Shenyang Institute of Automation,Chinese Academy Of Sciences
Not a member yet
23582 research outputs found
Sort by
Recent advance in cell patterning techniques: Approaches, applications and future prospects
No full textCell patterning technology is an effective technique to understand cell behaviors by precisely controlling the position of cells revealing the major contributions of cells in basic and applied biological research in vitro. Therefore, this technology has led to the development of various micro-nano manufacturing methods. In recent years, various methods have been continuously improved and innovated based on existing methods to overcome shortcomings. We outline the latest developments in cell patterning technology, including stencil patterning, microfluidics-based on traps and droplets, and surface modification techniques such as photolithography, microcontact printing, and laser technology. In the end, we introduce the application and future prospects of cell patterning in biological research.(c) 2021 Elsevier B.V. All rights reserved
An Efficient Online Trajectory Generation Method Based on Kinodynamic Path Search and Trajectory Optimization for Human-Robot Interaction Safety
No full textWith the rapid development of robot perception and planning technology, robots are gradually getting rid of fixed fences and working closely with humans in shared workspaces. The safety of human-robot coexistence has become critical. Traditional motion planning methods perform poorly in dynamic environments where obstacles motion is highly uncertain. In this paper, we propose an efficient online trajectory generation method to help manipulator autonomous planning in dynamic environments. Our approach starts with an efficient kinodynamic path search algorithm that considers the links constraints and finds a safe and feasible initial trajectory with minimal control effort and time. To increase the clearance between the trajectory and obstacles and improve the smoothness, a trajectory optimization method using the B-spline convex hull property is adopted to minimize the penalty of collision cost, smoothness, and dynamical feasibility. To avoid the collisions between the links and obstacles and the collisions of the links themselves, a constraint-relaxed links collision avoidance method is developed by solving a quadratic programming problem. Compared with the existing state-of-the-art planning method for dynamic environments and advanced trajectory optimization method, our method can generate a smoother, collision-free trajectory in less time with a higher success rate. Detailed simulation comparison experiments, as well as real-world experiments, are reported to verify the effectiveness of our method. © 2022 by the authors. Licensee MDPI, Basel, Switzerland.</p
Fault detection and isolation of actuator failures in jet engines using adaptive dynamic programming
No full textThis paper presents a adaptive dynamic programming-based fault detection and isolation (FDI) scheme to detect and isolate faults in an aircraft jet engine. To this end, the weights in Actor-Critic neural networks are first tuned to learn the input-output map of the jet engine considering its multiple working modes. The convergences of the trainings in Critic-Actor neural networks are strictly proved without knowing the drift dynamics and the input dynamics in the presence of unknown nonlinearities and approximation errors. Using the residuals that are generated by measuring the difference of each network output and the measured engine output, various criteria are established for accomplishing the fault diagnosis task, that addresses the problem of fault detection and isolation of the system components. A number of simulation studies are carried out for combustion chamber of a single-spool jet engine to demonstrate and illustrate the advantages, capabilities, and performance of our proposed fault diagnosis scheme. (c) 2021 Elsevier Inc. All rights reserved
Routing failure prediction and repairing for AUV-assisted underwater acoustic sensor networks in uncertain ocean environments
No full textUnderwater acoustic sensor networks (UASNs) provide temporary links, which is of great significance when it comes to dealing with abnormal situations or emergencies in Internet of underwater things (IoUT). However, UASNs are susceptible to changes and uncertainties in network topology, channel conditions, etc., which can easily lead to frequent link interruptions. In this paper, we introduce a link failure prediction mechanism and an autonomous underwater vehicle (AUV)-assisted routing holes repairing mechanism for routing design of UASNs in uncertain ocean environments, to save system energy consumption and improve network connectivity. The proposed link failure prediction mechanism takes into account residual energy of sensor nodes, node drifting information, and uncertain ocean ambient noise. When the energy of multiple sensor nodes is exhausted, the particle swarm optimization algorithm (PSO) is adopted to calculate the optimal repair location, and an AUV is used for fixed point repairing. The proposed method can effectively reduce the energy consumption of sensor nodes, increase the packet delivery ratio, and extend the life of entire network of UASNs. (C) 2021 Elsevier Ltd. All rights reserved
Probabilistic Riemannian quantification method with log-Euclidean metric learning
No full text在许多机器学习应用中，需要分析的数据可能由对称正定矩阵构成，而经典的欧氏机器学习算法处理这种数据的性能较差。针对此问题，提出一种新的基于对数欧氏度量学习的概率黎曼空间量化方法。该方法将对称正定矩阵看做对数欧氏度量下的黎曼流形上的点，采用对数欧氏度量学习距离函数将概率学习矢量量化方法从欧氏空间推广到对称正定黎曼空间。在BCI IV 2a脑电数据集上，该方法相较于概率学习矢量量化方法识别正确率提升20%，高于竞赛第一名；并且计算速度快，模型训练及测试时间分别为基于仿射不变度量的同类型算法的1%和10%。在BCI III IIIa和图像数据集ETH-80上也取得了较好的结果。</p
Automatic fault diagnosis method based on neural architecture search for industrial processes
No full text针对现有基于深度神经网络的工业过程故障诊断方法存在网络结构设计繁琐及参数寻优耗时等问题，提出了一种基于网络结构搜索的工业过程自动故障诊断方法(Automatic Fault Diagnosis,AutoFD)。该方法采用AutoFD网络结构搜索算法，来自动完成卷积神经网络的网络结构设计和网络参数寻优，在此基础上，首先通过在原始数据上施加操作生成新通道，接着利用表现预测加速获取通道适应性排序的过程，然后依据通道适应性排序，通过表现预测来快速选取最优卷积通道数，最终根据最优卷积通道来搜索表现最优的多通道卷积神经网络模型用于工业过程自动故障诊断。采用田纳西-伊斯曼(Tennessee Eastman,TE)工业过程和数值系统对提出的方法进行验证，结果表明，该方法可以实现网络结构自动设计及网络参数的自动寻优，并且具有优良的故障诊断性能。</p
The effects of thermal stratification on airborne transport within the urban roughness sublayer
No full textThe release of airborne hazardous substances in the urban roughness sublayer (RSL) is a risk to human health. The atmospheric dispersion of these materials should be contained to mitigate their adverse consequences. This study investigates the effects of atmospheric stability on the air, heat, and pollutant transport within RSL by a validated 3-D RANS code. A building morphology with height variance is proposed for better representing the real urban environment and characterizing the complex interactions between the roughness elements and the flow regimes. The thermal stratification stability test subject to the ground-inflow temperature difference is extended to span from strongly unstable to moderately stable. The quantitative indicators, including air exchange rate, heat removal rate, pollutant removal rate, heat transfer coefficient, and pollutant transfer coefficient, are introduced and analyzed. The correlations between indicators and thermal stabilities are established, which provides explicit expressions to describe the influence of the changing bulk Richardson number (Rb). Results show that the design of height-variance elements enables a stronger lateral momentum towards the target street canyon, which suppresses the spanwise dispersion of pollutants. The discussions upon heat and pollutant transport based on stability categories show a high Rb-dependence of heat/mass transfer efficiency. The stability threshold is demonstrated to be Rb ≈ 0.7, where the flow speed near the ground approaches zero. As a result, the high temperature gradient is formed and acts as positive feedback to facilitate a more stratified condition. The accurate and rational correlations obtained in this study will save the computational cost considerably for further research. Also, the available results will be a reference for environmentally friendly designs.</p
Unsupervised learning of depth estimation from imperfect rectified stereo laparoscopic images
No full textBackground: Learning-based methods have achieved remarkable performances on depth estimation. However, the premise of most self-learning and unsupervised learning methods is built on rigorous, geometrically-aligned stereo rectification. The performances of these methods degrade when the rectification is not accurate. Therefore, we explore an approach for unsupervised depth estimation from stereo images that can handle imperfect camera parameters. Methods: We propose an unsupervised deep convolutional network that takes rectified stereo image pairs as input and outputs corresponding dense disparity maps. First, a new vertical correction module is designed for predicting a correction map to compensate for the imperfect geometry alignment. Second, the left and right images, which are reconstructed based on the input image pair and corresponding disparities as well as the vertical correction maps, are regarded as the outputs of the generative term of the generative adversarial network (GAN). Then, the discriminator term of the GAN is used to distinguish the reconstructed images from the original inputs to force the generator to output increasingly realistic images. In addition, a residual mask is introduced to exclude pixels that conflict with the appearance of the original image in the loss calculation. Results: The proposed model is validated on the publicly available Stereo Correspondence and Reconstruction of Endoscopic Data (SCARED) dataset and the average MAE is 3.054 mm. Conclusion: Our model can effectively handle imperfect rectified stereo images for depth estimation.</p
Virtual chain: A storage model supporting cross-blockchain transaction
No full textAt present, blockchain is not only applied in the financial field, but also extended to the medical, insurance, internet of things, and many other fields. Its independence leads to more and more significant problems of network isolation. To address the problem of the cooperative operation between different blockchains, this article proposes virtual chain, a storage model supporting cross-blockchain transaction. It not only ensures that cross-chain transactions are tamper-proof and traceable but also realizes completely decentralized storage. First, to tackle the multidimensional heterogeneous problem existing in cross-chain trading, a data structure based on prefix tree is proposed to manage cross-chain trading data on top of source chain and target chain. Second, a signature verification mechanism based on elliptic curve is proposed to meet the operational atomicity requirements of both parties of cross-chain transaction. Third, the data fusion technology is used to expand the block capacity indirectly to achieve the high throughput of cross-chain transactions. Finally, by using the credibility-based access control and block chain technology to ensure the security of the data storage model, and conduct security analysis from three aspects of tamper-proof, data protection and anti-conspiracy node attack. The experiment shows that the proposed cross-chain transaction storage model can realize the efficient storage of the heterogeneous data, and satisfy the high requirement of the cross-chain transaction. Moreover, the virtual chain storage model establishes a safe and reliable working environment. © 2020 John Wiley & Sons Ltd</p
Reliability Analysis of Self-propelled Artillery Coordinator Based on Cooperative Simulation Strategy
No full text为了准确计算某自行火炮协调器可靠性，基于协同仿真策略考虑机械、液压和控制系统的参数随机性建立协调器系统参数化模型。采用基于Kriging和Monte Carlo的方法对协调器进行位姿可靠度计算。为了快速提高Kriging模型的准确性，选择使学习函数值最小的样本点代入模型中。提出了一种学习停止条件，保证了样本点符号预测精度且学习次数明显减少。计算结果表明：协调器位姿可靠度为99.82%，所提方法和AK-MCS+U(Active learning and Kriging-based Monte-Carlo Simulation+U function)相比失效概率相差0.001%，功能函数调用次数减少了38.27%，计算时间减少了37.6%。本文方法较好的解决了工程上隐式且非线性程度较高，仿真时间过长的问题。</p