检索结果-内蒙古大学图书馆

Dynamic docking algorithm for UGV to UAV based on single planning under disturbed conditions

ISA TRANSACTIONS 2025年 157卷 496-509页

作者： Si, Jinge Xu, Yongkang Niu, Tianwei Wang, Liang Li, Bin Deng, Chencheng Wang, Shoukun Wang, Junzheng Beijing Inst Technol Sch Automat State Key Lab Intelligent Control & Decis Complex Beijing Peoples R China

This paper investigates the initial dynamic docking problem to mobile and trajectory-disturbed targets for tracking and recovering drones by Unmanned Ground Vehicles (UGVs). First, the target status is estimated by employing the Extended Kalman Filter (EKF). Then, the drone's perturbation is mapped to a dynamic docking point, quantifying the target motion deviation. Within this point frame, an initial path planning algorithm based on Bezier curve with polar offset points is designed. By integrating the initial path and real-time target states into the control layer, a docking control algorithm within single planning is developed. This algorithm enables the UGV to dynamically dock to a trajectory-disturbed drone in position and angle, while also achieving target tracking after simplification. Finally, simulations and experiments have been conducted to demonstrate its effectiveness.

关键词： Dynamic docking Extended Kalman Filter Bezier path planning Wheel robot control Dynamic target tracking

来源：评论

学校读者我要写书评

暂无评论

Targetless Lidar-Camera Calibration via Cross-Modality Structure Consistency

IEEE TRANSACTIONS ON INTELLIGENT VEHICLES

引用

IEEE TRANSACTIONS ON intelligent VEHICLES 2024年第1期9卷 2636-2648页

作者： Ou, Ni Cai, Hanyu Wang, Junzheng Beijing Inst Technol State Key Lab Intelligent Control & Decis Complex Beijing 100081 Peoples R China

Lidar and cameras serve as essential sensors for automated vehicles and intelligent robots, and they are frequently fused in complicated tasks. Precise extrinsic calibration is the prerequisite of Lidar-camera fusion. Hand-eye calibration is almost the most commonly used targetless calibration approach. This article presents a particular degeneration problem of hand-eye calibration when sensor motions lack rotation. This context is common for ground vehicles, especially those traveling on urban roads, leading to a significant deterioration in translational calibration performance. To address this problem, we propose a novel targetless Lidar-camera calibration method based on cross-modality structure consistency. Our proposed method utilizes cross-modality structure consistency and ensures global convergence within a large search range. Moreover, it achieves highly accurate translation calibration even in challenging scenarios. Through extensive experimentation, we demonstrate that our approach outperforms three other state-of-the-art targetless calibration methods across various metrics. Furthermore, we conduct an ablation study to validate the effectiveness of each module within our framework.

关键词： Laser radar Calibration Cameras Sensors Simultaneous localization and mapping Visualization Optimization Calibration lidar camera automated vehicles

来源：评论

学校读者我要写书评

暂无评论

Influence Enhanced Sparse Coordination Graphs for Multi-Agent Reinforcement Learning

引用

NEURAL NETWORKS 2025年 188卷 107454页

作者： Zhang, Xiwen Chen, Jie Gan, Ming-Gang Chen, Haoxiang Beijing Inst Technol Sch Automat State Key Lab Intelligent Control & Decis Complex Beijing 100081 Peoples R China

In contemporary Multi-Agent Reinforcement Learning (MARL), effectively enhancing the expressive capacity of value functions has been a persistent research focus. Many studies have employed value decomposition methods;however, due to the neglect of inter-agent collaboration, these methods fall short of achieving optimal performance. Subsequent research introduced coordination graphs into value decomposition methods;nevertheless, these approaches often rely on simplistic rules to evaluate inter-agent collaboration and fail to adequately describe the collaborative relationships of agents in complex environments. Consequently, we propose Influence Enhanced Sparse Coordination Graphs (IESCG) as a solution to provide insights into the aforementioned problem. In this study, we propose influence networks as quantitative descriptions of the importance of collaboration among agents, utilizing them as crucial basis for constructing the topology of Sparse Time-Varying Coordination Graphs. Additionally, we propose Recurrent Payoff Function Networks (RPFN) to incorporate temporal information while providing necessary input to influence networks. Furthermore, Sparse Graph Advantage Selection Coefficients (SGASC) are introduced to stabilize the overall value function across different time steps, ensuring training stability. Experimental investigations conducted on the StarCraft II micromanagement and MACO benchmark indicate that our algorithm not only accelerates convergence and improves winning probabilities but also exhibits more pronounced advantages in complex scenarios.

关键词： Multi-agent reinforcement learning Decentralized partially observable Markov decision process (Dec-POMDP) Coordination graphs Q-learning

来源：评论

学校读者我要写书评

暂无评论

A Velocity Tracking Approach Based on Neural Networks for a Multimodal Wheel-Legged Composite Gait on a Parallel Structure Robot

引用

IEEE-ASME TRANSACTIONS ON MECHATRONICS 2025年

作者： Xue, Junfeng Wang, Shoukun Xu, Yongkang Wang, Junzheng Beijing Inst Technol Sch Automat State Key Lab Intelligent Control & Decis Complex Beijing 100081 Peoples R China

Parallel-legged robots are favored by many researchers due to their potent environmental adaptability and load capacity. However, the work efficiency of the parallel leg structure is always confined due to its uptight workspace and extremely uneven force distributed among actuators. Therefore, this article proposes a new type of wheel-legged composite gait for the parallel structure, inspired by the characteristic of variation in the pitch angle of the foot-end when a human walks. First, the kinematic inverse solution of the wheel-legged composite gait is derived, where the reverse rotation of the wheel alleviates the slipping phenomenon of the foot-end on stance stage in humanoid gait from the kinematic level. Meanwhile, numerical simulation and experiments have verified that the proposed gait can enlarge the workspace of a single leg for 25.4% and alleviate the maximum force on the actuators for 16.4% minimizing actuator wear. Besides, a legged motion speed tracking approach is implemented for dynamics of the wheel-legged composite gait. Among them, a whole body dynamics model is constructed to adjust the neural network weights enabling real-time adjustment of incremental proportion-integration-differentiation control parameters based on the foot longitudinal force, to scale up its control efficiency. Finally, to ameliorate the issue of limited contact friction between the robot and the ground, an encoding approach to identify the sliding state of the robot's foot-end in the $X-Y$ plane has been proposed. This approach adaptively modifies both the longitudinal force threshold of the foot-end and the desired linear velocity of the robot body. These adjustments are guided by specific control rules, which aim to enhance the accuracy of speed tracking during legged motion across varying terrains. The resulting linear and yaw velocity steady errors are constrained to 0.05 m/s and 0.012 rad/s, respectively.

关键词： Composite gait multimodal wheel-legged robot neural network velocity tracking

来源：评论

学校读者我要写书评

暂无评论

The Behavior of Fluid Flow and Solute Transport in 3D Crossed Rock Fractures

引用

PROCESSES 2025年第1期13卷 67-67页

作者： Han, Xuefeng Xue, Kangsheng Zhang, Shaojie Chongqing Univ State Key Lab Coal Mine Disaster Dynam & Control Chongqing 400044 Peoples R China State Key Lab Intelligent Construct & Hlth Ope Xuzhou 221116 Peoples R China

Understanding the behavior of fluid flow and solute transport in fractured rock is of great significance to geoscience and engineering. The discrete fracture network is the predominate channel for fluid flow through fractured rock as the permeability of fracture is several magnitudes higher than that of the rock matrix. As the basic components of the fracture network, investigating the fluid flow in crossed fractures is the prerequisite of understanding the fluid flow in fractured rock. First, a program based on the successive random addition algorithm was developed to generate rough fracture surfaces. Next, a series of fracture models considering shear effects and different surface roughness were constructed. Finally, fluid dynamic analyses were performed to understand the role of flowrate and surface roughness in the evolution of flow field, concentration field, solute breakthrough, and solute mixing inside the crossed fractures. Results indicated that the channeling flow at the fracture intersection became more pronounced with the increasing P & eacute;clet number (Pe) and Joint Roughness Coefficient (JRC), the evolution of the concentration field was influenced by Pe and the distribution of the concentration field was influenced by JRC. For Pe < 10, the solute transport process was dominated by molecular diffusion. For 100 > Pe > 10, the solute transport process was in the complete mixing mode. In addition, for Pe > 100, the solute transport process was in the streamline routing mode. The concentration distribution was affected by the local aperture at the fracture intersection corresponding to different surface roughness. Meanwhile, the solute mixing equation was improved based on this result. The research results are beneficial for further revealing the mechanism of fluid flow and solute transport phenomenon in fractured rock.

关键词： 3D crossed fractures fluid flow behavior solute transport behavior the Peclet number channeling flow mixing ratio

来源：评论

学校读者我要写书评

暂无评论

Online spatio-temporal action detection with adaptive sampling and hierarchical modulation

引用

MULTIMEDIA SYSTEMS 2024年第6期30卷 1-15页

作者： Su, Shaowen Gan, Minggang Beijing Inst Technol Sch Automat State Key Lab Intelligent Control & Decis Complex Beijing 100081 Peoples R China

Online spatio-temporal action detection (OSTAD) is a crucial task in video understanding, responsible for identifying and categorizing action instances in video streams in an online manner. This paper presents a novel approach that employs adaptive sampling and hierarchical modulation to enhance OSTAD capabilities. Traditional methods, often constrained by fixed sampling rates, may lead to redundancy in scenarios with slower action speeds and overlook essential details in faster-moving sequences. Our innovative dynamic sampling strategy, informed by speed estimation, adaptively adjusts sampling intervals based on speed attention and visual differential features, thereby optimizing the informational content of each sampled video clip. Additionally, our method incorporates a hierarchical modulation mechanism that synergizes high-level semantic and low-level spatial information, significantly enhancing action localization and classification accuracy. The adaptive sampling network with hierarchical modulation, underpinned by these advancements, demonstrates substantial improvements on benchmark datasets such as JHMDB21 and UCF24, proving our methods' efficacy in handling diverse and dynamic action sequences in an online setting.

关键词： Spatio-temporal action detection Speed estimation Adaptive sampling Hierarchical modulation Online

来源：评论

学校读者我要写书评

暂无评论

Content Temporal Relation Network for temporal action proposal generation

引用

PATTERN RECOGNITION 2024年 149卷

作者： Gan, Ming-Gang Zhang, Yan Beijing Inst Technol Sch Automat State Key Lab Intelligent Control & Decis Complex Beijing 100081 Peoples R China

Temporal action proposal generation is an essential step for untrimmed video analysis and gains much attention from academia. However, most of the prior works predict the confidence score of each proposal separately and neglect the relations between proposals, limiting their performance. In this work, we design a novel Content Temporal Relation Network (CTRNet) to generate temporal action proposals by exploring the content and temporal semantic relations between proposals simultaneously. Specifically, we design a proposal feature map generation layer to convert the temporal semantic relations of proposals into spatial relations. Based on the proposal feature map, we propose a content-temporal relation module, which applies a novel adaptive -dilated convolution to model the temporal semantic relations between proposals and designs a content-adaptive convolution operation to explore the content semantic relation between proposals. Considering the temporal and content semantic relations between proposals, CTRNet has learned discriminative proposal features to improve performance. Extensive experiments are performed on two mainstream temporal action detection datasets, and CTRNet significantly outperforms the previous state-of-the-art methods. The codes are available at https://***/YanZhang-bit/CTRNet.

关键词： Temporal action proposal generation Temporal action detection Untrimmed video analysis Proposal-proposal relations

来源：评论

学校读者我要写书评

暂无评论

Energy-efficient on/off control in serial production lines with Bernoulli machines

引用

FLEXIBLE SERVICES AND MANUFACTURING JOURNAL 2024年第1期36卷 103-128页

作者： Wang, Xiaohan Dai, Yaping Jia, Zhiyang Beijing Inst Technol Sch Automat State Key Lab Intelligent Control & Decis Complex Beijing 100081 Peoples R China

Achieving energy-efficient production has always been a critical issue for the manufacturing industry. In many production systems, the waste of energy mainly results from unbalanced machine efficiency, which brings about a considerable amount of idle time. One of the most important approaches to addressing this problem is to adopt an On/Off control policy on those underutilized machines. In this paper, serial production lines with multiple unreliable machines and finite buffers are considered. All machines are assumed to obey the Bernoulli reliability model, which is commonly used when the machine downtime is relatively short and comparable to the production cycle time (e.g., engine assembly). Besides, considering that in industrial practice, machines usually have to go through a preparation phase during mode switching, a non-negligible warm-up (cool-down) period is taken into account. An On/Off control policy is calculated for lines with two or three machines based on a Markov decision process (MDP) model, then the results are extended to multi-machine lines using a decomposition procedure. Numerical experiments show that the proposed policy achieves better system performance compared to the threshold policy that is commonly used in energy-efficient On/Off control.

关键词： Production system engineering Energy-efficient manufacturing On off control Bernoulli reliability model Serial production line

来源：评论

学校读者我要写书评

暂无评论

Harvesting Energy from Atmospheric Water: Grand Challenges in Continuous Electricity Generation

引用

ADVANCED MATERIALS 2024年第12期36卷 2211165-2211165页

作者： Tan, Jin Wang, Xiang Chu, Weicun Fang, Sunmiao Zheng, Chunxiao Xue, Minmin Wang, Xiaofan Hu, Tao Guo, Wanlin Nanjing Univ Aeronaut & Astronaut Key Lab Intelligent Nano Mat & Devices State Key Lab Mech & Control Mech Struct Minist Educ Nanjing 210016 Peoples R China Nanjing Univ Aeronaut & Astronaut Inst Frontier Sci Key Lab Intelligent Nano Mat & Devices State Key Lab Mech & Control Mech StructMinist Ed Nanjing 210016 Peoples R China

Atmospheric water is ubiquitous on earth and extensively participates in the natural water cycle through evaporation and condensation. This process involves tremendous energy exchange with the environment, but very little of the energy has so far been harnessed. The recently emerged hydrovoltaic technology, especially moisture-induced electricity, shows great potential in harvesting energy from atmospheric water and gives birth to moisture energy harvesting devices. The device performance, especially the long-term operational capacity, has been significantly enhanced over the past few years. Further development;however, requires in-depth understanding of mechanisms, innovative materials, and ingenious system designs. In this review, beginning with describing the basic properties of water, the key aspects of the water-hygroscopic material interactions and mechanisms of power generation are discussed. The current material systems and advances in promising material development are then summarized. Aiming at the chief bottlenecks of limited operational time, advanced system designs that are helpful to improve device performance are listed. Especially, the synergistic effect of moisture adsorption and water evaporation on material and system levels to accomplish sustained electricity generation is discussed. Last, the remaining challenges are analyzed and future directions for developing this promising technology are suggested.

关键词： continuous power generation moisture adsorption moisture energy harvesting devices water evaporation

来源：评论

学校读者我要写书评

暂无评论

Online Hierarchical Linking of Action Tubes for Spatio-Temporal Action Detection Based on Multiple Clues

引用

IEEE ACCESS 2024年 12卷 54661-54672页

作者： Su, Shaowen Zhang, Yan Beijing Inst Technol Sch Automat State Key Lab Intelligent Control & Decis Complex Beijing 100081 Peoples R China

The spatio-temporal action detection task requires the output of the temporal and spatial positions as well as the action category of the target action instances in the form of action tubes. However, the current definition of video-level metrics in spatio-temporal action detection tasks is not sufficiently clear and unified to fully describe the ability of network models to perform spatio-temporal detection. Furthermore, existing tube linking methods are not only heavily dependent on the quality of the detection stage but also lack reliable linking criteria, resulting in poor tube linking performance. To address these issues, this study proposes a hierarchical linking method based on multiple clues. This method first dynamically utilizes various correlation clues at two levels, including appearance features, spatial overlap, motion prediction, category scores, tube length, and tube confidence status, to reduce the negative impact of unreliable information on the correlation. Then, it employs inter-class correlation to handle the mutual influence between different categories, followed by joint probability data association to address the mutual influence between correlated objects, ultimately achieving robust and accurate online linking of action tubes. The method is experimentally compared with other correlation methods on the untrimmed UCF24 and MultiSports datasets, demonstrating state-of-the-art tube link performance. We also conducted ablation experiments to explore the impact of the different modules and stages in the proposed tube-linking method.

关键词： MCHL spatio-temporal action detection linking method untrimmed video mAP

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：