检索结果-内蒙古大学图书馆

MPOC-SLAM: An RGB-D SLAM System With Motion Probability and Object Category in High Dynamic Environments

IEEE-ASME TRANSACTIONS ON MECHATRONICS 2025年第2期30卷 1061-1071页

作者： Wu, Songyang Zhang, Xuebo Zhang, Shiyong Song, Zhixing Wang, Runhua Yuan, Jing Nankai Univ Inst PLXSPACE Robot & Automat Informat Syst Coll Artificial Intelligence Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China

The performance of simultaneous localization and mapping (SLAM) is significantly impacted by dynamic objects. Existing dynamic SLAM methods still suffer from some major issues, such as insufficient use of semantic information, losing track when dynamic objects occupy the field of view, and inability to achieve real-time processing. In this article, we present SLAM with motion probability and object category (MPOC-SLAM), an RGB-D SLAM system capable of high-performance, real-time work in high dynamic environments. Considering intensive semantic information, our method introduces the concepts of motion probability (MP) and object category (OC), and tightly integrates them within the whole SLAM system. First, a priority feature extraction strategy is proposed based on MP, and then a dual-constraint feature matching method is designed with the adaptive probability threshold and category consistency. Both methods reduce the influence of dynamic objects in the extraction and association process. Second, to speed up the process of calculating reprojection errors, a semantic-guided geometric method with the partitioning hypothetico-deductive method is presented. The semantic-guided geometric method and instance segmentation are then fused through the variance weighted fusion with their confidences. Finally, as a more important part of SLAM, dynamic-aware weights are added to the original pose optimization, which assigns larger weights to static feature points and map points to reduce their alteration during optimization. We perform evaluations on the TUM public dataset and real-world environments. Compared with ORB-SLAM2, the percentage reduction of root mean square error of absolute trajectory error is as high as 98.72%. Results show that MPOC-SLAM realizes a significant improvement in accuracy as well as a more robust and real-time performance compared with other state-of-the-art dynamic SLAM methods.

关键词： Dynamic environments instance segmentation motion probability (MP) object category (OC) visual simultaneous localization and mapping (SLAM)

来源：评论

学校读者我要写书评

暂无评论

STC-TEB: Spatial-Temporally Complete Trajectory Generation Based on Incremental Optimization

引用

IEEE robotICS AND AUTOMATION LETTERS 2025年第2期10卷 1289-1296页

作者： Zhu, Zeqing Zhang, Qianyi Song, Yinuo Yang, Yifan Liu, Jingtai Inst Robot & Automat Informat Syst Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China Nankai Univ TBI Ctr Tianjin 300350 Peoples R China

In the context of indoor crowd navigation for mobile robots, the generation of spatial-temporally complete trajectories gives the robot a wider range of options, thereby improving the robustness and safety of navigation. Focused on this topic, this letter presents an incremental optimization framework that initially searches for complete trajectories in spatial dimension and subsequently optimizes them iteratively in spatial-temporal dimension. It benefits from a discount factor design to increase obstacle velocities incrementally, an adaptive determination of the discount factor step to ensure optimization robustness and a dragging force mechanism to adjust trajectories according to obstacle velocities at each step. Compared to the baseline algorithm EGO-TEB and Graphic-TEB, simulations and experiments conducted in lobbies and corridors, with both pedestrians and static obstacles, demonstrate that the proposed STC-TEB algorithm achieves the highest scenario success rate, the highest trajectory completeness rate, nearly the shortest time to the goal, and competitive optimization time.

关键词： Trajectory robots Optimization Pedestrians Collision avoidance Navigation Indexes Planning Robustness Force Human-aware motion planning crowd navigation spatial-temporally completeness incremental optimization collision avoidance

来源：评论

学校读者我要写书评

暂无评论

Mee-SLAM: Memory efficient endoscopic RGB SLAM with implicit scene representation

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 269卷

作者： Zhou, Yuchen Li, Teng Dai, Yu Zhang, Jianxun Nankai Univ Coll Artificial Intelligence Tianjin 300350 Peoples R China Inst Robot & Automat Informat Syst Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China

Endoscopic dense simultaneous localization and mapping (SLAM) plays a critical role in robot assisted surgery. Recently, SLAM systems based on neural implicit representation have demonstrated superior localization and real-time mapping capabilities. However, due to the memory constraint in medical use, the lack of true depth in monocular endoscopy, and the existence of interferences in the working environment, dense reconstruction that balances quality and efficiency remain challenges. In this paper, we propose a globally consistent RGB SLAM system for simultaneous pose optimization and memory-efficient real-time reconstruction. Our method consists of three key components. To achieve better pose estimation, a visual odometry is designed to address distractors, and integrated with a loop closure. Online full bundle adjustment establishes a globally consistent map and reduces memory consumption through proposed keyframe management. Based on dynamically updated poses, a memory-efficient real-time rendering module is proposed to generate high quality dense maps. Evaluations on a public EndoSLAM dataset and areal endoscopic surgery dataset demonstrate that the proposed method achieves a superior balance inaccuracy and efficiency, laying the foundation for real-time application of surgical robots.

关键词： Endoscopy Simultaneous localization and mapping Implicit neural network

来源：评论

学校读者我要写书评

暂无评论

RotInv-PCT: Rotation-Invariant Point Cloud Transformer via feature separation and aggregation

引用

NEURAL NETWORKS 2025年 185卷 107223页

作者： He, Cheng Zhao, Zhenjie Zhang, Xuebo Yu, Hang Wang, Runhua Nankai Univ Inst Robot & Automat Informat Syst Coll Artificial Intelligence Tianjin Peoples R China Nankai Univ Tianjin Key Lab Intelligent Robot Tianjin Peoples R China

The widespread use of point clouds has spurred the rapid development of neural networks for point cloud processing. A crucial property of these networks is maintaining consistent output results under random rotations of the input point cloud, namely, rotation invariance. The dominant approach achieves rotation invariance is to construct local coordinate systems for computing invariant local point cloud coordinates. However, this method neglects the relative pose relationships between local point cloud structures, leading to a decline in network performance. To address this limitation, we propose a novel Rotation-Invariant Point Cloud Transformer (RotInv-PCT). This method extracts the local abstract shape features of the point cloud using Local Reference Frames (LRFs) and explicitly computes the spatial relative pose features between local point clouds, both of which are proven to be rotation-invariant. Furthermore, to capture the long-range pose dependencies between points, we introduce an innovative Feature Aggregation Transformer (FAT) model, which seamlessly fuses the pose features with the shape features to obtain a globally rotation-invariant representation. Moreover, to manage large-scale point clouds, we utilize hierarchical random downsampling to gradually decrease the scale of point clouds, followed by feature aggregation through FAT. To demonstrate the effectiveness of RotInv-PCT, we conducted comparative experiments across various tasks and datasets, including point cloud classification on ScanObjectNN and ModelNet40, part segmentation on ShapeNet, and semantic segmentation on S3DIS and KITTI. Thanks to our provable rotation-invariant features and FAT, our method generally outperforms state-of-the-art networks. In particular, we highlight that RotInv-PCT achieved a 2% improvement in real-world point cloud classification tasks compared to the strongest baseline. Furthermore, in the semantic segmentation task, we improved the performance on the S3DIS datase

关键词： Deep learning Point cloud classification and segmentation Rotation invariance Transformer

来源：评论

学校读者我要写书评

暂无评论

NKhex: A New Miniature Hexapod Crawling robot With Visual Perception and Target Tracking

引用

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS 2025年第1期72卷 693-702页

作者： Wu, Yinan Deng, Haoyue Liu, Huawang Zhao, Xingqi Fang, Yongchun Inst Robot & Automat Informat Syst Tianjin 300350 Peoples R China Nankai Univ Coll Artificial Intelligence Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China

Scaling a large crawling robot to the centimeter-level size is a major technical challenge, which mainly includes the research on mechanical structure design, independent battery life, maneuverability, and perception. For most miniature crawling robots, it is difficult to synchronously balance the aspects of simple structure, good maneuverability, and visual perception. Aiming at such problems, this article designs a composite-cam- driven miniature hexapod crawling robot, named NKhex, with the abilities of visual perception and target tracking. The size of the microrobot is centimeter-level, and only two micromotors are used as drivers. The tripod gait motion of the robot can be realized with the six outriggers through the composite cam transmission mechanism. The robot is equipped with a miniature camera and has the functions of visual perception and object recognition via training neural networks, based on which it can achieve trajectory tracking and moving target tracking through closed-loop control. Besides, the robot is able to crawl and turn quickly in narrow space and outdoor environment, thus presenting potential applications in emergency rescue and data collection in dangerous areas. Sufficient experiments are carried out to verify the motion performance and perception ability of the robot.

关键词： robots Legged locomotion Motors Visual perception Target tracking Collision avoidance Task analysis Composite cam crawling robot hexapod microrobot visual perception

来源：评论

学校读者我要写书评

暂无评论

Enhanced robust performance oriented integrated planning and control of a robotic manipulator with online instantaneous violent disturbances

引用

CONTROL ENGINEERING PRACTICE 2025年 160卷

作者： Huo, Zixuan Yuan, Mingxing Huang, Junsheng Zhang, Shuaikang Zhang, Xuebo Nankai Univ Inst Robot & Automat Informat Syst Coll Artificial Intelligence Tianjin Peoples R China Nankai Univ Tianjin Key Lab Intelligent Robot Tianjin Peoples R China

robotic manipulators suffer from various modeling uncertainties, which generally deteriorate motion control performance. These uncertainties are commonly treated as a lumped disturbance which is then addressed by those disturbance estimation and attenuation control (DEAC) approaches. Although existing DEAC algorithms have shown their effectiveness of rejecting normal disturbances with moderate amplitudes and slow variations, they cannot work well in the presence of an instantaneous violent disturbance (IVD). An IVD is typically characterized by the short duration and large amplitude which exceeds the control input limit of a robotic joint. Consequently, the actual trajectory of a robotic manipulator will deviate from its desired trajectory significantly. Given this issue, an enhanced robust performance oriented two-loop framework which integrates minimum-time trajectory planning and nonlinear control is proposed in this paper. Specifically, a nonlinear adaptive robust controller is synthesized in the inner loop to handle both structured and unstructured uncertainties, while a synchronized trajectory planning algorithm is devised in the outer loop to force the deviated trajectory converging to the desired trajectory in minimum time. Comparative experiments on a robotic manipulator under IVDs show that the deviated trajectory is recovered fastest by the proposed approach.

关键词： Nonlinear control systems Adaptive robust control robotic manipulator Synchronized trajectory planning

来源：评论

学校读者我要写书评

暂无评论

A COLREGs-based path-planning method for collision avoidance considering path cost through reinforcement learning

引用

OCEAN ENGINEERING 2025年 325卷

作者： Song, Wanping Chen, Zengqiang Sun, Mingwei Wang, Yongshuai Sun, Qinglin Nankai Univ Coll Artificial Intelligence Tianjin 300350 Peoples R China Key Lab Intelligent Robot Tianjin Tianjin 300350 Peoples R China

The autonomous collision avoidance algorithm, a core element in path-planning, ensures the safe navigation of unmanned surface vehicles (USVs). This paper proposes a path-planning method based on the International Maritime Collision Avoidance Regulations (COLREGs), which can generate the yaw reference in real-time, considering the optimized path cost through reinforcement learning (RL). The COLREGs, as the surface vehicle code of conduct, help to establish an encounter model integrating three typical encounter scenarios and determine avoidance tasks. The path-planning method acquires adaptive capabilities, and optimizes the avoidance path cost via RL, for more efficient avoidance. The effectiveness of the path-planning method is verified by simulations, which show the optimal path cost and reduced yaw distance in static collision avoidance, as well as successful avoidance completion and compliance with the COLREGs in typical encounter scenarios.

关键词： Unmanned surface vehicle Autonomous systems Path planning Collision avoidance Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

robotic Fast Dual-Arm Patch Clamp System for Mechanosensitive Excitability Research of Neurons

引用

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 2025年第2期72卷 822-832页

作者： Ma, Biting Qiu, Jinyu Cui, Chaoyu Li, Ke Li, Ruimin Li, Minghui Liu, Yuzhu Fu, Shaojie Sun, Mingzhu Zhao, Xin Zhao, Qili Nankai Univ Inst Robot & Automatic Informat Syst Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China Nankai Univ Engn Res Ctr Trusted Behav Intelligence Minist Educ Tianjin 300350 Peoples R China Nankai Univ Inst Intelligence Technol & Robot Syst Shenzhen Res Inst Shenzhen 518083 Peoples R China Nankai Univ Natl Key Lab Intelligent Tracking & Forecasting In Key Lab Intelligent Robot 2Tianjin Tianjin Peoples R China Nankai Univ Natl Key Lab Intelligent Tracking & Forecasting In Tianjin 300350 Peoples R China Beijing Inst Technol Beijing Adv Innovat Ctr Intelligent Robots & Syst Beijing 100811 Peoples R China

Objective: A robotic fast dual-arm patch clamp system with controllable mechanical stimulation is proposed in this paper for mechanosensitive excitability research of neurons in brain slice. Methods: First, a kinematic model of a dual-arm patch clamp system combined with Monte Carlo method is developed to calculate the workspaces of recording micropipette and stimulation micropipette, and optimize the length of end effector for reducing collision incidences during operation. Then, a quantitative stimulation method to cells using one micropipette is developed based on pressing depth control. Finally, a fast robotic dual-arm patch clamp operation process is proposed based on a three-stage motion control of dual micropipettes to approach target cells and form whole-cell recording with quantitative mechanical stimulation. Results: Experimental results on 50 pyramidal neurons in the primary visual cortex of mouse brain slices demonstrate that this system achieves a threefold throughput with a 37% improvement in the success rate of the contact process and a 42% improvement in the success rate of whole-cell recording in comparison to manual operation. With these advantages, a mechanical stimulation-regulated increase in neuron excitability is observed in primary visual cortex. The experimental results also show that the sodium ion current may be more sensitive to mechanical stimulation than potassium ion current. Conclusion: Our system significantly improves the efficiency of mechanical stimulation induced excitability research of neurons in brain slices. Significance: Our methods have the potential to investigate pathological and pathogenic mechanisms of mechanosensitive ion channel dysfunction-induced diseases in the future.

关键词： Mechanical stimulation to neurons robotic patch clamp robotic cell manipulation workspace calculation whole-cell recording Mechanical stimulation to neurons robotic patch clamp robotic cell manipulation workspace calculation whole-cell recording

来源：评论

学校读者我要写书评

暂无评论

Autonomous Landing of the Quadrotor on the Mobile Platform via Meta Reinforcement Learning

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 2269-2280页

作者： Cao, Qianqian Liu, Ziyi Yu, Hai Liang, Xiao Fang, Yongchun Nankai Univ Coll Artificial Intelligence Inst Robot & Automat Informat Syst Tianjin 300350 Peoples R China Nankai Univ Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China

Landing a quadrotor on a mobile platform moving with various unknown trajectories presents special challenges, including the requirements of fast trajectory planning/replanning, accurate control, and the adaptability for different target trajectories, especially when the platform is non-cooperative. However, previous works either assume the platform moves along a predefined trajectory or decouple planning from control which may cause a delay in tracking. In this work, we integrate planning and control into a unified framework and present an efficient off-policy Meta-Reinforcement Learning (Meta-RL) algorithm that enables a quadrotor (agent) to land on a mobile platform with various unknown trajectories autonomously. In our approach, we disentangle task-specific policy parameters by a separate adapter network to shared low-level parameters and learn a probabilistic encoder to extract common structures across different tasks. Specifically, during meta-training, we sample different trajectories from the task distribution, and then the probabilistic encoder accumulates the necessary statistics from past experience into the latent variables that enable the policy to perform the task. At meta-testing time, when the quadrotor is faced with an unseen trajectory, the latent variables can be sampled according to past interactions between the quadrotor and the mobile platform and held constant during an episode, enabling rapid trajectory-level adaptation. We assume similar tasks share a common low-dimensional structure in the representation of the policy network and the task-specific information is learned in the head of the policy. Accordingly, we further propose a separate adapter net as a supervised learning problem. The adapter net learns the weights of the policy's output layer for each meta-training task given by the environment interactions from the agent. When adapting to a new task during meta-testing, we fix the shared model layers and predict the head weights for th

关键词： Quadrotor meta reinforcement learning autonomous landing trajectory planning and control

来源：评论

学校读者我要写书评

暂无评论

G²VD Planner: Efficient Motion Planning WithGrid-Based Generalized Voronoi Diagrams

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 3743-3755页

作者： Wen, Jian Zhang, Xuebo Bi, Qingchen Liu, Hui Yuan, Jing Fang, Yongchun Nankai Univ Inst Robot & Automat Informat Syst Coll Artificial Intelligence Tianjin 300350 Peoples R China Nankai Univ Tianjin Key Lab Intelligent Robot Tianjin 300350 Peoples R China

In this paper, an efficient motion planning approach with grid-based generalized Voronoi diagrams (G( 2) VD) is newly proposed for mobile robots. Different from existing approaches, the novelty of this work is twofold: 1) a new state lattice-based path searching approach is proposed, in which the search space is reduced to a novel Voronoi corridor to further improve the search efficiency;2) an efficient quadratic programming-based path smoothing approach is presented, wherein the clearance to obstacles is considered to improve the path clearance of hard-constrained path smoothing approaches. We validate the efficiency and smoothness of our approach in various challenging simulation scenarios and outdoor environments. It is shown that the computational efficiency is improved by 17.1% in the path searching stage, and path smoothing with the proposed approach is 6.6 times faster than an advanced sparse-banded structure-based path smoothing approach and 53.3 times faster than the popular timed-elastic-band planner. A video showing outdoor navigation on our campus is available at https://***/iMXGthgvp58.

关键词： Smoothing methods Planning robots Search problems Mobile robots Safety Heuristic algorithms Autonomous navigation mobile robots motion planning path planning path optimization

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：