检索结果-内蒙古大学图书馆

A Novel Pose Estimation Method of Object in robotic Manipulation Using Vision-Based Tactile Sensor 5th

学校读者我要写书评

暂无评论

A Novel Pose Estimation Method of Object in Robotic Manipula...

5th International Conference on Cognitive Systems and Signal Processing, ICCSIP 2020

作者： Zhao, Dan Sun, Fuchun Zhou, Quan Wang, Zongtao Beijing National Research Center for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing100083 China AnHui Province Key Laboratory of Special Heavy Load Robot Anhui University of Technology Ma Anshan China Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University Qinhuangdao066000 China

ISBN: (纸本)9789811623356

The computer vision techniques have been widely used in the robotic manipulation for perception and positioning. However, duo to the inaccuracy of camera calibration and measurement, there is a greatly need for more accurate method to estimate the object pose especially for robotic precise manipulation. In this paper, we propose a novel pose estimation method of object in robotic manipulation using a vision-based tactile sensor. The structure of the tactile sensor and the proposed model are introduced in detail, and verification experiments have been performed. The results show the advantages and better performance of the proposed method. © 2021, Springer Nature Singapore Pte Ltd.

关键词： robot vision

Offboard Occupancy Refinement with Hybrid Propagation for Autonomous Driving

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Shi, Hao Wang, Song Zhang, Jiaming Yin, Xiaoting Wang, Zhongdao Wang, Guangming Zhu, Jianke Yang, Kailun Wang, Kaiwei The State Key Laboratory of Extreme Photonics and Instrumentation The National Engineering Research Center of Optical Instrumentation Zhejiang University Hangzhou310027 China The School of Robotics The National Engineering Research Center of Robot Visual Perception and Control Technology Hunan University Changsha410082 China The College of Computer Science and Technology Zhejiang University Hangzhou310027 China The Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology Karlsruhe76131 Germany Huawei Noah’s Ark Lab. Canada The Department of Engineering University of Cambridge CambridgeCB2 1PZ United Kingdom

Vision-based occupancy prediction, also known as 3D Semantic Scene Completion (SSC), presents a significant challenge in computer vision. Previous methods, confined to onboard processing, struggle with simultaneous geometric and semantic estimation, continuity across varying viewpoints, and single-view occlusion. Our paper introduces OccFiner, a novel offboard framework designed to enhance the accuracy of vision-based occupancy predictions. OccFiner operates in two hybrid phases: 1) a multi-to-multi local propagation network that implicitly aligns and processes multiple local frames for correcting onboard model errors and consistently enhancing occupancy accuracy across all distances. 2) the region-centric global propagation, focuses on refining labels using explicit multi-view geometry and integrating sensor bias, especially to increase the accuracy of distant occupied voxels. Extensive experiments demonstrate that OccFiner improves both geometric and semantic accuracy across various types of coarse occupancy, setting a new state-of-the-art performance on the SemanticKITTI dataset. Notably, OccFiner elevates vision-based SSC models to a level even surpassing that of LiDAR-based onboard SSC models. Furthermore, OccFiner is the first to achieve automatic annotation of SSC in a purely vision-based approach. Quantitative experiments prove that OccFiner successfully facilitates occupancy data loop-closure in autonomous driving. Additionally, we quantitatively and qualitatively validate the superiority of the offboard approach on city-level SSC static maps. The source code will be made publicly available at https://***/MasterHow/OccFiner. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

$\mathcal {H}_{\infty }$ State Estimation for PDT-Switched Coupled Neural Networks Under Round-Robin Protocol: A Cooperation-Competition-Based Mechanism

学校读者我要写书评

暂无评论

IEEE Transactions on Network Science and engineering 2022年第2期10卷 911-921页

作者： Hao Shen Yinsheng Song Jing Wang Ju H. Park AnHui Province Key Laboratory of Special Heavy Load Robot Anhui University of Technology Ma'anshan China Key Laboratory of Advanced Control and Optimization for Chemical Processes Ministry of Education East China University of Science and Technology Shanghai China School of Electrical and Information Engineering Anhui University of Technology Ma'anshan China School of Information Science and Engineering East China University of Science and Technology Shanghai China Department of Electrical Engineering Yeungnam University Kyongsan Republic of Korea

This paper is concentrated on the $\mathcal {H_{\infty }}$ state estimation problem for switched coupled neural networks based on a Takagi-Sugeno fuzzy model. Notably, the time-variant network topology with alternate fast switchings and slow ones is described suitably by a persistent dwell-time rule, and the interactive dynamics with both cooperative properties and antagonistic ones among nodes are featured comprehensively by the switching signed graph. In view of the communication pressures brought by network-induced problems and the requirements in digital control, the round-robin protocol and logarithmic quantization are flexibly integrated for more transmission efficiency and fewer data collisions. Thereafter, by utilizing a relaxed multiple Lyapunov function method and some novel matrix process techniques, sufficient criteria guaranteeing the exponential stability in a globally uniform sense with a prescribed $\mathcal {H_{\infty }}$ performance level of the estimation error system are established. Finally, the synthesized analysis of the proposed method is presented with an illustrative example.

关键词： Switches Topology Switching circuits Protocols Neurons State estimation Quantization (signal)

MADDPG Algorithm for Coordinated Welding of Multiple robots

学校读者我要写书评

暂无评论

MADDPG Algorithm for Coordinated Welding of Multiple Robots

作者： Weibin Chen Lei Hua Leixin Xu Benshun Zhang Mengmeng Li Tao Ma Yang-Yang Chen School of Automation Southeast University Robot Department Jiangsu Automation Research Institute School of Automation Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University

This paper deals with the coordinated welding problem of multi-robot systems by applying a deep reinforcement learning algorithm which is called multi-agent deep deterministic policy gradient(MADDPG).It is assumed that the states and actions of robots are continuous and each robot can only get local information of its neighbors.A novel reward composed of trajectory optimization,coordinated welding and collision avoidance is designed,which yield multi-robot systems to arrive the welding targets quickly,achieve the simultaneous welding for a weld line and avoid collision among *** results are provided to demonstrate the effectiveness of the proposed control algorithm.

关键词： Deep multi-agent reinforcement learning MADDPG Coordinated welding

Measurement of surface stress-strain distribution inside a grain of carbon steel via optical microscope

学校读者我要写书评

暂无评论

Measurement 2025年 255卷

作者： I Dewa Made Oka Dharmawan Jinyi Lee Department of Control Instrumentation and Robot Engineering Graduate School of Chosun University Gwangju 61452 Republic of Korea Interdisciplinary Program in IT-BIO Convergence System Chosun University Gwangju 61452 Republic of Korea Department of Electronic Engineering Chosun University Gwangju 61452 Republic of Korea

The advancement of alloys and steel in engineering highlights the critical need for robust methods to ensure structural integrity. While traditional Non-Destructive Testing and Evaluation (NDT/E) techniques provide valuable insights, they often fail to detect pre-existing stress conditions that contribute to fatigue cracks. Recent approaches have addressed this limitation by analyzing stress and strain at the grain level using advanced measurement systems—often based on diffraction techniques—integrated with Finite Element Method (FEM) and Digital Image Correlation (DIC) platforms. However, the application of these techniques for in-service inspections remains limited by challenges in accessibility, cost, and scalability. To overcome these issues, an integrated 2D surface system combining optical microscopy with DIC and grain-level analysis within the FEM framework is proposed. This method enables efficient visualization of grain-level deformation and stress distribution, thereby providing deeper insights into the microstructural behavior of polycrystalline materials under external loads. The proposed methodology is theoretically outlined and validated through simulations and experimental analyses, demonstrating its potential as a practical solution for in-service inspections. Ultimately, this approach is expected to advance inspection techniques by offering an accessible and scalable tool for monitoring microstructural behavior and identifying potential failure mechanisms through surface analysis.

关键词：

CoBEV: Elevating Roadside 3D Object Detection with Depth and Height Complementarity

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Shi, Hao Pang, Chengshan Zhang, Jiaming Yang, Kailun Wu, Yuhao Ni, Huajian Lin, Yining Stiefelhagen, Rainer Wang, Kaiwei The State Key Laboratory of Extreme Photonics and Instrumentation Zhejiang University Hangzhou310027 China The Institute for Anthropomatics and Robotics Karlsruhe Institute of Technology Karlsruhe76131 Germany The Department of Engineering Science University of Oxford OxfordOX1 3PJ United Kingdom The School of Robotics Hunan University Changsha410012 China The National Engineering Research Center of Robot Visual Perception and Control Technology Hunan University Changsha410082 China Shanghai SUPREMIND Technology Co. Ltd. Shanghai201210 China

Roadside camera-driven 3D object detection is a crucial task in intelligent transportation systems, which extends the perception range beyond the limitations of vision-centric vehicles and enhances road safety. While previous studies have limitations in using only depth or height information, we find both depth and height matter and they are in fact complementary. The depth feature encompasses precise geometric cues, whereas the height feature is primarily focused on distinguishing between various categories of height intervals, essentially providing semantic context. This insight motivates the development of Complementary-BEV (CoBEV), a novel end-to-end monocular 3D object detection framework that integrates depth and height to construct robust BEV representations. In essence, CoBEV estimates each pixel's depth and height distribution and lifts the camera features into 3D space for lateral fusion using the newly proposed two-stage complementary feature selection (CFS) module. A BEV feature distillation framework is also seamlessly integrated to further enhance the detection accuracy from the prior knowledge of the fusion-modal CoBEV teacher. We conduct extensive experiments on the public 3D detection benchmarks of roadside camera-based DAIR-V2X-I and Rope3D, as well as the private Supremind-Road dataset, demonstrating that CoBEV not only achieves the accuracy of the new state-of-the-art, but also significantly advances the robustness of previous methods in challenging long-distance scenarios and noisy camera disturbance, and enhances generalization by a large margin in heterologous settings with drastic changes in scene and camera parameters. For the first time, the vehicle AP score of a camera model reaches 80% on DAIR-V2X-I in terms of easy mode. The source code will be made publicly available at CoBEV. Copyright © 2023, The Authors. All rights reserved.

关键词： Feature Selection

A new calibration method and experimental study for kinematic parameters of industrial robot

学校读者我要写书评

暂无评论

A new calibration method and experimental study for kinemati...

2020 International Conference on Artificial Intelligence and Electromechanical Automation, AIEA 2020

作者： Zhao, Honghua Zhao, Jian Lin, Yifei Han, Qifei Chen, Mianpeng University of Jinan School of Mechanical Engineering Jinan China Shandong University School of Control Science and Engineering Jinan China Shandong Institute of Intelligent Robot and Applied Technology Jining China GSK CNC Equipment Co. Ltd. CNC System R and D Department Guangzhou China

ISBN: (纸本)9781728182889

The accuracy of industrial robot kinematics parameters directly affects its absolute positioning accuracy. Based on Modify Denavit-Hartenberg (MD-H) kinematics theory, a more accurate calibration method for kinematic parameters of industrial robots is proposed in this paper. It takes the kinematics parameters, deceleration ratio and coupling coefficient of the robot as error parameters to establish the error model. The calibration experiment and accuracy verification experiment are designed by laser tracker. After the introduction of the reduction ratio and the coupling coefficient calibration, the absolute positioning error of the robot in all directions has been greatly improved, and the overall average accuracy of the robot is increased by 91.42% compared with the uncalibrated, and is increased by 81.02% compared with the without introducing the deceleration ratio and coupling coefficient. © 2020 IEEE.

关键词： Kinematics

Learning compliant grasping and manipulation by teleoperation with adaptive force control

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zeng, Chao Li, Shuang Jiang, Yiming Li, Qiang Chen, Zhaopeng Yang, Chenguang Zhang, Jianwei Department of Informatics Universiẗat Hamburg Germany National Engineering Laboratory for Robot Visual Perception and Control Hunan University China Center for Cognitive Interaction Technology Bielefeld University Germany Bristol Robotics Laboratory University of the West of England United Kingdom

In this work, we focus on improving the robot's dexterous capability by exploiting visual sensing and adaptive force control. TeachNet, a vision-based teleoperation learning framework, is exploited to map human hand postures to a multifingered robot hand. We augment TeachNet, which is originally based on an imprecise kinematic mapping and position-only servoing, with a biomimetic learning-based compliance control algorithm for dexterous manipulation tasks. This compliance controller takes the mapped robotic joint angles from TeachNet as the desired goal, computes the desired joint torques. It is derived from a computational model of the biomimetic control strategy in human motor learning, which allows adapting the control variables (impedance and feedforward force) online during the execution of the reference joint angle trajectories. The simultaneous adaptation of the impedance and feedforward profiles enables the robot to interact with the environment in a compliant manner. Our approach has been verified in multiple tasks in physics simulation, i.e., grasping, opening-a-door, turning-a-cap, and touching-a-mouse, and has shown more reliable performances than the existing position control and the fixed-gain-based force control approaches. Copyright © 2021, The Authors. All rights reserved.

关键词： Remote control

Collision-free formation control of multiple nano-quadrotors

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Nguyen, Anh Tung Lee, Ji-Won Nguyen, Thanh Binh Hong, Sung Kyung Korea Republic of Human-Robot Interaction Research Center Korea Institute of Robotics and Technology Convergence Pohang37553 Korea Republic of Department of Control Engineering and Automation Thuyloi University 175 Tay Son Dong Da Hanoi Viet Nam

The utilisation of unmanned aerial vehicles has witnessed significant growth in real-world applications including surveillance tasks, military missions, and transportation deliveries. This letter investigates practical problems of formation control for multiple nano-quadrotor systems. To be more specific, the first aim of this work is to develop a theoretical framework for the time-varying formation flight of the multi-quadrotor system regarding anti-collisions. In order to achieve this goal, the finite cut-off potential function is devoted to avoiding collisions among vehicles in the group as well as between vehicles and an obstacle. The control algorithm navigates the group of nano-quadrotors to asymptotically reach an anticipated time-varying formation. The second aim is to implement the proposed algorithm on Crazyflies nano-quadrotors, one of the most ubiquitous indoor experimentation platforms. Several practical scenarios are conducted to tendentiously expose anti-collision abilities among group members as well as between vehicles and an obstacle. The experimental outcomes validate the effectiveness of the proposed method in the formation tracking and the collision avoidance of multiple nano-quadrotors. © 2021, CC BY.

关键词： Collision avoidance