检索结果-内蒙古大学图书馆

Chinese Control Conference (CCC)

作者： Lu Cheng Yin-Bo Liu Qing-Hao Meng Institute of Robotics and Autonomous Systems Tianjin Key Laboratory of Process Measurement and Control School of Electrical and Information Engineering Tianjin University Tianjin China

ISBN: (数字)9789881563903

ISBN: (纸本)9781728165233

To detect parts per billion (ppb)-level volatile organic compounds (VOCs) in automobiles, a pre-concentrator is generally needed for electronic noses (e-noses), which can enrich ppb-level concentration to parts per million (ppm) level. However, the current e-nose chamber structure could dilute the pre-concentrated gases and then the detection performance is reduced. To solve this problem, a novel chamber design was proposed in this paper. The core design idea was to reduce the space for the gas flow in the chamber, taking into account the shape and size of the gas sensors. Specially, for the pin-type gas sensors, a series of cylindrical structure design was adopted, and for the micro-electro-mechanical system (MEMS) patch-type sensors, the headspace was compressed. The results of computational fluid dynamics (CFD) based simulation showed that the concentration of pre-concentrated gases in the novel chamber was higher than that of the traditional chamber. The designed chambers were manufactured using polylactic acid (PLA) material by 3D printing. Experimental gases were prepared from polyurethane (PU) leather and polyvinyl chloride (PVC) leather to verify the performance of the chambers, and the results demonstrated that the proposed new e-nose chamber was superior to the traditional one in terms of the pre-concentration effect, repeatability and classification performance.

关键词： Gas detectors Sensor arrays Micromechanical devices Fluid flow Gases Automobiles

来源：评论

学校读者我要写书评

暂无评论

Indoor Layout Estimation by Fusing Monocular RGB Image Features Extracted with HRNet

Indoor Layout Estimation by Fusing Monocular RGB Image Featu...

引用

Chinese Control Conference (CCC)

作者： Rong-Ze Huang Yin-Bo Liu Meh Jabeen Qing-Hao Meng Institute of Robotics and Autonomous Systems Tianjin Key Laboratory of Process Measurement and Control School of Electrical and Information Engineering Tianjin University Tianjin China

ISBN: (数字)9789881563903

ISBN: (纸本)9781728165233

Indoor layout estimation (ILE) is a challenging research that segments the natural indoor scene image into floor, walls and ceiling, and it has a wide range of application prospects in the fields such as scene understanding, reconstruction, robot positioning, and virtual reality. An ILE method by merging multiple features from a monocular red-green-blue (RGB) image was proposed. Firstly, the High-resolution networks (HRNet) was used to extract the features of indoor corner keypoints (CKs) and informative edges (IEs) in the form of heatmaps. Secondly, the extracted features were then selected through a room type (RT) classification network. Finally, a feature fusion algorithm was proposed, which obtained the initial layout hypothesis through CK features and then continuously optimized the layout hypothesis with IE features. The proposed ILE method was validated on the large scene understanding challenge (LSUN) dataset.

关键词：

来源：评论

学校读者我要写书评

暂无评论

D-VPnet: A Network for Real-time Dominant Vanishing Point Detection in Natural Scenes

arXiv

引用

arXiv 2020年

作者： Bo Liu, Yin Ming, Zeng Meng, Qing-Hao Institute of Robotics and Autonomous Systems Tianjin Key Laboratory of Process Measurement and Control School of Electrical and Information Engineering Tianjin University Tianjin300072 China

As an important part of linear perspective, vanishing points (VPs) provide useful clues for mapping objects from 2D photos to 3D space. Existing methods are mainly focused on extracting structural features such as lines or contours and then clustering these features to detect VPs. However, these techniques suffer from ambiguous information due to the large number of line segments and contours detected in outdoor environments. In this paper, we present a new convolutional neural network (CNN) to detect dominant VPs in natural scenes, i.e., the Dominant Vanishing Point detection Network (D-VPnet). The key component of our method is the feature line-segment proposal unit (FLPU), which can be directly utilized to predict the location of the dominant VP. Moreover, the model also uses the two main parallel lines as an assistant to determine the position of the dominant VP. The proposed method was tested using a public dataset and a Parallel Line based Vanishing Point (PLVP) dataset. The experimental results suggest that the detection accuracy of our approach outperforms those of state-of-the-art methods under various conditions in real-time, achieving rates of 115fps. Copyright © 2020, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Unstructured Road Vanishing Point Detection Using the Convolutional Neural Network and Heatmap Regression

arXiv

引用

arXiv 2020年

作者： Liu, Yin-Bo Zeng, Ming Meng, Qing-Hao Institute of Robotics and Autonomous Systems Tianjin Key Laboratory of Process Measurement and Control School of Electrical and Information Engineering Tianjin University Tianjin300072 China

Unstructured road vanishing point (VP) detection is a challenging problem, especially in the field of autonomous driving. In this paper, we proposed a novel solution combining the convolutional neural network (CNN) and heatmap regression to detect unstructured road VP. The proposed algorithm firstly adopts a lightweight backbone, i.e., depthwise convolution modified HRNet, to extract hierarchical features of the unstructured road image. Then, three advanced strategies, i.e., multi-scale supervised learning, heatmap super-resolution, and coordinate regression techniques are utilized to achieve fast and high-precision unstructured road VP detection. The empirical results on Kong’s dataset show that our proposed approach enjoys the highest detection accuracy compared with state-of-the-art methods under various conditions in real-time, achieving the highest speed of 33 fps. Copyright © 2020, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Bayesian deep multi-instance learning for student performance prediction based on campus big data

引用

Neurocomputing 2025年 647卷

作者： Jiayang Huang Keyi Yang Quan Wang Pengfei Yang Ziling Ruan Jiaxi Wang Zhi-Qiang Zhang School of Computer Science and Technology Xidian University Xi’an 710126 China The Key Laboratory of Smart Human-Computer Interaction and Wearable Technology of Shaanxi Province Xi’an 710126 China School of Electronic and Electrical Engineering Institute of Robotics Autonomous Systems and Sensing University of Leeds Leeds UK

Problem: Predicting student performance using campus big data and deep learning methods has emerged as a promising alternative to traditional psychological assessments, which are often delayed and subjective. However, existing deep learning approaches face significant challenges, including the sparsity of campus big data and susceptibility to overfitting. Objective: To address these issues, this study aims to develop a robust and effective framework for predicting student performance by overcoming data sparsity and overfitting problems. Method: We propose a novel Bayesian Deep Multi-Instance Learning (Bay-DeepMIL) framework. The method integrates multi-instance learning (MIL) to handle data sparsity and incorporates a Bayesian framework to treat network parameters as probabilistic distributions, enhancing robustness and mitigating overfitting. Additionally, a Bayesian multi-head attention mechanism is introduced to dynamically assign importance to different instances, improving the extraction of key information from multi-instance data. Results: Extensive experiments demonstrate that Bay-DeepMIL outperforms state-of-the-art methods in prediction accuracy. The framework also provides uncertainty estimates for each prediction, offering valuable confidence measures and decision support for educational stakeholders. Conclusion: The proposed Bay-DeepMIL framework not only advances the technical capabilities of predictive models but also provides practical tools for enhancing educational decision-making. This study underscores the effectiveness of integrating Bayesian inference with multi-instance learning to address core challenges such as data sparsity and model overfitting in student performance prediction.

关键词： Academic performance Bayesian neural network Deep learning Educational data mining Multi-instance learning

来源：评论

学校读者我要写书评

暂无评论

Human Intention Recognition for Human Aware Planning in Integrated Warehouse systems

Human Intention Recognition for Human Aware Planning in Inte...

引用

Mediterranean Conference on Control and Automation (MED)

作者： Tomislav Petković Jakub Hvězda Tomáš Rybecký Ivan Marković Miroslav Kulich Libor Přeučil Ivan Petrović Faculty of Electrical Engineering and Computing University of Zagreb Laboratory for Autonomous Systems and Mobile Robotics Unska 3 Zagreb Croatia Czech Institute of Informatics Robotics and Cybernetics Czech Technical University in Prague Prague Czech Republic

ISBN: (数字)9781728157429

ISBN: (纸本)9781728157436

With the substantial growth of logistics businesses the need for larger and more automated warehouses increases, thus giving rise to fully robotized shop-floors with mobile robots in charge of transporting and distributing goods. However, even in fully automatized warehouse systems the need for human intervention frequently arises, whether because of maintenance or because of fulfilling specific orders, thus bringing mobile robots and humans ever closer in an integrated warehouse environment. In order to ensure smooth and efficient operation of such a warehouse, paths of both robots and humans need to be carefully planned; however, due to the possibility of humans deviating from the assigned path, this becomes an even more challenging task. Given that, the supervising system should be able to recognize human intentions and its alternative paths in real-time. In this paper, we propose a framework for human deviation detection and intention recognition which outputs the most probable paths of the humans workers and the planner that acts accordingly by replanning for robots to move out of the human's path. Experimental results demonstrate that the proposed framework increases total number of deliveries, especially human deliveries, and reduces human-robot encounters.

关键词： Robot kinematics Frequency modulation Planning Task analysis Hidden Markov models Trajectory

来源：评论

学校读者我要写书评

暂无评论

Multi-agent Gaussian Process Motion Planning via Probabilistic Inference

引用

IFAC-PapersOnLine 2018年第22期51卷 160-165页

作者： Petrovi, Luka Markovi, Ivan Seder, Marija University of Zagreb Faculty of Electrical Engineering and Computing Laboratory for Autonomous Systems and Mobile Robotics Croatia

This paper deals with motion planning for multiple agents by representing the problem as a simultaneous optimization of every agent's trajectory. Each trajectory is considered as a sample from a one-dimensional continuous-time Gaussian process (GP) generated by a linear time-varying stochastic differential equation driven by white noise. By formulating the planning problem as probabilistic inference on a factor graph, the structure of the pertaining GP can be exploited to find the solution efficiently using numerical optimization. In contrast to planning each agent's trajectory individually, where only the current poses of other agents are taken into account, we propose simultaneous planning of multiple trajectories that works in a predictive manner. It takes into account the information about each agent's whereabouts at every future time instant, since full trajectories of each agent are found jointly during a single optimization procedure. We compare the proposed method to an individual trajectory planning approach, demonstrating significant improvement in both success rate and computational efficiency. © 2018

关键词： Motion planning Computational efficiency Continuous time systems Differential equations Gaussian distribution Gaussian noise (electronic) Multi agent systems One dimensional Optimization Stochastic systems Trajectories White noise Factor graphs Numerical optimizations Optimization procedures Probabilistic inference Simultaneous optimization Stochastic differential equations Trajectory optimization Trajectory Planning

来源：评论

学校读者我要写书评

暂无评论

Low-cost Scalable People Tracking System for Human-Robot Collaboration in Industrial Environment

引用

Procedia Manufacturing 2020年 51卷 116-124页

作者： Matteo Terreran Edoardo Lamon Stefano Michieletto Enrico Pagello Intelligent Autonomous Systems Laboratory (IAS-Lab) Dept. of Information Engineering (DEI) University of Padova Padova Italy Robotics and Automation Dept. of Information Engineering Universitä degli Studi di Pisa Pisa Italy

Human-robot collaboration is one of the key elements in the Industry 4.0 revolution, aiming to a close and direct collaboration between robots and human workers to reach higher productivity and improved ergonomics. The first step toward such kind of collaboration in the industrial context is the removal of physical safety barriers usually surrounding standard robotic cells, so that human workers can approach and directly collaborate with robots. Anyway, human safety must be granted avoiding possible collisions with the robot. In this work, we propose the use of a people tracking algorithm to monitor people moving around a robot manipulator and recognize when a person is too close to the robot while performing a task. The system is implemented by a camera network system positioned around the robot workspace, and thoroughly evaluated in different industry-like settings in terms of both tracking accuracy and detection delay.

关键词： People tracking human-robot collaboration low-cost industrial

来源：评论

学校读者我要写书评

暂无评论

Corrections to “a Sim-to-Real Deep Learning-Based Framework for autonomous Nano-Drone Racing”

引用

IEEE robotics and Automation Letters 2024年第10期9卷 8426-8426页

作者： Lorenzo Lamberti Elia Cereda Gabriele Abbate Lorenzo Bellone Victor Javier Kartsch Morinigo Michał Barciś Agata Barciś Alessandro Giusti Francesco Conti Daniele Palossi Department of Electrical Electronic and Information Engineering University of Bologna Bologna Italy Dalle Molle Institute for Artificial Intelligence USI-SUPSI Lugano Switzerland Autonomous Robotics Research Center Technology Innovation Institute Abu Dhabi UAE Integrated Systems Laboratory ETH Zürich Zürich Switzerland

Presents corrections to the article “a Sim-to-Real Deep Learning-Based Framework for autonomous Nano-Drone Racing”.

关键词： Robots Technological innovation IEEE Associate Members

来源：评论

学校读者我要写书评

暂无评论

Computationally efficient dense moving object detection based on reduced space disparity estimation

引用

IFAC-PapersOnLine 2018年第22期51卷 360-365页

作者： Popovi, Goran Hadviger, Antea Markovi, Ivan Petrovi, Ivan University of Zagreb Faculty of Electrical Engineering and Computing Laboratory for Autonomous Systems and Mobile Robotics Zagreb Croatia

Computationally efficient moving object detection and depth estimation from a stereo camera is an extremely useful tool for many computer vision applications, including robotics and autonomous driving. In this paper we show how moving objects can be densely detected by estimating disparity using an algorithm that improves complexity and accuracy of stereo matching by relying on information from previous frames. The main idea behind this approach is that by using the ego-motion estimation and the disparity map of the previous frame, we can set a prior base that enables us to reduce the complexity of the current frame disparity estimation, subsequently also detecting moving objects in the scene. For each pixel we run a Kalman filter that recursively fuses the disparity prediction and reduced space semi-global matching (SGM) measurements. The proposed algorithm has been implemented and optimized using streaming single instruction multiple data instruction set and multi-threading. Furthermore, in order to estimate the process and measurement noise as reliably as possible, we conduct extensive experiments on the KITTI suite using the ground truth obtained by the 3D laser range sensor. Concerning disparity estimation, compared to the OpenCV SGM implementation, the proposed method yields improvement on the KITTI dataset sequences in terms of both speed and accuracy. © 2018

关键词： Computer vision Computational complexity Computational efficiency Motion estimation Object detection Object recognition Stereo image processing Stereo vision 3D laser range sensors Computationally efficient Computer vision applications Dense disparity estimation Detecting moving objects Moving object detection Semi global matching Single instruction multiple data instructions

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：