State Estimation Method for GNSS/INS/Visual Multi-sensor Fusion Based on Factor Graph Optimization for Unmanned System

With the development of unmanned driving technology,intelligent robots and drones,high-precision localization,navigation and state estimation technologies have also made great progress.Traditional global navigation satellite system/inertial navigation system(GNSS/INS)integrated navigation systems can provide high-precision navigation information continuously.However,when this system is applied to indoor or GNSS-denied environments,such as outdoor substations with strong electromagnetic interference and complex dense spaces,it is often unable to obtain high-precision GNSS positioning data.The positioning and orientation errors will diverge and accumulate rapidly,which cannot meet the high-precision localization requirements in large-scale and long-distance navigation scenarios.This paper proposes a method of high-precision state estimation with fusion of GNSS/INS/Vision using a nonlinear optimizer factor graph optimization as the basis for multi-source optimization.Through the collected experimental data and simulation results,this system shows good performance in the indoor environment and the environment with partial GNSS signal loss.

State-of-health estimation for fast-charging lithium-ion batteries based on a short charge curve using graph convolutional and long short-term memory networks

A fast-charging policy is widely employed to alleviate the inconvenience caused by the extended charging time of electric vehicles. However, fast charging exacerbates battery degradation and shortens battery lifespan. In addition, there is still a lack of tailored health estimations for fast-charging batteries;most existing methods are applicable at lower charging rates. This paper proposes a novel method for estimating the health of lithium-ion batteries, which is tailored for multi-stage constant current-constant voltage fast-charging policies. Initially, short charging segments are extracted by monitoring current switches,followed by deriving voltage sequences using interpolation techniques. Subsequently, a graph generation layer is used to transform the voltage sequence into graphical data. Furthermore, the integration of a graph convolution network with a long short-term memory network enables the extraction of information related to inter-node message transmission, capturing the key local and temporal features during the battery degradation process. Finally, this method is confirmed by utilizing aging data from 185 cells and 81 distinct fast-charging policies. The 4-minute charging duration achieves a balance between high accuracy in estimating battery state of health and low data requirements, with mean absolute errors and root mean square errors of 0.34% and 0.66%, respectively.

融合Graph state LSTM与注意力机制的跨句多元关系抽取

已有的跨句多元关系抽取工作将输入文本表示为集成句内和句间依赖关系的复杂文档图,但图中包含的噪声信息会影响关系抽取的效果。针对这种情况,该文利用Graph state LSTM获得上下文信息,再分别利用词级注意力机制或位置感知的注意力机制,自动聚焦在对关系抽取起到决定性作用的关键词上,降低噪声信息的影响。并且比较了两种注意力机制对使用Graph state LSTM进行关系抽取的影响。通过在一个重要的精确医学数据集上进行实验,验证了该文所提出模型的有效性。

Feasible schemes for preparing all five-photon graph states

We propose feasible experimental schemes for preparing all five-photon graph states. Our schemes require only linear optical elements, photon detectors and post-selection, which are available in current experiment so that these schemes are within the reach of the current technology.

Generation of various multiatom entangled graph states via resonant interactions

In this paper, a scheme for generating various multiatom entangled graph states via resonant interactions is proposed. We investigate the generation of various four-atom graph states first in the ideal case and then in the case in which the cavity decay and atomic spontaneous emission are taken into consideration in the process of interaction. More importantly, we improve the possible distortion of the graph states coming from cavity decay and atomic spontaneous emission by performing appropriate unitary transforms on atoms. The generation of multiatom entangled graph states is very important for constructing quantum one-way computer in a fault-tolerant manner. The resonant interaction time is very short, which is important in the sense of decoherence. Our scheme is easy and feasible within the reach of current experimental technology.

Feasible schemes for preparation of all five-atom graph states

We propose feasible schemes for preparation of all five-atom graph states by cavity quantum electrodynamics （QED）. Our schemes require only the atom-cavity interaction with a large detuning which is available in current experiment so that these schemes are within the reach of the current technology.

Measures of genuine multipartite entanglement for graph states

Graph states are special multipartite entangled states that have been proven useful in a variety of quantum information tasks. We address the issue of characterizing and quantifying the genuine multipartite entanglement of graph states up to eight qubits. The entanglement measures used are the geometric measure, the relative entropy of entanglement, and the logarithmic robustness, have been proved to be equal for the genuine entanglement of a graph state. We provide upper and lower bounds as well as an iterative algorithm to determine the genuine multipartite entanglement.

Improved frequency standard via weighted graph states

We study the spin squeezing property of weighted graph states,which can be used to improve sensitivity in interferometry.We study the time evolution of spin squeezing under local decoherence acting independently on each qubit.Based on the analysis,the spin squeezing of the weighted graph states is somehow robust in the presence of decoherence and the decoherence limit in the improvement of the interferometric sensitivity is still achievable.Furthermore,one can obtain the optimal improvement of sensitivity by tuning the weighted of each edges of the weighted graph state.

Entanglement and Closest Product States of Graph States with 9 to 11 Qubits

The numbers of local complimentary inequivalent graph states for 9, 10 and 11 qubit systems are 440, 3132, 40457, respectively. We calculate the entanglement, the lower and upper bounds of the entanglement and obtain the closest product states for all these graph states. New patterns of closest product states are analyzed.

Generation of Multi-ion Graph States with Trapped Ions in Thermal Motion

We propose a scheme for generating multi-ion graph states using many trapped ions in thermal motion.Our generation scheme is insensitive to external state since the interaction between ions and laser fields does not involvethe external degree of freedom.The scheme can be well realized within the current experimental technique.

基于多维能力和知识图谱-多层感知机的变压器运行状态画像构建方法

利用大数据和画像技术对电力变压器运行状态进行准确评价有利于保障电力系统的安全稳定运行。针对电力变压器运行状态传统评价方法存在的评价维度过于单一、主观性较强等不足,提出了一种基于多维能力和知识图谱-多层感知机的变压器运行状态画像构建方法。首先,构建了由绝缘水平、负载能力、抗短路能力、能效等级和调压能力五个能力构成的变压器运行状态画像体系;然后,融合知识图谱(knowledge graph,KG)与多层感知机(multilayer perceptron,MLP),建立了一种变压器运行状态画像分析模型;最后,基于某地区1368台110kV变压器的实际运行数据,开展了变压器运行状态画像的实例分析,并与随机森林(random forest,RF)和支持向量机(support vector machine,SVM)方法的画像分析结果进行对比。研究结果表明,所提方法对变压器运行状态画像的准确率达到96.35%,优于RF算法(准确率89%)和SVM算法(准确率77%),为电力变压器的运行状态评价提供了一种新思路。

一种用于多域对话状态追踪的知识增强与自注意力引导的图神经网络

对话状态追踪是对话系统的重要组成部分,旨在从用户与系统的对话中跟踪用户意图,其通常表示为槽位-槽值对序列.近年来,深度神经网络模型在对话状态追踪问题上取得了较大进展.然而,现有模型在槽位相关性建模方面还存在可拓展性差与易引入噪声等问题.针对上述问题,本文提出了一种知识增强与自注意力引导的图神经网络KESA-GNN(Knowledge-Enhanced&Self-Attention Guided Graph Neural Network).首先,KESA-GNN通过外部知识嵌入增强槽的语义表征提升多头自注意力机制对槽位间相关性的辨别能力.其次,为了精确建模槽位间的诸如共指、共现等相关性,提出了一种自注意力引导的图神经网络建模槽位相关性.该网络采用多头注意力机制获得槽位间的注意力矩阵以及槽位表征,通过Max-N Relation算法获得注意力矩阵中强相关关系集,将稠密的注意力矩阵稀疏化,从而引导图神经网络中强相关槽位间的信息传播,降低无关槽位的噪声影响.最后,KESA-GNN采用门控融合机制过滤槽位多头注意力和图神经网络输出的槽位表征,从而获取更准确的槽位表征向量,进一步提升了KESA-GNN的鲁棒性.在多域对话数据集上的实验结果表明,KESA-GNN模型的性能优于最新的基线模型.

基于态靶辨治代谢疾病智能辅助诊疗系统的研究与设计

目的:基于“态靶辨证”中医诊疗思维,利用人工智能技术为代谢病中医辅助诊疗提供智能化支持。方法:通过自然语言处理技术,对医学文献和专家经验进行信息抽取和实体识别,构建包含疾病状态、治疗靶点及相关辨证规则的术语知识库。在此基础上,利用图数据库和深度学习算法,构建态靶辨证知识图谱,表达不同疾病状态与治疗靶点之间的复杂关系。结果:通过模型训练和知识推理,结合病例的具体症状和体质,推荐相应的中医治疗方案,提高了代谢病辅助诊疗的准确性和效率。结论:本研究为中医辨证施治提供了新的方法和思路,不断迭代优化算法,拓展临床应用,推动中医药辅助诊疗的发展和完善,为中医药传承创新奠定良好的技术基础。

美国蒙特克莱尔州立大学图论教学特点研究及启示

图论是理工科专业学生开设的一门重要基础课程。该文研究美国蒙特克莱尔州立大学图论课程的教学方法,发现其在教学过程中融入大量应用实例,增强学生的学习主动性,完美达到理论联系实际的教学目标。分析其应用型教学特点,可以在国内图论课程的教学中,注重引入相关应用及研究理念,激发学生学习兴趣,明确图论课程的实际应用价值,对于图论课程的教学改革和课程建设具有一定的参考意义。

考虑运行状态信息的综合能源系统图强化学习优化调度

“双碳”背景下,异质能源的耦合加剧迫使综合能源系统(integrated energy system, IES)拓扑朝着更复杂、更灵活的方向不断演变。然而,现有优化调度方法对非欧网络拓扑知识及其异质潮流约束考虑不足。针对这一问题,提出一种基于图强化学习的综合能源系统优化调度方法。首先,基于图理论在保证节点多样状态的情况下,将异质能源网络拓扑转换为网络图模型。其次,通过建立基于真实图映射的状态-动作-奖励的框架,利用图强化学习的方法学习图模型的非欧拓扑信息,将异质潮流知识加入系统节点运行状态,从而实现IES的安全优化调度。最后,利用某工业园区的真实数据进行仿真验证,所提方法相对于传统方法有效缓解了节点电压越限的问题。结果表明,所提方法能够在考虑IES真实拓扑运行状态信息和异质潮流安全的情况下实现IES的优化调度。

静息态功能磁共振成像在癫痫中的研究进展

癫痫是一种慢性神经系统疾病,其分类广泛、机制复杂,具有反复发作和不可预测的特点,对患者的生活造成了一定的影响。深度了解其病理生理机制对疾病的治疗和提高患者生活质量十分重要,静息态功能磁共振成像(resting-state functional magnetic resonance imaging,rs-fMRI)现已成为探究癫痫脑功能改变的有效方法。目前基于rs-fMRI研究癫痫的数据分析方法主要有低频振幅(amplitude of low frequency fluctuation,ALFF)、局部一致性(regional homogeneity,ReHo)、功能连接(function connection,FC)和图论分析。本文将对rs-fMRI的各种分析方法在癫痫中的应用作一综述,为癫痫的病理生理机制及术前定位、治疗提供重要参考。

基于知识图谱和熵值优化的高桩码头模糊评估

高桩码头应用广泛,在其服役期内可能受到超载、淤积、环境腐蚀等因素影响,出现结构性能退化,对其进行服役性态评估具有重要价值。基于已有行业规范与工程案例整理了中国常见码头结构形式的服役性态评估指标体系,采用Top-down方式创建指标体系的知识图谱,并利用visNetwork工具包编写指标体系内核代码与BS架构交互界面。结合AHP-FCE法与评估指标进行码头服役性态评估:建立评语集与评估等级集进而得到模糊评估向量;采用熵值法(EM)对层次分析(AHP)中的权值进行修正,从而实现客观赋权。最后,以中国东南沿海某海港高桩码头结构为例进行流程演示,结合该工程特点从知识图谱中抽取安全评估指标并建立该工程分级评估系统,分层计算模糊评估向量并得到码头整体安全等级为A级。该结果与工程检测机构评估结果一致,验证了所建知识图谱的适用性与修正的AHP-FCE法的正确性,说明该方法可推广应用于其他各类码头结构的运维评估。

ErlangShen:基于流水线低访问开销的图式区块链高效事务执行机制

基于有向无环图(DAG)的图式区块链能够显著提升系统性能,已成为近年来业界的研究热点。相较于传统串行化的链式区块链,图式区块链可在单位时间内并发处理大量事务从而提升吞吐量。随着事务量的激增,图式区块链面临事务执行效率低的瓶颈问题,即海量事务执行对状态数据访问的需求大幅增加,导致高昂的输入/输出(I/O)开销。实现低I/O访问主要包括两方面的全新挑战:一方面,图式区块链若直接采用传统的事务预取机制,将因执行逻辑不一致引入大量的陈旧读;另一方面,针对不同账户的状态访问会在默克尔树的高层节点中造成重复的I/O开销。为此,本文设计基于流水线的图式区块链高效事务执行机制—ErlangShen,包括Epoch粒度的状态预取机制和默克尔高层路径缓存机制来分别减少陈旧读的数量和重复的I/O开销。具体而言,ErlangShen充分分析并利用了事务访问频次的冷热特征,将访问热事务的逻辑执行与冷事务的状态预取并行化,以避免状态预取对事务执行的影响。此外,为了进一步提升事务执行的吞吐量,根据访问冷热状态事务的特性设计了定制化的并发控制方法。实验结果表明,ErlangShen机制能够减少约90%的陈旧读数量,与最新图式区块链事务处理机制Nezha相比,可将性能提升3~4倍。

基于局部攻击图的最小关键漏洞集分析方法

为缓解攻击图应用在工业互联网安全防护中的状态空间爆炸问题,提出一种基于局部攻击图的最小关键漏洞集分析方法。提出一种以重要资产节点为目标的局部攻击图生成算法,通过裁剪不可达目标节点的攻击路径缓解状态空间爆炸问题;基于局部攻击图生成过程中得到的攻击路径漏洞集直接进行最小关键漏洞集分析,节省传统分析方法在搜索关键漏洞过程中对攻击图进行二次遍历的时空开销。在此基础上,通过工业网络实例进行分析并开展相关工作比较,实验结果表明,所提方法合理可行,可高效分析网络系统中的最小关键漏洞集。

基于符号执行的智能合约重入漏洞检测

在智能合约安全问题中,利用重入漏洞是最具破坏性的攻击之一。针对目前相关检测工作漏报率和误报率高的问题,提出一种基于符号执行的重入漏洞检测方法。该方法基于静态符号执行技术,在模拟以太坊虚拟机指令执行过程中,通过将可能被外部合约多次调用的公有函数控制流子图连接到被调用合约的控制流图,构建出能够模拟重入攻击的完全控制流图,再结合合约状态一致性检测,实现同函数、跨函数和跨合约等3种不同类型的重入漏洞检测。基于该方法设计的检测工具Lucifer与相关工作Oyente、Securify、DefectChecker、Sailfish在已知标签数据集、漏洞注入数据集、自定义数据集和以太坊智能合约真实数据集上进行对比,实验结果表明,Lucifer在误报率、漏报率和容错性上均分别获得第一或者第二的成绩,部分检测情形准确率达到100%,由综合评价指标可以看出,Lucifer的检测率优于现有检测工具,在对于部分特定重入合约的情形尤其在与互斥锁和函数修饰符有关的重入漏洞的识别中有较好的识别能力,在检测时长上,Lucifer的检测时间较久但也在可控范围,并未出现检测超时。