Article
Robotics
Sheng Yu, Di-Hua Zhai, Yuanqing Xia, Haoran Wu, Jun Liao
Summary: In this letter, a novel grasp detection neural network called Squeeze-and-Excitation ResUNet (SE-ResUNet) is developed. It can generate grasp poses from RGB-D images and predict the quality scores of each grasp pose. The experimental results show high accuracy and real-time performance, and the proposed method outperforms other methods in comparison study.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Article
Multidisciplinary Sciences
Di Lu, Shuli Cheng, Liejun Wang, Shiji Song
Summary: In this paper, a new deep learning-based method for change detection is proposed, which utilizes multi-scale feature fusion and distribution strategies to improve the accuracy of change region detection. Experimental results demonstrate that this method outperforms other comparative methods.
SCIENTIFIC REPORTS
(2022)
Article
Automation & Control Systems
Sheng Yu, Di-Hua Zhai, Yuanqing Xia
Summary: This paper proposes a new robotic grasp detection network called SKGNet, which achieves high accuracy and real-time performance by integrating attention mechanism and multi-scale fusion features. Training and testing on public datasets, SKGNet achieves superior accuracy and high detection speed compared to existing methods. Real-world experiments demonstrate the generalization performance and effectiveness of SKGNet.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2022)
Article
Engineering, Multidisciplinary
Xinheng Yuan, Hao Yu, Houlin Zhang, Li Zheng, Erbao Dong, Heng'an Wu
Summary: This paper proposes a grasp detector for predicting multi-scale grasp rectangles on different sizes of objects and various cameras and grippers. The detector achieves high accuracy by conducting independent predictions on different scales and using a fully matching model and background classifier to ensure efficiency. Experimental results demonstrate that the detector achieves high accuracy on different datasets and can predict grasps on multiple objects.
CMES-COMPUTER MODELING IN ENGINEERING & SCIENCES
(2022)
Article
Automation & Control Systems
Dexin Wang, Chunsheng Liu, Faliang Chang, Nanjun Li, Guangxin Li
Summary: This article presents a pixel-level grasp detection method based on deep neural network, which achieves high accuracy and fast speed on RGB images through novel representation models, adaptive attribute models, and feature fusion and grasp-aware network.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Automation & Control Systems
Hu Cao, Guang Chen, Zhijun Li, Qian Feng, Jianjie Lin, Alois Knoll
Summary: This study proposes an efficient grasp detection network for robotic grasp tasks. The network uses a lightweight generative structure to achieve a balance between high grasp confidence and fast inference speed. It introduces a Gaussian kernel-based grasp representation for encoding training samples and employs receptive field blocks and attention mechanisms for improved feature discriminability and semantic information fusion. Experimental results demonstrate excellent performance on Cornell, Jacquard, and extended OCID grasp datasets with accuracy of 97.8%, 95.6%, and 76.4% respectively.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2023)
Article
Chemistry, Multidisciplinary
Amna Mujahid, Muhammad Aslam, Muhammad Usman Ghani Khan, Ana Maria Martinez-Enriquez, Nazeef Ul Haq
Summary: The main contribution of this research is the collected dataset of human gestures, which is of high-quality and low noise. The study focuses on the application of confidence determination during social issues and achieves remarkable results by combining CNN and LSTM models for training and testing.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Zhihua Xu, Shangwei Lan, Zhijing Yang, Jiangzhong Cao, Zongze Wu, Yongqiang Cheng
Summary: A multi-stage balanced R-CNN (MSB R-CNN) for defect detection is proposed in this paper based on Cascade R-CNN, which uses deformable convolution, balanced feature pyramid, and balanced loss to improve the accuracy and convergence effect of the model.
Review
Computer Science, Artificial Intelligence
Ning Cao, Shujuan Ji, Dickson K. W. Chiu, Maoguo Gong
Summary: This study investigates the detection of deceptive reviews in online platforms and proposes a feature fusion strategy that combines feature representations from different models. Experimental results demonstrate the superior performance of this method compared to others.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Sukhendra Singh, Manoj Kumar, Abhay Kumar, Birendra Kumar Verma, S. Shitharth
Summary: Worldwide, pneumonia is a leading cause of infant mortality. Chest X-rays are used by experienced radiologists to diagnose pneumonia and other respiratory diseases, but the complexity of the diagnostic procedure often leads to disagreement among radiologists. Early diagnosis is crucial for mitigating the impact of the disease, and computer-aided diagnostics, such as Quaternion neural networks, improve the accuracy of diagnosis. Our proposed QCSA network combines spatial and channel attention mechanisms with Quaternion residual network to classify chest X-Ray images for Pneumonia detection, achieving a high accuracy rate of 94.53% and an AUC of 0.89. This approach shows promising potential for detecting pneumonia.
SCIENTIFIC REPORTS
(2023)
Article
Engineering, Civil
Muneeb Ahmed, Sarfaraz Masood, Musheer Ahmad, Ahmed A. Abd El-Latif
Summary: Fatigue, drowsiness, and distraction are the main causes of road accidents worldwide. Existing solutions either extract physiological signals of the driver or use computer vision techniques, but they have limited performances. Therefore, this study proposes an ensemble deep learning architecture to determine the state of the driver by incorporating features from the eyes and mouth. The model achieves high accuracy when trained and evaluated on the NTHU-DDD video dataset.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Environmental Sciences
Chujie Xu, Xiangtao Zheng, Xiaoqiang Lu
Summary: This paper explores a cross-domain ship detection task, adapting the detector from labeled optical images to unlabeled SAR images. A multi-level alignment network is proposed to achieve cross-domain detection and reduce domain shift. Experimental results demonstrate the effectiveness of the method.
Article
Biology
Amir Ebrahimi, Suhuai Luo, Raymond Chiong
Summary: The study examined the effectiveness of applying deep sequence-based network models for AD detection, addressing the classification accuracy issue of 2D and 3D CNNs in AD detection by handling the MRI feature sequences generated by CNNs.
COMPUTERS IN BIOLOGY AND MEDICINE
(2021)
Article
Mathematics
Xiaobao Yang, Wentao Wang, Junsheng Wu, Chen Ding, Sugang Ma, Zhiqiang Hou
Summary: This paper proposes a feature pyramid network method called MLA-Net, which uses a local attention mechanism to establish the correlation between multi-level features and improve object detection performance. Extensive experiments show that the proposed method achieves significant improvements on multiple datasets.
Article
Mathematics
Xiaochen Ju, Xinxin Zhao, Shengsheng Qian
Summary: This paper proposes a novel method called Transformer-based Multi-scale Fusion Model (TransMF) for crack detection. It combines an Encoder Module, Decoder Module, and Fusion Module to capture long-range dependencies, eliminate background noise, and enhance the detection of cracks. Experimental results show that TransMF outperforms existing baselines.
Article
Engineering, Aerospace
Lu Chen, Panfeng Huang, Jia Cai, Zhongjie Meng, Zhengxiong Liu
AEROSPACE SCIENCE AND TECHNOLOGY
(2016)
Article
Engineering, Aerospace
Panfeng Huang, Lu Chen, Bin Zhang, Zhongjie Meng, Zhengxiong Liu
INTERNATIONAL JOURNAL OF AEROSPACE ENGINEERING
(2017)
Article
Engineering, Aerospace
Fan Zhang, Panfeng Huang, Lu Chen, Jia Cai
JOURNAL OF AEROSPACE ENGINEERING
(2018)
Review
Engineering, Mechanical
Panfeng Huang, Fan Zhang, Lu Chen, Zhongjie Meng, Yizhai Zhang, Zhengxiong Liu, Yongxin Hu
NONLINEAR DYNAMICS
(2018)
Article
Automation & Control Systems
Lu Chen, Panfeng Huang, Zhou Zhao
ROBOTICS AND AUTONOMOUS SYSTEMS
(2018)
Article
Automation & Control Systems
Yuanhao Li, Panfeng Huang, Zhiqiang Ma, Lu Chen
Summary: The vision-based grasp detection method is important in researching the grasping task of robots. However, these research methods perform worse in practice compared to state-of-the-art accuracy on public datasets, mainly due to shifts in data distribution and the sensitivity of neural network-based methods to small data changes. The evaluation metrics of existing models do not reflect the actual robustness of the method. Therefore, a new solution is proposed, which includes a benchmark for verifying realistic robustness and a method for improving model robustness by transferring texture knowledge from other images.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhou Zhao, Panfeng Huang, Lu Chen
2017 IEEE INTERNATIONAL CONFERENCE ON REAL-TIME COMPUTING AND ROBOTICS (RCAR)
(2017)
Proceedings Paper
Computer Science, Artificial Intelligence
Lu Chen, Panfeng Huang, Zhou Zhao
2017 IEEE INTERNATIONAL CONFERENCE ON REAL-TIME COMPUTING AND ROBOTICS (RCAR)
(2017)
Proceedings Paper
Robotics
Jia Cai, Panfeng Huang, Lu Chen, Bin Zhang
2015 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (ROBIO)
(2015)
Proceedings Paper
Robotics
Lu Chen, Panfeng Huang, Jia Cai
2015 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND BIOMIMETICS (ROBIO)
(2015)
Article
Automation & Control Systems
Runwei Guan, Shanliang Yao, Lulu Liu, Xiaohui Zhu, Ka Lok Man, Yong Yue, Jeremy Smith, Eng Gee, Yutao Yue
Summary: With the development of Unmanned Surface Vehicles (USVs), the perception of inland waterways has become significant. Traditional RGB cameras cannot work effectively in adverse weather and at night, which has led to the emergence of 4D millimeter-wave radar as a new perception sensor. However, the radar suffers from water-surface clutter and irregular shape of point cloud. To address these issues, this paper proposes a high-performance panoptic perception model called Mask-VRDet, which fuses features of vision and radar using graph neural network.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Adrien Le Reun, Kevin Subrin, Anthony Dubois, Sebastien Garnier
Summary: This study aims to evaluate the quality and health of aerospace parts using a high-dimensional robotic cell. By utilizing X-ray Computed Tomography devices, the interior of the parts can be reconstructed and anomalies can be detected. A methodology is proposed to assess both the raw process capability and the improved process capability, with three strategies developed to improve the robot behavior model and calibration.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Weiming Ba, Jung-Che Chang, Jing Liu, Xi Wang, Xin Dong, Dragos Axinte
Summary: This paper proposes a hybrid scheme for kinematic control of continuum robots, which avoids errors through tension supervision and accurate piecewise linear approximation. The effectiveness of the controller is verified on different continuum robotic systems.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Gabriele Abbate, Alessandro Giusti, Viktor Schmuck, Oya Celiktutan, Antonio Paolillo
Summary: In this study, a learning-based approach is proposed to predict the probability of human users interacting with a robot before the interaction begins. By considering the pose and motion of the user, the approach labels the robot's encounters with humans in a self-supervised manner. The method is validated and deployed in various scenarios, achieving high accuracy in predicting user intentions to interact with the robot.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Tiago Cortinhal, Eren Erdal Aksoy
Summary: This work presents a new depth-and semantics-aware conditional generative model, named TITAN-Next, for cross-domain image-to-image translation between LiDAR and camera sensors. The model is able to translate raw LiDAR point clouds to RGB-D camera images by solely relying on semantic scene segments, and it has practical applications in fields like autonomous vehicles.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Marios Krestenitis, Emmanuel K. Raptis, Athanasios Ch. Kapoutsis, Konstantinos Ioannidis, Elias B. Kosmatopoulos, Stefanos Vrochidis
Summary: This paper addresses the issue of informative path planning for a UAV used in precision agriculture. By using a non-uniform scanning approach, the time spent in areas with minimal value is reduced, while maintaining high precision in information-dense regions. A novel active sensing and deep learning-based coverage path planning approach is proposed, which adjusts the UAV's speed based on the quantity and confidence level of identified plant classes.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Shota Kokubu, Pablo E. Tortos Vinocour, Wenwei Yu
Summary: In this study, a new modular soft actuator was proposed to improve the support performance of soft rehabilitation gloves (SRGs). Objective evaluations and clinical tests were conducted to demonstrate the effectiveness and functionality of the proposed actuator and SRG.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Jinliang Zhu, Yuanxi Sun, Jie Xiong, Yiyang Liu, Jia Zheng, Long Bai
Summary: This paper proposes an active prosthetic knee joint with a variable stiffness parallel elastic actuation mechanism. Numerical verifications and practical experiments demonstrate that the mechanism can reduce torque and power, thus reducing energy consumption and improving the endurance of the prosthetic knee joint.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Yong You, Jingtao Wu, Yunlong Meng, Dongye Sun, Datong Qin
Summary: A new power-cycling variable transmission (PCVT) is proposed and applied to construction vehicles to improve transmission efficiency. A shift correction strategy is developed based on identifying the changes in construction vehicles' mass and gradient. Simulation results show that the proposed method can correct shift points, improve operation efficiency, and ensure a safer operation process.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Shaorui Liu, Wei Tian, Jianxin Shen, Bo Li, Pengcheng Li
Summary: This paper proposes a two-objective optimization technique for multi-robot systems, addressing the issue of balancing productivity and machining performance in high-quality machining tasks.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Pengchao Ding, Faben Zhu, Hongbiao Zhu, Gongcheng Wang, Hua Bai, Han Wang, Dongmei Wu, Zhijiang Du, Weidong Wang
Summary: We propose an autonomous approaching scheme for mobile robot traversing obstacle stairwells, which overcomes the restricted field of vision caused by obstacles. The scheme includes stair localization, structural parameter estimation, and optimization of the approaching process.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Pedro Azevedo, Vitor Santos
Summary: Accurate detection and tracking of vulnerable road users and traffic objects are vital tasks for autonomous driving and driving assistance systems. This paper proposes a solution for object detection and tracking in an autonomous driving scenario, comparing different object detectors and exploring the deployment on edge devices. The effectiveness of DeepStream technology and different object trackers is assessed using the KITTI tracking dataset.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)
Article
Automation & Control Systems
Benjamin Beiter, Divya Srinivasan, Alexander Leonessa
Summary: Powered exoskeletons can significantly reduce physical workload and have great potential impact on future labor practices. To truly assist users in achieving task goals, a shared autonomy control framework is proposed to separate the control objectives of the human and exoskeleton. Positive Power control is introduced for the human-based controller, while 'acceptance' is used as a measure of matching the exoskeleton's control objective to the human's. Both control objectives are implemented in an optimization-based Whole-Body-Control structure. The results verify the effectiveness of the control framework and its potential for improving cooperative control for powered exoskeletons.
ROBOTICS AND AUTONOMOUS SYSTEMS
(2024)