feature maps of the low-level and the high-level are spliced, so he output feature map retains the rich context information. Because it contains the information in the low-level feature map and the high-level feature map, which are location information and semantic information, respectively. Finally, based on the idea of dilated convolution, the receptive field enhancement module is designed to make the receptive field of the feature map larger. And the feature information is filtered to reduce the aliasing effect when the feature map is spliced. After the experimental result on remote sensing image dataset NWPU vhr-10, the average detection accuracy of the algorithm is 90%, which improves the object detection accuracy, especially the detection accuracy of the small objects. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Error Modeling Method of Aircraft HWIL Simulation system Based on BP Wavelet Neural Network 18th

Error Modeling Method of Aircraft HWIL Simulation System Ba...

引用

18th Chinese Intelligent systems Conference, CISC 2022

作者： Tong, Jiahui Liu, Jin Li, Jing Zhang, Qing Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing100190 China Beijing Electro-Mechanical Engineering Institute Beijing102299 China

ISBN: (纸本)9789811962028

In order to ensure the confidence of hardware-in-the-loop(HWIL) simulation results, the simulation equipment and system are required to simulate the realistic flight environment;However, the additional error introduced by the simulation system in the guidance and control loop should be minimized. By studying the error model of the aircraft HWIL simulation system and analyzing the impact of the HWIL simulation system on the guidance and control quality, we can recognize the real performance of the guidance and control system, find out the key factors affecting the performance of the HWIL simulation system, provide decision-making basis for the development of simulation equipment, and improve and enhance the reliability of the simulation system. Based on BP wavelet neural network, the mathematical modeling of the aircraft HWIL simulation system error is carried out, and the modeling method and process are introduced in detail. A new method of aircraft HWIL simulation system error modeling with input and output data is proposed, and the applicability of modeling method is analyzed, and satisfactory accuracy results are obtained. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Study on the Radio Frequency Simulation Method of Moving Ship for UAV-Borne Imaging Radar 18th

Study on the Radio Frequency Simulation Method of Moving Shi...

引用

18th Chinese Intelligent systems Conference, CISC 2022

作者： Diao, Guijie Gong, Ningbo Ni, Hong Liu, Zhe Tong, Jiahui Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing100190 China Beijing Electro-Mechanical Engineering Institute Beijing100083 China

ISBN: (纸本)9789811962257

A radio-frequency (RF) simulation method for UAV-borne imaging radar for moving ships at the sea-surface is proposed, which mainly solves the RF radar echo simulation problem for moving target at sea-surface. Based on the electromagnetic (EM) scattering model of sea-surface and ship target, the time-varying sea-surface scattering echo and moving ship target scattering echo are established respectively. Then, the scattering echo of a composite scene consisting of sea-surface background and ship-target is obtained by vector summation. Finally, the RF simulation system based on microwave anechoic chamber is constructed, and the simulation tests are carried out. Simulation results are presented to demonstrate the validity and applicability of the proposed method. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Radar imaging

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：