基于时空聚类的出租车载客热点区域挖掘研究

针对出租车运营过程缺少路径优化指导造成运营能力分布不均、空载率高的问题,本文以成都市安装有GPS设备的出租车所采集的轨迹数据为研究对象,以提高出租车效益为目标,采用了一种基于网格的出租车载客热点聚类算法,通过对出租车GPS轨迹数据进行处理和聚类分析,充分挖掘出租车载客热点区域,从而为出租车的运营者和管理者提供信息决策服务。     

基于Spark的城市居民出行时空特征快速提取

利用交通轨迹挖掘人类活动规律是实现智慧城市与生活的基础,然而常规方法已无法满足高效处理的大数据量要求。以海量出租车轨迹数据分析为例,设计了基于Spark平台的城市居民出行时空特征快速提取方法。借助Spark分布式计算框架,将核心算子匹配应用于二次排序、数据清洗及O/D点提取等预处理工作,并在时间分割和网格剖分统计的基础上完成时空特征提取。利用2014年8月份成都市出租车轨迹数据开展实验验证,结果表明,相较于传统单机环境,在确保准确性的基础上,该方法能够大幅提高特征提取的效率;集群节点一定时,数据规模越大优势越明显;数据规模足够大时,Spark集群节点数量越多,耗时越少。     

轨迹延续性与影像特征相似性结合的城市道路提取

城市道路区域检测是城市土地管理、交通规划等领域的迫切需求,而传统城市道路区域检测多使用轨迹提取、遥感解译、人工采集等单独方式,在自动化程度或提取质量上存在一定的局限性。本文结合GNSS轨迹点与高分遥感影像各自的数据优势,提出一种基于轨迹延续性与影像特征相似性的遥感影像道路区域检测方法。该方法以出租车GNSS轨迹点构建轨迹特征栅格,基于轨迹延续性在平均方向特征栅格中划分路段对象,利用道路对象的光谱特征向轨迹无法覆盖的小区内部进行拓展,以获得提取区域内较为完整的道路信息。试验证明：本文方法可以有效降低道路的同物异谱现象及阴影、树木遮挡的影响,高效地提取高分遥感影像中的道路区域。与传统的遥感影像分类方法相比,具有更高的精度与自动化程度,相较于深度学习模型具有更广的适应性。     

空间认知导向下利用分层强化学习的最优路径规划

针对空间认知导向下模型驱动型路径规划和人们认知偏好多样性之间的矛盾,提出了一种基于分层强化学习的交互学习型路径规划方法。该方法将最优路径标准转换为路口处转向决策的瞬时奖励值,并通过预学习和实时学习两个阶段实现高效地发现总奖励值最大的最优路径策略。其中,预学习阶段自动发现子目标节点,并构建包含局部最优策略的子任务;实时学习阶段利用预定义策略实现高效的Q值更新,并根据Q值追溯最优路径。实验表明,该方法具有足够好的实时性和最优性。     

基于改进的粒子群优化神经网络算法的海平面变化预测分析

针对验潮站潮位预报的需求，提出一种基于分群策略的粒子群优化神经网络(SSPSO-BP)的预报方法。该方法通过建立多个不同功能且具有交流能力的智能粒子群，经SSPSO和BP的两次优化，构建潮高预报模型。实验研究表明，SSPSO-BP模型在Oga站的潮位资料上高、低潮位间的时刻基本保持一致，高潮时刻最大潮高差为7.37 cm,低潮时刻最大潮高差为4.21 cm,该模型比标准BP神经网络及PSO优化神经网络在准确度和精度上有了很大的提高，其平均绝对误差、均方误差相较于BP神经网络分别提升了16.2%、79.2%,相较于PSO-BP神经网络提升了13.9%、79.6%。     

面向建筑物边缘特征保持的遥感影像变化检测研究

随着深度学习技术在遥感变化检测领域的发展,许多基于卷积神经网络(CNNs)的变化检测算法得到了广泛的应用,但在建筑物变化检测中,仍存在样本不平衡、边界模糊等问题。为解决上述问题,提出基于孪生U-Net和边缘优化策略(Siamese U-Net and Edge Refinement Strategies, Siam-Unet-ERS)的建筑物变化检测方法。首先提出了边界优化策略,共享建筑物变化检测主任务和边界提取子任务之间的空间特征,以实现提升建筑物边界定位精度的目的。同时,设计了残差捷径模块,在不降低模型性能的同时提升模型收敛速率。网络在LEVIR-CD和WHU数据集上都经过验证和测试,分别达到了97.79%和98.92%的总体精度。实验结果表明,Siam-Unet-ERS方法优于一些以前较先进的变化检测方法。     

改进的半参数模型法在GNSS时间序列季节性信号提取中的应用

GNSS基准站坐标时间序列具有明显的季节性变化。基于谐波函数模型的最小二乘拟合方法只能得到固定振幅的季节性信号,而真实的季节性信号其振幅是变化的。采用半参数模型进行季节性信号提取时又存在最优平滑因子确定困难、迭代速度慢的问题。提出一种赋相对权比的半参数模型,采用迭代更快速的黄金分割法与改进效率法相结合的策略确定最优平滑因子。通过模拟数据实验,验证了改进模型的可用性。实验表明:改进方法的计算效率明显提高,对于10 a长度的模拟数据,相较GCV函数法搜索速度提高68.1%,相较L-曲线法速度提高25.8%;计算精度较最小二乘法和半参数模型法均有提高,所得残差中没有明显的季节性信号。     

基于奇异值分解的DS-InSAR相位优化

高质量监测点空间密度是时序InSAR开展形变监测的重要指标，依靠分布式散射体DS(Distributed Scatters)开展InSAR形变监测可有效解决传统时序InSAR监测点空间密度不够的缺陷，但分布式散射体干涉相位极易受到去相干影响，造成干涉相位失真，因此分布式散射体相位优化是DS-InSAR技术的关键，针对这一情况，本文提出一种新的基于奇异值分解的DS相位优化方法，该方法利用同质像元时序相位重构相位矩阵，对矩阵进行主成分分析得到优化相位。采用模拟数据和33景覆盖郑州东部白沙镇Sentinel-1A数据对提出方法的可靠性与有效性进行验证和分析。采用时序平均相位标准偏差、平均相位梯度、平均残差点数目作为干涉图优化效果评价指标，提出方法优化后干涉图相较于原始干涉图分别下降了15.61%、25.81%、44.84%，与对比DS相位优化方法相比，提出方法对干涉图DS相位优化效果更好，特别是在一些低相干区域仍然可获得较好DS相位优化结果；提出方法在降低DS相位噪声同时可较好地保持地物细节信息。此外，相较于常规PS(Permanent Scatters)技术形变监测结果，本文方法高质量监测...     

顾及耕地占补平衡的土地利用变化及其效应多情景模拟

占补平衡是中国耕地保护的重要措施之一。在耕地保护政策从传统数量规模保护为主向数量-质量-生态一体化保护的发展过程中,较少有研究系统地分析不同占补平衡策略的未来潜在影响作用。以中国山东省招远市为例,构建了顾及耕地占补平衡策略的多情景土地利用变化模拟模型,通过设计基准情景、耕地数量保护、耕地数量-质量保护、耕地数量-生态保护和耕地数量-质量-生态保护5个情景,分析2030年不同耕地保护策略下耕地规模、生产力和生境质量的演化规律。结果表明,相较于基准情景,耕地占补平衡情景下耕地数量平均提高了7.2%,粮食产量、生境质量分别平均提升了0.63%和0.70%,其中耕地数量-质量-生态保护情景下粮食产量和生境质量均高于平均水平。因此,耕地数量-质量-生态保护情景更有利于多种目标的协同优化。本研究为系统地对比评估多种占补平衡策略,优化耕地保护政策提供了方法支持。     

交通网络旅行商路径优化的遗传禁忌搜索算法

旅行商路径优化问题是经典的网络分析问题之一。由于旅行商问题具有NP Hard特性,主要通过智能优化方法或启发式算法来获得近似最优解。然而,单一智能优化方法存在运算量过大、参数选择苛刻,对初值依赖性强等缺陷,很难快速实现全局优化。结合多种优化机制和邻域搜索结构设计混合启发式算法可在一定程度上解决这一问题。本文结合遗传算法的全局寻优能力和禁忌搜索的记忆功能,设计实现了一种基于分散集中策略的禁忌遗传算法,即采用遗传变异算子作为分散策略构造邻域,开辟新的搜索空间,有效提升获得全局最优解的概率;将禁忌搜索作为集中策略进行局部寻优,避免迂回探测,充分体现禁忌搜索较强的“爬山”能力,并通过实际交通网络和不同规模的节点集合,从求解精度、稳定性和效率三个方面对算法进行了评价。结果表明,本文提出的交通网络旅行商路径优化的禁忌遗传算法平均求解精度比禁忌搜索算法提高了9%,略优于ArcGIS;当与ArcGIS求解的TSP路径长度差异在1%以内时,禁忌搜索算法已经难以获得对应精度的TSP路径,而禁忌遗传算法效率比遗传算法提高了50%。且禁忌遗传算法具有很好的并行化潜力。     

A network distance and graph-partitioning-based clustering method for improving the accuracy of urban hotspot detection

Clustering is an important approach to identifying hotspots with broad applications, ranging from crime area analysis to transport prediction and urban planning. As an on-demand transport service, taxis play an important role in urban systems, and the pick-up and drop-off locations in taxi GPS trajectory data have been widely used to detect urban hotspots for various purposes. In this work, taxi drop-off events are represented as linear features in the context of the road network space. Based on such representation, instead of the most frequently used Euclidian distance, Jaccard distance is calculated to measure the similarity of road segments for cluster analysis, and further, a network distance and graph-partitioning-based clustering method is proposed for improving the accuracy of urban hotspot detection. A case study is conducted using taxi trajectory data collected from over 6500 taxis during one week, and the results indicate that the proposed method can identify urban hotspots more precisely.     

武汉市出租车轨迹二氧化碳排放的时空模式分析

车辆尾气是道路交通二氧化碳（CO₂）排放的重要来源，目前的研究主要集中在区域CO₂排放清单的计算和分析上，鲜有从微观层面上进行CO₂排放的反演分析。采用微观尾气排放模型对出租车行程轨迹的CO₂排放量进行定量反演，进而从点、线、面3个粒度对武汉市出租车行程轨迹CO₂排放的时空模式进行分析。研究结果表明，利用出租车轨迹可以有效地反演出租车CO₂排放量，并发现其在不同日期和时段具有明显的规律性。通过时空聚类技术发现了点粒度下的出租车CO₂排放的类簇数目存在一定的时空变化规律，采用数据场模型展示了线粒度下的道路线段出租车CO₂排放强度存在明显的时空分布规律，利用时空自相关技术揭示了面粒度下的区域出租车CO₂排放量具有较高的时空正相关性。研究成果可以为城市减排措施制定等提供辅助支持。     

Online monitoring of local taxi travel momentum and congestion effects using projections of taxi GPS-based vector fields

Ubiquitous taxi trajectory data has made it possible to apply it to different types of travel analysis. Of interest is the need to allow someone to monitor travel momentum and associated congestion in any location in space in real time. However, despite an abundant literature in taxi data visualization and its applicability to travel analysis, no easy method exists. To measure taxi travel momentum at a location, current methods require filtering taxi trajectories that stop at a location at a particular time range, which is computationally expensive. We propose an alternative, computationally cheaper way based on preprocessing vector fields from the trajectories. Algorithms are formalized for generating vector kernel density to estimate a travel-model-free vector field-based representation of travel momentum in an urban space. The algorithms are shared online as an open source GIS 3D extension called VectorKD. Using 17 million daily taxi GPS points within Beijing over a 4-day period, we demonstrate how to generate in real time a series of projections from a continuously updated vector field of taxi travel momentum to query a point of interest anywhere in a city, such as the CBD or the airport. This method allows a policy-maker to automatically identify temporal net influxes of travel demand to a location. The proposed methodology is shown to be over twenty times faster than a conventional selection query of trajectories. We also demonstrate, using taxi data entering the Beijing Capital International Airport and the CBD, how we can quantify in nearly real time the occurrence and magnitude of inbound or outbound queueing and congestion periods due to taxis cruising or waiting for passengers, all without having to fit any mathematical queueing model to the data.     

出租车轨迹数据挖掘进展

大数据、物联网与精密定位技术的发展推动了城市感知的进步。随着社会活动的与日俱增,出租车轨迹数据不仅记录了出租车的行车轨迹,还蕴藏着道路交通状态、城市居民出行规律、城市结构及其他社会问题。通过各种数据分析与挖掘手段对出租车轨迹数据进行深入探究,对于智能交通、城市规划等有着重要意义。本文综述了近十年国内外基于出租车轨迹大数据的相关研究,按照空间统计方法、时间序列方法、图论与网络方法及机器学习方法等4类,详细阐述各类方法的研究现状。随后,本文分析了现有研究的应用领域、热点主题和发展趋势。最后,本文指出了出租车轨迹数据挖掘研究领域面临的挑战和未来研究方向。     

Extracting commuter-specific destination hotspots from trip destination data – comparing the boro taxi service with Citi Bike in NYC

ABSTRACT

Taxi trajectories from urban environments allow inferring various information about the transport service qualities and commuter dynamics. It is possible to associate starting and end points of taxi trips with requirements of individual groups of people and even social inequalities. Previous research shows that due to service restrictions, boro taxis have typical customer destination locations on selected Saturdays: many drop-off clusters appear near the restricted zone, where it is not allowed to pick up customers and only few drop-off clusters appear at complicated crossing. Detected crossings imply recent infrastructural modifications. We want to follow up on these results and add one additional group of commuters: Citi Bike users. For selected Saturdays in June 2015, we want to compare the destinations of boro taxi and Citi Bike users. This is challenging due to manifold differences between active mobility and motorized road users, and, due to the fact that station-based bike sharing services are restricted to stations. Start and end points of trips, as well as the volumes in between rely on specific numbers of bike sharing stations. Therefore, we introduce a novel spatiotemporal assigning procedure for areas of influence around static bike sharing stations for extending available computational methods.     

基于北斗/GPS双模的纯电动乘用车行车状态数据采集终端设计及其车载信息服务应用

基于北斗/GPS双模定位方式,采用北斗定位优先策略,设计了纯电动乘用车行车状态数据采集终端,并建立了相应的应用平台。通过纯电动乘用车车载信息服务框架的设计,列举了基于北斗/GPS双模的纯电动乘用车行车状态数据采集终端的车载信息服务方式与具体内容,为纯电动乘用车电池、电机、电能管理等提供一种可参考的方案。     

网络约束下局部空间同位模式的扫描统计方法

空间同位模式挖掘旨在发现空间数据库中频繁发生在邻近位置的地理事件。由于空间异质性,地理事件在不同区域邻近出现的频繁程度亦存在差异,进而形成局部同位模式。现有局部同位模式挖掘方法多基于欧氏空间的平面假设,难以客观揭示网络空间（如城市道路）内地理事件间的局部同位规律,因此基于空间扫描统计思想,提出了一种网络约束下的局部同位模式挖掘方法。首先,发展了网络约束下的路径扩展方法,识别可能存在局部网络空间同位模式的候选路径;其次,基于网络约束下的二元泊松分布构建显著性检验的零模型,判别候选路径中局部网络空间同位模式的有效性。通过模拟实验与北京市出租车供需模式分析,发现该方法比现有方法得到的结果更精细、更客观,能够有效地挖掘网络约束下的局部同位模式。     

Inferring trip purposes and uncovering travel patterns from taxi trajectory data

Global positioning system-enabled vehicles provide an efficient way to obtain large quantities of movement data for individuals. However, the raw data usually lack activity information, which is highly valuable for a range of applications and services. This study provides a novel and practical framework for inferring the trip purposes of taxi passengers such that the semantics of taxi trajectory data can be enriched. The probability of points of interest to be visited is modeled by Bayes’ rules, which take both spatial and temporal constraints into consideration. Combining this approach with Monte Carlo simulations, we conduct a study on Shanghai taxi trajectory data. Our results closely approximate the residents’ travel survey data in Shanghai. Furthermore, we reveal the spatiotemporal characteristics of nine daily activity types based on inference results, including their temporal regularities, spatial dynamics, and distributions of trip lengths and directions. In the era of big data, we encounter the dilemma of “trajectory data rich but activity information poor” when investigating human movements from various data sources. This study presents a promising step toward mining abundant activity information from individuals’ trajectories.     

Discovering spatial and temporal patterns from taxi-based Floating Car Data: a case study from Nanjing

Floating Car Data (FCD) refers to the trajectories of vehicles equipped with Global Positioning System-enabled devices that automatically record location-related data within a short time interval. As taxies in Chinese cities continually drive along the streets seeking passengers, FCD can easily traverse the entire street network in a city on a daily basis. Taking advantage of this situation, this study extracted passenger pickup and drop-off locations from FCD sourced from 6445 taxis over a 2-week period in Nanjing, China to discover human behavioral patterns and the dynamics behind them. In this study, road nodes are converted to the points, based on which Thiessen polygons are generated to divide the study area into small areas with the goal of exploring the spatial distribution of pickup and drop-off locations. Moran’s I index is used to calculate the spatial autocorrelation of the spatial distribution of pickup and drop-off locations, and hot spot analysis is used to identify statistically significant spatial clusters of hot and cold spots. The spatial and temporal patterns of FCD in the study area are investigated, and the results show that: (1) the temporal patterns show a strong daily rhythm, (2) the spatial patterns show that the number of pickup and drop-off locations gradually diminish from the downtown areas to the outer suburbs, (3) the spatiotemporal patterns exhibit large differences over time, and (4) the driving forces explored by regression models indicate that population density and transportation density are consistent with the population distribution, but per capita disposable income is not consistent with the population distribution.