Article
Engineering, Electrical & Electronic
Huei-Yung Lin, Shih-Cheng Liang, Yu-Kai Chen
Summary: This paper introduces a robotic grasping system with multi-view depth image acquisition, utilizing a series of algorithms for noise removal and target pose estimation to ultimately increase grasping efficiency and demonstrate feasibility through experimentation.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Civil
Yujie Li, Zhiyun Yin, Yuchao Zheng, Huimin Lu, Tohru Kamiya, Yoshihisa Nakatoh, Seiichi Serikawa
Summary: In this paper, a new multi-class dataset ICD-4 is proposed for 6D object pose estimation. An innovative pose estimation network called PoseMLP is also introduced, which uses residual MLP modules to directly predict the 6D pose estimation. Experimental results demonstrate the effectiveness and reliability of the proposed method.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Munkhtulga Byambaa, Gou Koutaki, Lodoiravsal Choimaa
Summary: This study proposes a method for 6D pose estimation of transparent objects, which uses a deep neural network to estimate 2D keypoints and uses the PnP algorithm to estimate the 6D pose of the object. The experiments demonstrate that the method is capable of grasping transparent objects from different backgrounds and outperforms other 6D pose estimation methods, with the potential to be applied to real-world images.
Article
Computer Science, Artificial Intelligence
Shuangjie Yuan, Zhenpeng Ge, Lu Yang
Summary: This paper presents a Single-Camera Multi-View (SCMV) method that utilizes a fixed monocular camera and the initiative motion of a robotic manipulator to capture multi-view RGB image sequences, achieving more accurate 6DoF pose estimation results.
FRONTIERS IN NEUROROBOTICS
(2023)
Article
Robotics
Zhenwei Liu, Junyi Geng, Xikai Dai, Tomasz Swierzewski, Kenji Shimada
Summary: This study introduces a robotic system that automatically removes unfused powder from the surface of 3D-printed parts. The system utilizes a visual perception technology that can track the position of the parts in real-time and estimate the completion percentage of depowdering. Experiments show that the system is capable of handling parts with various shapes without causing any damage.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Automation & Control Systems
Hae-June Park, Bo-Hyeon An, Su-Bin Joo, Oh-Won Kwon, Min Young Kim, Joonho Seo
Summary: This paper introduces an algorithm to control a robotic prosthetic hand using deep learning to select grasping pose and time from 2D images and 3D point clouds. The algorithm consists of four steps: acquiring images and point clouds, object recognition, pose selection, and time determination.
INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS
(2022)
Article
Computer Science, Interdisciplinary Applications
Chungang Zhuang, Shaofei Li, Han Ding
Summary: This article proposes a deep learning-based pose estimation method using point cloud as input, which includes instance segmentation and instance point cloud pose estimation. Experimental results demonstrate that this method can effectively and robustly predict the poses of objects in cluttered and occluded scenes.
ROBOTICS AND COMPUTER-INTEGRATED MANUFACTURING
(2023)
Article
Robotics
Dong Li, Quan Mu, Yilin Yuan, Shiwei Wu, Ye Tian, Hualin Hong, Qian Jiang, Fei Liu
Summary: This letter presents a novel method for achieving high precision 6D pose estimation by exploiting the reprojection of 3D edges onto binocular RGB image pairs. The proposed method includes detection, pose initialization, and pose refinement phases, improved initial pose estimation network and a novel pose optimization technique are introduced to enhance the accuracy and precision of the estimations. Experimental results demonstrate the effectiveness of the method.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Robotics
Haixin Yu, Shoujie Li, Houde Liu, Chongkun Xia, Wenbo Ding, Bin Liang
Summary: In this research, we propose TGF-Net, a monocular instance-level 6D pose estimation method for transparent objects based on geometric fusion. TGF-Net learns edge features and surface fragments of transparent objects to reduce the influence of appearance changes on pose estimation. We also generate a high-fidelity synthetic dataset called Trans6D-32 K using Blender, containing rendered RGB images and poses information of transparent objects in various backgrounds, perspectives, and lighting conditions. Experimental results on Trans6D-32 K dataset and real-world scenarios demonstrate the good performance and application value of TGF-Net in transparent object manipulation.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Agriculture, Multidisciplinary
Xing Wang, Hanwen Kang, Hongyu Zhou, Wesley Au, Chao Chen
Summary: Field robotic harvesting is a promising technique in agricultural industry. This study proposes a geometry-aware network, A3N, and a global-to-local scanning strategy to enable robots to accurately recognize and retrieve fruits in complex field environments.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Computer Science, Information Systems
Junyan Ge, Lingbo Mao, Jinlong Shi, Yan Jiang
Summary: In this paper, a visual strategy is designed to solve the key problem of realizing stable robotic grasping in cluttered scenes. This strategy improves the instability of grasping caused by stacking objects through steps such as generating synthetic dataset, fusion of color information and re-encoded depth information, and point cloud registration, realizing 6-DoF grasping.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Automation & Control Systems
Penglei Liu, Qieshi Zhang, Jun Cheng
Summary: This article introduces a new method to predict the 6D pose of objects from RGB images using a depth prediction network and fusion of depth and RGB information. The proposed approach is evaluated on three benchmark datasets and applied on the UR5 robot platform, showing its effectiveness.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Review
Computer Science, Artificial Intelligence
Guoguang Du, Kai Wang, Shiguo Lian, Kaiyong Zhao
Summary: This paper provides a comprehensive overview of vision-based robotic grasping, focusing on key tasks such as object localization, object pose estimation, and grasp estimation. Various methods combining these tasks, including traditional and deep learning-based approaches, are reviewed in detail. Challenges and future directions in the field are also highlighted.
ARTIFICIAL INTELLIGENCE REVIEW
(2021)
Article
Computer Science, Information Systems
Ji Liang, Jiguang Zhang, Bingbing Pan, Shibiao Xu, Guangheng Zhao, Ge Yu, Xiaopeng Zhang
Summary: This paper proposes a novel low-cost robotic grasping framework, capable of achieving high degree of freedom grasping in a wild environment. By utilizing global localization and 3D point cloud reconstruction, the framework enables rough and fine localization of grasping targets, ultimately realizing visual 6-DoF robotic grasping.
Article
Computer Science, Artificial Intelligence
Martin Fisch, Ronald Clark
Summary: This paper introduces a novel approach for estimating the full position and rotation of skeletal joints using only single-frame RGB images. The method utilizes virtual markers to generate sufficient information for accurately inferring rotations. Experimental results demonstrate significant improvements in joint angles and joint positions estimation.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Engineering, Electrical & Electronic
Yuhai Wei, Hui Zhang, Hang Zhong, Li Liu, Yiming Jiang, Yaonan Wang
Summary: This paper proposes an efficient Double Simultaneous Majorization Particle Filter algorithm for localization and mapping of a mobile robot. By using pose majorization and weight majorization algorithms, the accuracy of robot localization and particle-carried map is improved. The proposed adaptive hierarchical resampling method maintains particles with higher weights.
IEEE SENSORS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Zhiqiang Miao, Hang Zhong, Jie Lin, Yaonan Wang, Rafael Fierro
Summary: This paper addresses the problem of formation control of a swarm of quadrotor UAVs with a group leader using only position and attitude measurements. A distributed formation controller is proposed, which achieves formation and attitude control through position and attitude controllers respectively, and overcomes the lack of linear and angular velocity measurements with auxiliary dynamic systems.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
(2022)
Article
Automation & Control Systems
Ningbin Lai, Yanjie Chen, Jiacheng Liang, Bingwei He, Hang Zhong, Yaonan Wang
Summary: In this paper, an onboard-eye-to-hand visual servo and task coordination control strategy is proposed for the aerial manipulator to enhance the accurate manipulation and flight stability. By establishing an error equation and adopting a multi-task coordinated control scheme, the aerial manipulator achieves precise positioning and grasping, and the stability of the closed-loop system is analyzed using the Lyapunov method.
Article
Automation & Control Systems
Zhiqiang Miao, Hang Zhong, Yaonan Wang, Hui Zhang, Haoran Tan, Rafael Fierro
Summary: This article proposes a solution for formation control of mobile robots based on image, using a monocular camera and complying with field-of-view constraints. A low-complexity image-based visual servo controller is introduced to achieve desired relative position on the image plane and solve FOV constraints. The effectiveness and performance of the proposed controller are verified through simulations and experiments.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Automation & Control Systems
Jie Lin, Yaonan Wang, Zhiqiang Miao, Hang Zhong, Rafael Fierro
Summary: This article addresses the vision-based landing problem of a low-cost quadrotor on an unknown moving platform. A robust landing controller is developed, which consists of the design of the low-complexity outer-loop controller and the geometric inner-loop attitude controller. The proposed control strategy exhibits strong robustness and has been demonstrated to be effective through numerical simulations and experiments.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Engineering, Civil
Hui Zhang, Yanan Song, Yurong Chen, Hang Zhong, Li Liu, Yaonan Wang, Thangarajah Akilan, Q. M. Jonathan Wu
Summary: This study proposes a multi-model rail surface defect detection system based on convolutional neural networks, which can rapidly and accurately identify various defects on rail surfaces and improve the detection performance.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Hui Zhang, Xidong Zhou, Hang Zhong, He Xie, Wei He, Xuan Tan, Yaonan Wang
Summary: This article introduces a low-cost UWB-odometer fusion method that enables global localization using only one UWB anchor, improving positioning accuracy and eliminating the influence of cumulative odometer error.
IEEE SENSORS JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Junfei Yi, Hui Zhang, Jianxu Mao, Yurong Chen, Hang Zhong, Yaonan Wane
Summary: This study proposes an end-to-end deep learning method with adaptive convolution and multiscale attention to quickly detect foreign particles in liquid pharmaceuticals. By using pixel-adaptive feature extraction and multiscale attention-based feature fusion, the method achieves a low missed detection rate and has a fast processing speed on a liquid pharmaceutical dataset.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Automation & Control Systems
Jingmou Nie, Yaonan Wang, Yang Mo, Zhiqiang Miao, Yiming Jiang, Hang Zhong, Jie Lin
Summary: This article proposes a control scheme based on hierarchical quadratic programming-prescribed performance control (HQP-PPC) to address the problem of redundant wheeled mobile manipulators satisfying multiple physical constraints and avoiding obstacles while performing tasks. The first layer of HQP achieves trajectory tracking, obstacle avoidance, and physical constraints, while the second layer guides obstacle avoidance direction and produces a smoother planned path. The prescribed performance function (PPF) ensures tracking performance, satisfying the physical constraints and obstacle avoidance constraints in actual motion. The proposed method is verified to be effective and superior through simulations and experiments.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Computer Science, Artificial Intelligence
Hang Zhong, Yaonan Wang, Zhiqiang Miao, Ling Li, Shuangwen Fan, Hui Zhang
Summary: This paper proposes a homography-based visual servo control approach for stabilizing underactuated unmanned aerial vehicles in GPS-denied environments. A virtual homography matrix is introduced, and a new visual error vector is defined to obtain the passivity-like visual error equation. The control approach includes an online translation velocity estimator and a nonlinear backstepping visual servo control method to achieve fast maneuvering without linear velocity feedback. It guarantees global stability in large offset, in contrast to conventional solutions that only ensure local stability. The robustness and performance of the approach are demonstrated through simulations and experiments with a realistic quadrotor, including an autonomous tracking experiment with unpredictable target speed.
IEEE TRANSACTIONS ON INTELLIGENT VEHICLES
(2023)
Article
Automation & Control Systems
Yanjie Chen, Yangning Wu, Zhenguo Zhang, Zhiqiang Miao, Hang Zhong, Hui Zhang, Yaonan Wang
Summary: This article proposes an image-based visual servoing (IBVS) control strategy for an unmanned aerial manipulator (UAM) system to track and grasp a moving target. A robust-adaptive velocity observer is designed to estimate the relative velocity between the tracked target and the UAM platform, and an IBVS controller using the onboard camera is proposed for moving target tracking. The stability of the proposed IBVS control strategy is analyzed using Lyapunov theory, and comparative simulations are provided to demonstrate the target tracking performance.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Automation & Control Systems
Jiacheng Liang, Hang Zhong, Yaonan Wang, Yanjie Chen, Junhao Zeng, Jianxu Mao
Summary: This article investigates an adaptive force tracking impedance control strategy for an aerial manipulator in uncertain contact environments. An adaptive impedance control method is proposed to achieve aerial interaction and maintain a stable contact force. A robust pose tracking controller is designed to ensure tracking performance, which includes a barrier function-based position controller and an adaptive attitude controller. The stability and convergence of the proposed strategy are analyzed mathematically, and simulations and experiments are conducted to validate the feasibility and performance.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Automation & Control Systems
Jiacheng Liang, Yaonan Wang, Hang Zhong, Yanjie Chen, Hongwen Li, Jianxu Mao, Wei Wang
Summary: This article investigates a robust variable impedance control methodology for aerial manipulators to realize compliant and safe interaction tasks. The proposed method includes a simplified impedance controller and an improved attitude control approach, which result in low implementation costs and global flight attitude stability without singularities or ambiguities.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)