检索结果-内蒙古大学图书馆

IEEE International Conference on Mechatronics and Automation

作者： Biying Xu Xuehe Zhang Xuan Yu Hegao Cai Kuan Zhang Jie Zhao Jizhuang Fan State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798350388077

ISBN: (纸本)9798350388084

The steam generator tube sheet quadruped maintenance robot is a novel mobile robot designed for the inspection of steam generator heat transfer tubes. This paper presents a motion planning method for this type of robot. The trajectory of the robot's base is planned through kinematic solutions search. Additionally, a foothold motion evaluation system is established using the Technique for Order of Preference by Similarity to Ideal Solution algorithm to plan the movement of foothold points. For a given motion path, our proposed method divides the path into multiple straight-line segments, and then performs separate planning for turning foothold points and straight-line foothold points. The experiments demonstrate that the proposed algorithm is adaptable to steam generator tube sheets and performs well in environments with obstacles since the algorithm proposed has a planning success rate of 100% for the tube sheet robot. The experiments validate the effectiveness and feasibility of the proposed method.

关键词： Robot motion Turning Generators Electron tubes Planning Maintenance Trajectory

来源：评论

学校读者我要写书评

暂无评论

Helium Loss for Robust keypoints Detection with Convolutional Neural Networks

Helium Loss for Robust Keypoints Detection with Convolutiona...

引用

Automation, Control and robotics Engineering (CACRE), International Conference on

作者： Haode Huo Tianran Wang Linjie Zhao State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798350350302

ISBN: (纸本)9798350350319

In this paper, a novel loss function for key point detection, named Helium Loss is proposed. All our experiments are based on Ultralytics' YOLOv8n-pose architecture. Initially, this paper analyze various existing loss functions, including L1 loss, L2 loss, and smooth L1 loss. There is a problem that current loss functions do not account for issues related to the quality of dataset annotations and unclear boundaries of targets. To address this challenge, this paper introduce a loss function that incorporates partial trust in annotated key points and utilizes Gaussian distribution to mitigate the impact of manual annotations.

关键词： Training Computer vision Annotations Computational modeling Pose estimation Computer architecture Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

A Robotic system for Brachytherapy: Validation in Animal Experiments

A Robotic System for Brachytherapy: Validation in Animal Exp...

引用

IEEE International Conference on Control and robotics Engineering (ICCRE)

作者： Changle Li Leifeng Zhang Xuehe Zhang Gangfeng Liu Chong Yao State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798350372694

ISBN: (纸本)9798350372700

In this paper we proposed a brachytherapy robotic system. The system uses a 7-Dof robotic arm and a set of tools to complete puncture and seeds implantation. We designed end tools for puncture and seed implantation separately. A workflow based on this system has also been proposed. Finally, we used this system to conduct six animal experiments. The experimental results showed that the average error of seed implantation was 2.48mm, the minimum error is only 0.5mm, the maximum error reached 5.2mm.

关键词： Animals Manipulators Brachytherapy

来源：评论

学校读者我要写书评

暂无评论

RAM-NAS: Resource-aware Multiobjective Neural Architecture Search Method for Robot Vision Tasks

RAM-NAS: Resource-aware Multiobjective Neural Architecture S...

引用

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Shouren Mao Minghao Qin Wei Dong Huajian Liu Yongzhuo Gao State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798350377705

ISBN: (纸本)9798350377712

Neural architecture search (NAS) has shown great promise in automatically designing lightweight models. However, conventional approaches are insufficient in training the supernet and pay little attention to actual robot hardware resources. To meet such challenges, we propose RAM-NAS, a resource-aware multi-objective NAS method that focuses on improving the supernet pretrain and resource-awareness on robot hardware devices. We introduce the concept of subnets mutual distillation, which refers to mutually distilling all subnets sampled by the sandwich rule. Additionally, we utilize the Decoupled Knowledge Distillation (DKD) loss to enhance logits distillation performance. To expedite the search process with consideration for hardware resources, we used data from three types of robotic edge hardware to train Latency Surrogate predictors. These predictors facilitated the estimation of hardware inference latency during the search phase, enabling a unified multi-objective evolutionary search to balance model accuracy and latency trade-offs. Our discovered model family, RAM-NAS models, can achieve top-1 accuracy ranging from 76.7% to 81.4% on ImageNet. In addition, the resource-aware multi-objective NAS we employ significantly reduces the model’s inference latency on edge hardware for robots. We conducted experiments on downstream tasks to verify the scalability of our methods. The inference time for detection and segmentation is reduced on all three hardware types compared to MobileNetv3-based methods. Our work fills the gap in NAS for robot hardware resource-aware.

关键词： Training Accuracy Image edge detection Scalability Robot vision systems Predictive models Hardware Neural architecture search Robots Testing

来源：评论

学校读者我要写书评

暂无评论

Large Fixture Calibration Method Based on Point Cloud Scanning for Robotic Drilling

Large Fixture Calibration Method Based on Point Cloud Scanni...

引用

Automation, Control and robotics Engineering (CACRE), International Conference on

作者： Yongzhuo Gao Haibo Gao Dexin Cheng Wei Dong the State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798350350302

ISBN: (纸本)9798350350319

The rocket fuel tank requires the splicing of multiple sub-parts, which is difficult to be processed in an integrated manner. Among them, the short shell parts of the storage tank play a linking role and need to be drilled on the surface. Due to the large size of the tooling of the parts, there are problems such as large installation and positioning errors and low precision of robot automatic drilling. Therefore, a robot automatic hole making calibration method is proposed to solve the current problems. In order to achieve automated control, a large-scale robot automatic drilling system must have large-scale flexible tooling, control algorithms, human-computer interaction interfaces, execution units, and integrated tool rooms. It is essential to build effective calibration procedures as a foundation for this automation. The calibration plane is fitted, positioned, and calibrated using the point cloud scanning method. Subsequently, the hand-eye calibration method is employed to establish the relationship between the coordinate systems. Using this approach, the calibration process is ultimately finished, resulting in the successful implementation of automated drilling. The experimental findings demonstrate that, in the designated quadrant, the drilling rate of the robot can achieve 5 holes per minute. Furthermore, the average variance of the hole position is below 0.2mm, and the highest error is 0.5mm. These results align with the requirements of practical production.

关键词： Drilling Point cloud compression Rockets Automation Service robots Robot kinematics Splicing

来源：评论

学校读者我要写书评

暂无评论

PMM-Net: Single-stage Multi-agent Trajectory Prediction with Patching-based Embedding and Explicit Modal Modulation

arXiv

引用

arXiv 2024年

作者： Liu, Huajian Dong, Wei Fan, Kunpeng Wang, Chao Gao, Yongzhuo State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

Analyzing and forecasting trajectories of agents like pedestrians plays a pivotal role for embodied intelligent applications. The inherent indeterminacy of human behavior and complex social interaction among a rich variety of agents make this task more challenging than common time-series forecasting. In this letter, we aim to explore a distinct formulation for multi-agent trajectory prediction framework. Specifically, we proposed a patching-based temporal feature extraction module and a graph-based social feature extraction module, enabling effective feature extraction and cross-scenario generalization. Moreover, we reassess the role of social interaction and present a novel method based on explicit modality modulation to integrate temporal and social features, thereby constructing an efficient single-stage inference pipeline. Results on public benchmark datasets demonstrate the superior performance of our model compared with the state-of-the-art methods. The code is available at: ***/TIB-K330/pmm-net. Copyright © 2024, The Authors. All rights reserved.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

Development of a Cable-Driven Upper Limb Exosuit for Both Pulling and Lifting

Development of a Cable-Driven Upper Limb Exosuit for Both Pu...

引用

Mechatronics and robotics Engineering (ICMRE), International Conference on

作者： Songhao Guo Tianjiao Zheng Haotian Ju Jie Zhao Yanhe Zhu State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798350394276

ISBN: (纸本)9798350394283

This paper presents an upper limb exosuit for human augmentation. The exosuit utilizes a low-cost air pressure sensing glove based on the principle of air pressure to detect hand grasping states. It achieves assistance in upper limb elbow flexion movements during activities such as pulling and lifting through a driving module and wire rope for power transmission. To verify the effectiveness of the upper limb exosuit, pull-ups and lifting experiments were performed on five healthy participants with and without the exosuit. The surface electromyographic (sEMG) signals were recorded from selected muscles (Brachioradialis (BR), biceps brachii (BB) and latissimus dorsi (LD)) to assess how well the exosuit is helping the body. The results indicate a significant decrease in both the mean and peak value of the average normalized sEMG signals of the selected muscles during the pullup and load-lifting experiments. The finding indicates that the designed upper limb exosuit system can reduce muscle activation during pulling up and lifting tasks, effectively providing assistance. Future research directions include further optimizing the exosuit’s mechanical structure, improving the air pressure sensing glove’s state recognition accuracy and validating the application effectiveness of the exoskeleton suit in specific scenarios such as rock climbing and rope climbing.

关键词： Mechatronics Wires Power transmission Grasping Muscles Robot sensing systems Rocks

来源：评论

学校读者我要写书评

暂无评论

Design and Analysis of Cellular Multidimensional Morphing Linkage of Aircraft Fuselage

Design and Analysis of Cellular Multidimensional Morphing Li...

引用

Mechatronics and robotics Engineering (ICMRE), International Conference on

作者： Hualiang Liu Hong Xiao Guang Yang Hongwei Guo Rongqiang Liu Zongquan Deng State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

ISBN: (数字)9798350394276

ISBN: (纸本)9798350394283

Firstly, this paper proposes two types of variable cross-section morphing for the aircraft fuselage: upper convex and below convex morphing. Secondly, an innovative design of a cellular multidimensional morphing linkage combining planar six-bar linkages and a 6-UPS parallel linkage is proposed. The DOF of the linkage is analyzed, and kinematic simulation analysis of the morphing linkage is carried out. Finally, a prototype of the morphing fuselage is developed, and through morphing capacity test, it is verified that the morphing fuselage can achieve reversible upper convex and below convex morphing, with maximum radial morphing rates of 11.9% and 11.4%, verifying the feasibility of the linkage design.

关键词： Couplings Mechatronics Simulation Prototypes Kinematics Aircraft Robots

来源：评论

学校读者我要写书评

暂无评论

Aggressive and robust low-level control and trajectory tracking for quadrotors with deep reinforcement learning

引用

Neural Computing and Applications 2025年第3期37卷 1223-1240页

作者： Chen, Shiyu Li, Yanjie Lou, Yunjiang Lin, Ke Department of Control Science and Engineering Harbin Institute of Technology Shenzhen Shenzhen518055 China Guangdong Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics Shenzhen518055 China

Executing accurate trajectory tracking tasks using a high-performance low-level controller is crucial for quadrotors to be applied in various scenarios, especially those involving uncertain disturbances. However, due to the uncertainties in disturbed environments, developing effective low-level controllers with traditional model-based control schemes is challenging. This paper presents an aggressive and robust reinforcement learning (RL)-based low-level control policy for quadrotors. The policy maps the observed quadrotor state directly to motor thrust commands, without requiring the quadrotor dynamics. Additionally, a trajectory generation pipeline is developed to improve the accuracy of trajectory tracking tasks based on differential flatness. With the learned low-level control policy, extensive simulations and real-world experiments are implemented to validate the performance of the policy. The results indicate that our RL-based low-level control policy outperforms traditional proportional–integral–derivative (PID) control methods and related learning-based policies in terms of accuracy and robustness, particularly in environments with uncertain disturbances. Furthermore, the proposed RL-based control policy exhibits an aggressive response in trajectory tracking, even when the speed of the desired trajectory is increased to 6 m/s. Moreover, the learned policy demonstrates strong vibration suppression capabilities and enables the quadrotor to recover to a hovering state from random initial conditions with shorter response time. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Development of an Integrated Grapple Chain for a Simultaneous Three-Fingered End-Effector 16th

Development of an Integrated Grapple Chain for a Simultaneou...

引用

16th International Conference on Intelligent robotics and Applications, ICIRA 2023

作者： Wu, Jun He, Yu Wang, He Fan, Shaowei School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu611731 China State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin150001 China

ISBN: (纸本)9789819965007

This paper presents the development of an integrated grapple chain for a simultaneous three-fingered end-effector. The geometrical constraint with four stages is firstly defined. Then, a grapple finger is presented to fulfill the proposed four stages with an approximately straight line actuation. Then, the actuation chain of linkage is designed by modification, optimization and combination. At last, Simulation indicates that the grapple finger can conduct planar and linear motion, which contributed to restrain the target gradually. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： End effectors

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：