Shenyang Institute of Automation,Chinese Academy Of Sciences
Not a member yet
23582 research outputs found
Sort by
Efficient Attention Pyramid Network for Semantic Segmentation
No full textSemantic segmentation is a task that covers most of the perception needs of intelligent vehicles in an unified way. Recent studies witnessed that attention mechanisms achieve impressive performance in computer vision task. Current attention mechanisms based segmentation methods differ with each other in position and form of the attention mechanism, and perform differently in practice. This paper firstly introduces the effectiveness of multi-scale context features and attention mechanisms in segmentation tasks. We find that multi-scale and channel attention can play a vital role in constructing effective context features. Based on this analysis, this paper proposes an efficient attention pyramid network (EAPNet) for semantic segmentation. Specifically, to efficient handle the problem of segmenting objects at multiple scales, we design efficient channel attention pyramid (ECAP) which employ atrous convolution with channel attention in cascade or in parallel to capture multi-scale context by using multiple atrous rates. Furthermore, we propose a residual attention fusion block (RAFB), whose purpose is to simultaneously focus on meaningful low-level feature maps and spatial location information. At the same time, we will explore different channel attention modules and spatial attention modules, and describe their impact on network performance. We empirically evaluate our EAPNet on two semantic segmentation datasets, including PASCAL VOC 2012 and Cityscapes datasets. Experimental results show that without MS COCO pre-training and any post-processing, EAPNet achieved 81.7% mIoU on the PASCAL VOC 2012 validation set. With deeplabv3+ as the benchmark, EAPNet improve the model performance of more than 1.50% mIoU
Deep Learning for EMG-based Human-Machine Interaction: A Review
No full textElectromyography (EMG) has already been broadly used in human-machine interaction (HMI) applications. Determining how to decode the information inside EMG signals robustly and accurately is a key problem for which we urgently need a solution. Recently, many EMG pattern recognition tasks have been addressed using deep learning methods. In this paper, we analyze recent papers and present a literature review describing the role that deep learning plays in EMG-based HMI. An overview of typical network structures and processing schemes will be provided. Recent progress in typical tasks such as movement classification, joint angle prediction, and force/torque estimation will be introduced. New issues, including multimodal sensing, inter-subject/inter-session, and robustness toward disturbances will be discussed. We attempt to provide a comprehensive analysis of current research by discussing the advantages, challenges, and opportunities brought by deep learning. We hope that deep learning can aid in eliminating factors that hinder the development of EMG-based HMI systems. Furthermore, possible future directions will be presented to pave the way for future research
基于USV 与AUV 异构平台协同海洋探测系统研究综述
No full textWith sufficient literature retrieval and summery, this paper reviews the heterogeneous oceanographic exploration system based on the synergy of unmanned surface vehicle (USV) and multiple autonomous underwater vehicles (AUVs). Firstly, latest progress and achievements of heterogeneous oceanic exploration system are summarized and compared in terms of technical features and design ideas. In the following, research status of key technologies in coordinated control of the heterogeneous systems are elaborated. The task assignment, path planning and formation control are discussed in detail. Then the paper analyzes the technical problems of the exploration system from the aspects of external constraints from environment and hardware and the auxiliary technology of ocean exploration system. Finally, the future vision for the oceanographic exploration system is prospected based on the analysis of the current progress and actual requirements.</p
Confining single-atom Pd on g-C3N4 with carbon vacancies towards enhanced photocatalytic NO conversion
No full textModification of a photocatalyst with single-atom noble metals can improve its activity while a remaining challenge is the stabilization of single atoms. As a proof of concept, g-C3N4 with desired amount of carbon defects was fabricated to manipulate the distribution of single-atom Pd by taking advantage of its affinity with carbon vacancy-resulted nitrogen atoms. The single-atom Pd was produced by photo-reduction, preferentially located on the carbon vacancy sites as supported by HAADF-STEM and XAFS analyses. As obtained photocatalyst showed high and stable photocatalytic activity in NO conversion; its activity is about 4.4 times higher than that of the pristine g-C3N4. The improved photoactivity was attributed to the preferential separation and transportation of the photo-generated charge carriers due to the introduction of single-atom Pd as evidenced by UV-vis, static and time-resolved photoluminescence spectroscopic analyses. The present work underlines the impetus of surface defect chemistry in the fabrication of single-atom catalysts
一种多AUV仿真平台
No full text本发明涉及一种多AUV仿真平台，该平台包括：虚拟环境系统，多AUV机载系统，视景系统，交互系统，动力学仿真系统；虚拟环境系统负责模拟海流、水声信道等海洋环境，生成AUV的传感器数据；AUV机载控制系统，可连接物理设备，主要负责接收由虚拟环境系统或真实传感器产生的传感器数据，进行数据处理、控制和多AUV间的协同；视景显示系统负责实时显示二维/三维场景，以及各AUV的运行参数和轨迹；交互系统负责为操作人员提供管理本系统的接口；动力学仿真系统进行AUV动力学仿真。本发明具有较强的系统性、展示性和可扩展性，能够较真实的验证多AUV系统的协同策略或单台AUV的控制策略，提高AUV任务方案的测试效率
Macro-micro quantitative feeding device for lining high-viscosity trace components
No full text本发明涉及一种衬层高粘度微量组分的宏微定量给料装置，包括高精度天平及设置于高精度天平上的给料容器、管道输送系统、控制系统、配重块及载体座，其中管道输送系统、控制系统及配重块设置于载体座的四周，给料容器设置于载体座的中间，给料容器、管道输送系统、控制系统、配重块及载体座全部置于高精度天平上用于称重；管道输送系统用于把从给料容器里抽出的高粘度液体输送到相应的位置；控制系统用于根据状态需求调节阀门和泵体的工作状态；配重块用于平衡整个载体座上的重量保证质心位置。本发明可以保证高粘度微量组分的高精度输送，具有广泛的使用范围，并且具有较小的尺寸
一种面向工业互联制造的移动开放平台及实现方法
No full text本发明提出了一种面向工业互联制造的移动应用开放平台及实现方法，该开放平台面向工业互联制造领域，提供SaaS模式的云应用服务，将应用软件统一部署在平台的服务器上，客户可以根据工作实际需求，通过互联网定购所需的移动应用以及算法，用户购买后部署到独立的租户空间内，为企业提供便捷的，高效的应用管理服务。此外，平台还提供基础工业建模，开放数据查询，移动终端设备同移动应用的双向通信交互以及信息获取，消息推送服务等各类接口，用户可以根据自身需求自行开发应用，也满足了其他应用供应商的开发需要
一种安全型堆垛机用多级货叉
No full text本发明涉及堆垛机货叉系统，特别涉及一种安全型堆垛机用多级货叉。包括载货台基座及设置于载货台基座上的伸缩货叉单元、驱动单元及货叉末端检测单元，其中驱动单元与伸缩货叉单元连接，用于驱动伸缩货叉单元进行伸缩；末端检测单元与伸缩货叉单元连接，用于检测伸缩货叉单元的伸缩位移。伸缩货叉单元包括依次滑动连接的下层货叉、中层货叉及上层货叉，其中下层货叉与载货台基座连接，上层货叉与货叉末端检测单元连接。本发明具有高度安全可靠性，可避免在托盘对象位置异常或传动机构意外机械疲劳损伤等工况下设备运行的安全性
Detection and Segmentation of Structured Light Stripe in Weld Image
No full text为了在复杂噪声环境下从焊缝图像中精确地提取结构光条纹,构建了语义分割与目标检测相结合的深度学习模型用于焊缝图像的检测。为了提高模型的检测速度,在语义分割分支中,通过添加并行下采样模块及缩减卷积核数量的策略对模型进行了优化,并使该分支与目标检测分支的特征提取部分共享权重。针对焊缝图像中结构光条纹与背景像素比例失衡而导致模型分割结果偏向负样本的问题,在损失函数中添加Dice系数来对模型进行修正。经实验验证,该方法在保证实时性的基础上,以较高的精度实现了结构光条纹的检测。</p