WO2023098076A1 - 一种基于多维度影响因素的道路安全评价方法及系统 - Google Patents

一种基于多维度影响因素的道路安全评价方法及系统 Download PDF

Info

Publication number
WO2023098076A1
WO2023098076A1 PCT/CN2022/103535 CN2022103535W WO2023098076A1 WO 2023098076 A1 WO2023098076 A1 WO 2023098076A1 CN 2022103535 W CN2022103535 W CN 2022103535W WO 2023098076 A1 WO2023098076 A1 WO 2023098076A1
Authority
WO
WIPO (PCT)
Prior art keywords
sub
road
traffic
area
safety
Prior art date
Application number
PCT/CN2022/103535
Other languages
English (en)
French (fr)
Inventor
郭延永
丁红亮
吴瑶
刘攀
刘佩
Original Assignee
东南大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 东南大学 filed Critical 东南大学
Priority to US17/925,748 priority Critical patent/US11887472B2/en
Publication of WO2023098076A1 publication Critical patent/WO2023098076A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/0104Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125Traffic data processing
    • G08G1/0129Traffic data processing for creating historical data or processing based on historical data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/11Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/0104Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125Traffic data processing
    • G08G1/0133Traffic data processing for classifying traffic situation
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/0104Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/0104Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G08G1/0141Measuring and analyzing of parameters relative to traffic conditions for specific applications for traffic information dissemination

Definitions

  • the present invention relates to the technical field of road safety, in particular to a road safety evaluation method and system based on multi-dimensional influencing factors.
  • the rate of car ownership has gradually increased, which not only causes road congestion, but also increases the incidence of road traffic accidents.
  • road safety analysis model includes two levels, one is the road safety analysis model at the macro level, and the other is the road safety analysis model at the micro level, but there is no relevant research or patent field.
  • the study comprehensively considers the correlation between the road safety analysis models at the macro and micro levels. Establishing a road safety analysis model only from a single-dimensional perspective will cause certain deviations in the analysis results.
  • the annual average daily traffic volume of motor vehicles is regarded as an effective safety risk exposure, which is of great significance for measuring the influencing factors and the mechanism of accidents.
  • relevant literature assumes that the impact of safety risk exposure is constant, and the impact should be elastic in nature, and each influencing factor will have similarities and differences as the annual average daily motor vehicle traffic volume changes.
  • the purpose of the present invention is to provide a road safety evaluation method and system based on multi-dimensional influencing factors to solve the problems in the prior art.
  • the present invention provides the following technical solutions:
  • the first aspect of the present invention proposes a road safety evaluation method based on multi-dimensional influencing factors.
  • a safety evaluation model is constructed through steps A to D, and the safety evaluation model is applied.
  • steps From E to step F obtain the influencing factors affecting the safety of each traffic road in the sub-area, and perform safety evaluation on the sub-area:
  • Step A for the sub-area, periodically obtain the historical traffic data of the sub-area within the preset time length, and the historical traffic data of each traffic road in the sub-area within the preset time length, and then enter step B;
  • Step B Based on the historical traffic data of the sub-area within the preset time period and the historical traffic data of each traffic road in the sub-area within the preset time length, obtain the safety risk exposure corresponding to the sub-area, and each Each traffic road corresponds to each safety risk exposure, and quantifies each safety risk exposure to obtain each classification variable T corresponding to each safety risk exposure, and then enters step C;
  • Step C for each traffic road included in the sub-area, based on the corresponding historical traffic data and each classification variable T obtained in step B, construct a road safety quantitative sub-model, that is, obtain the sub-area Road safety quantitative sub-models corresponding to each traffic road;
  • step D Based on the road safety quantitative sub-model corresponding to each traffic road in the sub-region and the historical traffic data of the sub-region, construct the regional safety quantitative sub-model corresponding to the sub-region, and then enter step D;
  • Step D For the sub-region, the model group composed of the regional safety quantitative sub-model corresponding to the sub-region and the road safety quantitative sub-model corresponding to each traffic road in the sub-region is used as the safety evaluation model corresponding to the sub-region, And the input volume of each sub-model in the model group is its corresponding historical traffic data;
  • Step E according to the method in step A to step C, based on the actual traffic data of the sub-area and the actual traffic data of each traffic road in the sub-area, obtain the regional safety quantification sub-model corresponding to the sub-area, and each road safety quantization sub-model model, then enter step F;
  • Step F For the sub-area, apply the safety evaluation model according to the method in step D, and use the constraint function as the target to solve the regional safety quantification sub-model corresponding to the sub-area and each road safety quantization sub-model, and obtain the affected sub-area road Influencing factors of safety, according to the influencing factors, the safety evaluation of the sub-area and each traffic road in the sub-area is carried out.
  • the aforementioned period obtains the historical traffic data of each sub-area within the limited area within the preset time length
  • the historical traffic data corresponding to each sub-area respectively includes: the population density N of the sub-area, the GDP of the sub-area, the sub-area Road network density K, annual average daily traffic volume of motor vehicles in the sub-region AADT1, proportion of green area in the sub-region L1, proportion of residential areas in the sub-region L2, proportion of non-residential areas in the sub-region L3, proportion of road area in the sub-region L4, And the average driving speed V in the sub-area;
  • the historical traffic data corresponding to each traffic road in each sub-region includes: traffic road length D, traffic road lane number J, traffic road width W, traffic road is equipped with special lane Q, annual average motor vehicle traffic road The daily traffic volume AADT2, the intersection density A of the traffic road, and the class D of the traffic road.
  • step B based on the historical traffic data of the sub-region within the preset time length and the historical traffic data of each traffic road in the sub-region within the preset time length, for each traffic corresponding to the sub-region road, according to the following formula:
  • step F the constraint function in the aforementioned step F is as follows:
  • the safety evaluation model is trained, and the regional safety quantification sub-model corresponding to the sub-region and the safety influence coefficient in each road safety quantization sub-model are solved under the constraint condition, and the safety influence coefficient is obtained in Significant degree within the 95% confidence interval.
  • the safety impact coefficient is positively significant within the 95% confidence interval, the traffic data corresponding to the safety impact coefficient will increase the incidence of traffic accidents on the road.
  • the safety impact coefficient is within the 95% confidence interval If the interval is negative and significant, the traffic data corresponding to the safety impact coefficient will reduce the incidence of traffic accidents on the traffic road.
  • the second aspect of the present invention proposes a road safety evaluation system based on multi-dimensional influencing factors, including:
  • processors one or more processors
  • the memory stores executable instructions. When the instructions are executed by one or more processors, the one or more processors execute the process including any one of the road safety evaluation methods.
  • a third aspect of the present invention provides a computer-readable medium storing software comprising instructions executable by one or more computers that, when executed by the one or more computers, perform any An operation of the road safety evaluation method.
  • a road safety evaluation method and system based on multi-dimensional influencing factors described in the present invention compared with the prior art by adopting the above technical scheme, has the following technical effects:
  • the present invention Based on the median of each traffic data, the present invention obtains the safety risk exposure corresponding to the sub-region and the safety risk exposure corresponding to each traffic road contained in the sub-region, and further obtains the corresponding safety risk exposure Classification variables, considering the elastic change of safety risk exposure, make the change of annual average daily motor vehicle traffic volume in the target area affected by various influencing factors, and the evaluation results of road safety are more objective and authentic.
  • Fig. 1 is a schematic flowchart of a road safety evaluation method according to an exemplary embodiment of the present invention.
  • the present invention proposes a road safety evaluation method based on multi-dimensional influencing factors, which can accurately judge the impact of each influencing factor on road accidents on the basis of considering the macroscopic and microscopic road safety analysis models for the limited area respectively.
  • For each sub-area build a safety evaluation model through steps A to D, apply the safety evaluation model, and obtain the influencing factors that affect the safety of each traffic road in the sub-area through the following steps E to F, and perform safety evaluation on the sub-area:
  • the research units are selected under the macro and micro dimensions, the research units under the macro dimension are determined as the traffic analysis area, and the research units under the micro dimension are determined as each research road section in the traffic analysis area.
  • Step A for the traffic analysis area, periodically obtain the historical traffic data of the traffic analysis area within the preset time period, the historical traffic data of each traffic road in the traffic analysis area within the preset time length, and the corresponding historical traffic data of each traffic analysis area
  • the traffic data include: the population density N of the traffic analysis area, the GDP of the traffic analysis area, the road network density K in the traffic analysis area, the annual average daily traffic volume of motor vehicles in the traffic analysis area AADT1, and the green area ratio of the traffic analysis area L1 , the proportion of residential areas in the traffic analysis area L2, the proportion of non-residential areas in the traffic analysis area L3, the proportion of road area in the traffic analysis area L4, and the average driving speed V in the traffic analysis area.
  • the historical sample data corresponding to the traffic analysis area is as follows: Table 1 shows:
  • the historical traffic data corresponding to each traffic road in the traffic analysis area includes: traffic road length D, traffic road lane number J, traffic road width W, traffic road is equipped with special lane Q, annual average motor vehicle traffic road Daily traffic volume AADT2, traffic road intersection density A, and traffic road grade D, for a single traffic analysis area, the historical traffic data of each traffic road contained in it is shown in Table 2:
  • Step B Based on the historical traffic data of the sub-area b1 within the preset time length and the historical traffic data of each traffic road in the sub-area within the preset time length, obtain the safety risk exposure corresponding to the sub-area and the information contained in the sub-area.
  • AADT i AADT1 or AADT2
  • AADT i AADT1
  • AADT i ' is all sub-regions within the limited area
  • AADT i AADT2
  • AADT i ′ is the median of the annual average daily traffic volume of motor vehicles on all traffic roads in the sub-region, and then enter step C.
  • Step C for each traffic road contained in the sub-area b1, based on the corresponding historical traffic data and each classification variable T obtained in step B, construct a road safety quantitative sub-model, that is, obtain the sub-area
  • the road safety quantization sub-models corresponding to each traffic road in taking the three road sections A1-A3 in sub-area b1 as an example, the corresponding road safety quantization sub-models are respectively:
  • AADT i ′ is the median of the annual average daily traffic volume of motor vehicles on all traffic roads in the sub-region;
  • the regional safety quantitative sub-model corresponding to the sub-region is constructed as in, At this time, AADT i ′ is the median of the annual average daily traffic volume of motor vehicles in all sub-areas within the limited area, and the regional safety quantification sub-model corresponding to the traffic area b1 is:
  • lnE1 1 lnE2 1 +lnE2 2 +lnE2 3 , then enter step D.
  • Step D For the sub-region, the model group composed of the regional safety quantitative sub-model corresponding to the sub-region and the road safety quantitative sub-model corresponding to each traffic road in the sub-region is used as the safety evaluation model corresponding to the sub-region, And the input volume of each sub-model in the model group is its corresponding historical traffic data;
  • Step E according to the method in step A to step C, based on the actual traffic data of the sub-area and the actual traffic data of each traffic road in the sub-area, obtain the regional safety quantification sub-model corresponding to the sub-area, and each road safety quantization sub-model model, then enter step F;
  • Step F For the sub-area, apply the safety evaluation model according to the method in step D, and use the constraint function as the target to solve the regional safety quantification sub-model corresponding to the sub-area and each road safety quantization sub-model, and obtain the affected sub-area road Influencing factors of safety, according to the influencing factors, the safety evaluation of the sub-area and each traffic road in the sub-area is carried out.
  • the influence mechanism of each influencing factor on road safety in different dimensions can be judged separately. If the coefficient of the influencing factor is positively significant in the 95% confidence interval, it means that the influencing factor will increase the accident rate on the traffic area or road section. If the coefficient of the influencing factor is negatively significant in the 95% confidence interval, it means that the influencing factor will reduce the occurrence of accidents in traffic areas or road sections.
  • the experimental verification of this invention is carried out under hypothetical data conditions. Taking the factor N of the traffic area as an example, if ⁇ 1 >0 under the 95% confidence interval, it means that the population density in the traffic area is positively related to the occurrence of road accidents. The greater the population density, the more accidents in the traffic area. If ⁇ 1 ⁇ 0 under the 95% confidence interval, it means that the population density in the traffic area is negatively correlated with the occurrence of road accidents. The greater the population density , the fewer accidents in the traffic area.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Operations Research (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种基于多维度影响因素的道路安全评价方法及系统,涉及道路安全技术领域,基于历史交通数据以及对应的安全影响因素分别构建不同维度下的安全评价模型,并对道路安全风险曝光量进行了弹性分类,通过约束函数链接宏观与微观维度下的安全评价模型并分别判断各安全影响因素的影响机理,具体地,分别针对限定区域范围内的各个子区域,构建并获得安全评价模型,应用安全评价模型,获得影响子区域中各交通道路安全的影响因素,对子区域进行安全评价。对道路安全的评价结果更为精确、全面、客观、具有真实性,适用范围更广。

Description

一种基于多维度影响因素的道路安全评价方法及系统 技术领域
本发明涉及道路安全技术领域,具体而言涉及一种基于多维度影响因素的道路安全评价方法及系统。
背景技术
随着社会经济的发展,小汽车拥有率逐步上升,不仅造成了道路的拥挤,与此同时道路交通事故发生率也逐步上升,为了减少道路事故的发生,提高道路安全性,相关研究领域提出了多种道路安全分析模型,其中道路安全分析模型包含两个层面,一个是宏观维度下的道路安全分析模型,一个是微观层面下的道路安全分析,但是不管在研究领域还是专利领域,都没有相关研究综合考虑宏观与微观层面下的道路安全分析模型之间的关联性。只从单维度视角建立道路安全分析模型会对分析结果造成一定的偏差。此外,年平均机动车日交通量被视为一种有效的安全风险曝光量,对衡量影响因素与事故产生机理具有重要意义。然而相关文献都假设安全风险曝光量的影响是恒定的,本质上该影响应该是弹性变化的,随着年平均机动车日交通量的变化各影响因素会有异同。
发明内容
本发明的目的在于提供一种基于多维度影响因素的道路安全评价方法及系统,以解决现有技术中的问题。
为实现上述目的,本发明提供如下技术方案:
本发明的第一方面提出一种基于多维度影响因素的道路安全评价方法,分别针对限定区域范围内的各个子区域,通过步骤A至步骤D构建安全评价模型,应用安全评价模型,通过以下步骤E至步骤F,获得影响子区域中各交通道路安全的影响因素,对子区域进行安全评价:
步骤A、针对子区域,周期获得子区域在预设时长内的历史交通数据、以及子区域内各个交通道路分别在预设时长内的历史交通数据,随后进入步骤B;
步骤B、基于子区域在预设时长内的历史交通数据、以及子区域内各个交通道路分别在预设时长内的历史交通数据,获得子区域对应的安全风险曝光量、以及子区域所含各个交通道路分别所对应的各个安全风险曝光量,并对各安全风险曝光量进行量化,得到各安全风险曝光量分别所对应的各个分类变量T,随后进入步骤C;
步骤C、分别针对该子区域所包含的各个交通道路,基于其所对应的各历史交通数据、以及步骤B中所获的各个分类变量T,构建道路安全量化子模型,即获得该子区域中各个交通道路分别所对应的道路安全量化子模型;
基于该子区域中各个交通道路分别所对应的道路安全量化子模型、以及该子区域的历史交通数据,构建该子区域所对应的区域安全量化子模型,随后进入步骤D;
步骤D、针对子区域,以该子区域所对应的区域安全量化子模型和该子区域中各个交通道路分别所对应的道路安全量化子模型构成的模型组作为该子区域对应的安全评价模型,并且模型组中各个子模型的输入量为其所对应的历史交通数据;
步骤E、按照步骤A至步骤C中方法,基于子区域的实际交通数据、以及子区域内各个交通道路的实际交通数据,获得子区域所对应的区域安全量化子模型、以及各个道路安全量化子模型,随后进入步骤F;
步骤F、针对该子区域,按照步骤D中方法应用安全评价模型,以约束函数为目标,对子区域对应的区域安全量化子模型、以及各个道路安全量化子模型进行求解,得到影响子区域道路安全的影响因素,根据影响因素对子区域以及该子区域内各个交通道路进行安全评价。
进一步地,前述周期获得限定区域范围内各个子区域在预设时长的历史交通数据,各子区域所对应的历史交通数据分别均包括:子区域的人口密度N、子区域的GDP、子区域内道路网密度K、子区域的机动车年平均日交通量AADT1、子区域绿化面积占比L1、子区域居住区占比L2、子区域非居住区占比L3、子区域道路面积占比L4、以及子区域内的平均行车速度V;
各子区域内的各个交通道路所对应的历史交通数据分别均包括:交通道路长度D、交通道路车道数J、交通道路宽度W、交通道路是否设置有专用车道Q、交通道路的机动车年平均日交通量AADT2、交通道路的交叉口密度A、以及交通道路等级D。
进一步地,前述的步骤B中,基于子区域在预设时长内的历史交通数据、以及子区域内各个交通道路分别在预设时长内的历史交通数据,针对该子区域分别所对应的各个交通道路,根据以下公式:
Figure PCTCN2022103535-appb-000001
得到该子区域、以及对应各个交通道路的风险曝光量分别所对应的各个分类变量T, 其中,AADT i为AADT1或AADT2,当AADT i=AADT1时,AADT i′为限定区域范围内所有子区域的机动车年平均日交通量的中位数,当AADT i=AADT2时,AADT i′为子区域内所有交通道路的机动车年平均日交通量的中位数。
进一步地,前述步骤C中,分别针对子区域所包含的各个交通道路,根据以下公式:
Figure PCTCN2022103535-appb-000002
获得各个交通道路分别所对应的各个道路安全量化子模型lnE2 n,其中,E2为交通道路在预设时间周期内的事故发生量,θ为交通道路的安全影响系数,ε n为道路安全量化子模型的误差项,n的取值范围为1至N,N为各个子区域中分别所包含的交通道路的总数,当交通道路设置有专用车道时Q=1,当交通道路设置有专用车道时Q=0,当道路等级为主干道时D=1,当道路等级为次干道时D=2,当道路等级为支路时D=3,其中,
Figure PCTCN2022103535-appb-000003
此时,AADT i′为子区域内所有交通道路的机动车年平均日交通量的中位数;
分别针对限定区域范围内各个子区域,根据以下公式:
Figure PCTCN2022103535-appb-000004
获得限定区域范围内各个子区域所对应的各个区域安全量化子模型lnE1 m,其中,E1为子区域在预设时间周期内的事故发生量,β为子区域的安全影响系数,ε m为区域安全量化子模型的误差项,m的的取值范围为1至M,M为限定区域范围所包含各个子区域的总数,其中,
Figure PCTCN2022103535-appb-000005
此时,AADT i′为限定区域范围内所有子区域的机动车年平均日交通量的中位数。
进一步地,前述的步骤F中约束函数如下:
Figure PCTCN2022103535-appb-000006
以该约束函数为训练目标,对安全评价模型进行训练,在约束条件下对子区域对应的区域安全量化子模型、以及各个道路安全量化子模型中的安全影响系数进行求解,获得安全影响系数在95%置信区间内的显著程度,当安全影响系数在95%置信区间内正 向显著,则安全影响系数对应的交通数据会增加交通道路上交通事故的发生率,当安全影响系数在95%置信区间内负向显著,则安全影响系数对应的交通数据会降低交通道路上交通事故的发生率。
本发明的第二方面提出一种基于多维度影响因素的道路安全评价系统,包括:
一个或多个处理器;
存储器,存储可被执行的指令,所述指令在通过一个或多个处理器执行时,一个或多个处理器执行包括任意一项所述道路安全评价方法的过程。
本发明的第三方面提出一种存储软件的计算机可读取介质,所述软件包括能通过一个或多个计算机执行的指令,所述指令在被所述一个或多个计算机执行时,执行任意一项所述道路安全评价方法的操作。
本发明所述一种基于多维度影响因素的道路安全评价方法及系统,采用以上技术方案与现有技术相比,具有以下技术效果:
本发明基于各个交通数据中的中位数,获得子区域对应的安全风险曝光量、以及子区域所含各个交通道路分别所对应的各个安全风险曝光量,进一步得到各个安全风险曝光量所对应的分类变量,考虑安全风险曝光量的弹性变化,使得在目标区域内年平均机动车日交通量的变化受各个影响因素的影响,得到对道路安全的评价结果更为客观,更具有真实性,同时,基于多维度考量在多维度条件下构建的安全量化模型,考虑到道路安全在宏观与微观条件下的关联性,对道路安全的评价结果更为精确、全面,该方法的适用范围更为广泛。
附图说明
图1为本发明示例性实施例的道路安全评价方法的流程示意图。
具体实施方式
为了更了解本发明的技术内容,特举具体实施例并配合所附图式说明如下。
在本发明中参照附图来描述本发明的各方面,附图中示出了许多说明性实施例。本发明的实施例不局限于附图所示。应当理解,本发明通过上面介绍的多种构思和实施例,以及下面详细描述的构思和实施方式中的任意一种来实现,这是因为本发明所公开的构思和实施例并不限于任何实施方式。另外,本发明公开的一些方面可以单独使用,或者与本发明公开的其他方面的任何适当组合来使用。
参照图1,本发明提出一种基于多维度影响因素的道路安全评价方法,能够结合考虑宏观与微观道路安全分析模型的基础上准确判断各影响因素对道路事故产生的影响 分别针对限定区域范围内的各个子区域,通过步骤A至步骤D构建安全评价模型,应用安全评价模型,通过以下步骤E至步骤F,获得影响子区域中各交通道路安全的影响因素,对子区域进行安全评价:
宏观与微观维度下选取研究单位,宏观维度下的研究单位确定为交通分析小区,微观维度下的研究单位确定为交通分析小区内的各个研究路段。
步骤A、针对交通分析小区,周期获得交通分析小区在预设时长内的历史交通数据、以及交通分析小区内各个交通道路分别在预设时长内的历史交通数据,各交通分析小区所对应的历史交通数据分别均包括:交通分析小区的人口密度N、交通分析小区的GDP、交通分析小区内道路网密度K、交通分析小区的机动车年平均日交通量AADT1、交通分析小区绿化面积占比L1、交通分析小区居住区占比L2、交通分析小区非居住区占比L3、交通分析小区道路面积占比L4、以及交通分析小区内的平均行车速度V,交通分析小区所对应的历史样本数据如表1所示:
表1交通小区样本数据统计表
样本编号 E1 N GDP K L1 L2 L3 L4 V AADT
b 1 E1 1 N 1 GDP 1 K 1 L1 1 L2 1 L3 1 L4 1 V 1 AADT 1
b 10 E1 10 N 10 GDP 10 K 10 L1 10 L2 10 L3 10 L4 10 V 10 AADT 10
b 200 E1 200 N 200 GDP 200 K 200 L1 200 L2 200 L3 200 L4 200 V 200 AADT 200
交通分析小区内的各个交通道路所对应的历史交通数据分别均包括:交通道路长度D、交通道路车道数J、交通道路宽度W、交通道路是否设置有专用车道Q、交通道路的机动车年平均日交通量AADT2、交通道路的交叉口密度A、以及交通道路等级D,针对单个交通分析小区,其内部所含各交通道路的历史交通数据如表2所示:
表2各路段样本数据统计表
样本编号 E2 T J W Q AADT2 A D
A 1 E2 1 T 1 J 1 W 1 Q 1 AADT2 1 A 1 D 1
A 10 E2 10 T 10 J 10 W 10 Q 10 AADT2 10 A 10 D 10
A 200 E2 200 T 200 J 200 W 200 Q 200 AADT2 200 A 200 D 200
选取交通小区b1作为本发明实施例的实例,随后进入步骤B。
步骤B、基于子区域b1在预设时长内的历史交通数据、以及子区域内各个交通道路分别在预设时长内的历史交通数据,获得子区域对应的安全风险曝光量、以及子区域所含各个交通道路分别所对应的各个安全风险曝光量,并对各安全风险曝光量进行量化,得到各安全风险曝光量分别所对应的各个分类变量T,基于中位数对道路安全风险曝光量进行分类,低于中位数的称之为低密度机动车日交通量,高于中位数的称之为高密度机动车日交通量;同时基于分类的风险曝光量赋予各研究单位分类变量T,处于高密度机动车日交通量的研究单位T=1,反之T=0,,针对该子区域分别所对应的各个交通道路,根据以下公式:
Figure PCTCN2022103535-appb-000007
得到该子区域b1、以及对应各个交通道路的风险曝光量分别所对应的各个分类变量T,其中,AADT i为AADT1或AADT2,当AADT i=AADT1时,AADT i′为限定区域范围内所有子区域的机动车年平均日交通量的中位数,当AADT i=AADT2时,AADT i′为子区域内所有交通道路的机动车年平均日交通量的中位数,随后进入步骤C。
步骤C、分别针对该子区域b1所包含的各个交通道路,基于其所对应的各历史交 通数据、以及步骤B中所获的各个分类变量T,构建道路安全量化子模型,即获得该子区域中各个交通道路分别所对应的道路安全量化子模型,以子区域b1中三条路段A1-A3为示例,其分别所对应的道路安全量化子模型分别为:
lnE2 1=θ 1T+θ 2J 13W 14Q 15AADT2 16A 17D 12
lnE2 2=θ 1T+θ 2J 23W 24Q 25AADT2 26A 27D 22
lnE2 3=θ 1T+θ 2J 33W 34Q 35AADT2 36A 37D 32
当交通道路设置有专用车道时Q=1,当交通道路设置有专用车道时Q=0,当道路等级为主干道时D=1,当道路等级为次干道时D=2,当道路等级为支路时D=3,其中,
Figure PCTCN2022103535-appb-000008
此时,AADT i′为子区域内所有交通道路的机动车年平均日交通量的中位数;
基于该子区域b1中各个交通道路分别所对应的道路安全量化子模型、以及该子区域的历史交通数据,构建该子区域所对应的区域安全量化子模型为
Figure PCTCN2022103535-appb-000009
Figure PCTCN2022103535-appb-000010
Figure PCTCN2022103535-appb-000011
其中,
Figure PCTCN2022103535-appb-000012
Figure PCTCN2022103535-appb-000013
此时,AADT i′为限定区域范围内所有子区域的机动车年平均日交通量的中位数,交通小区b1所对应的区域安全量化子模型为:
lnE1 1=β 1N 12GDP 13K 14AADT1 15L1 16L2 17L3 18L4 19V 11
其中,lnE1 1=lnE2 1+lnE2 2+lnE2 3,随后进入步骤D。
步骤D、针对子区域,以该子区域所对应的区域安全量化子模型和该子区域中各个交通道路分别所对应的道路安全量化子模型构成的模型组作为该子区域对应的安全评价模型,并且模型组中各个子模型的输入量为其所对应的历史交通数据;
步骤E、按照步骤A至步骤C中方法,基于子区域的实际交通数据、以及子区域内各个交通道路的实际交通数据,获得子区域所对应的区域安全量化子模型、以及各个道路安全量化子模型,随后进入步骤F;
步骤F、针对该子区域,按照步骤D中方法应用安全评价模型,以约束函数为目标,对子区域对应的区域安全量化子模型、以及各个道路安全量化子模型进行求解,得到影响子区域道路安全的影响因素,根据影响因素对子区域以及该子区域内各个交通道路进 行安全评价。
在约束条件下,可分别判断各影响因素在不同维度下对道路安全的影响机理,如果影响因素的系数在95%置信区间正向显著,则说明该影响因素会增加交通小区或路段上事故的发生,如果影响因素的系数在95%置信区间负向显著,则说明该影响因素会减少交通小区或路段上事故的发生。
本次发明的实验验证是在假设数据条件下进行的,以交通小区的因素N为例,如果在95%置信区间下β 1>0,则说明交通小区内的人口密度与道路事故的产生是正向相关,人口密度越大,交通小区内事故发生越多,如果在95%置信区间下β 1<0,则说明交通小区内的人口密度与道路事故的产生是负向相关,人口密度越大,交通小区内事故发生越少。
虽然本发明已以较佳实施例揭露如上,然其并非用以限定本发明。本发明所属技术领域中具有通常知识者,在不脱离本发明的精神和范围内,当可作各种的更动与润饰。因此,本发明的保护范围当视权利要求书所界定者为准。

Claims (7)

  1. 一种基于多维度影响因素的道路安全评价方法,其特征在于,分别针对限定区域范围内的各个子区域,通过步骤A至步骤D构建安全评价模型,应用安全评价模型,通过以下步骤E至步骤F,获得影响子区域中各交通道路安全的影响因素,对子区域进行安全评价:
    步骤A、针对子区域,周期获得子区域在预设时长内的历史交通数据、以及子区域内各个交通道路分别在预设时长内的历史交通数据,随后进入步骤B;
    步骤B、基于子区域在预设时长内的历史交通数据、以及子区域内各个交通道路分别在预设时长内的历史交通数据,获得子区域对应的安全风险曝光量、以及子区域所含各个交通道路分别所对应的各个安全风险曝光量,并对各安全风险曝光量进行量化,得到各安全风险曝光量分别所对应的各个分类变量T,随后进入步骤C;
    步骤C、分别针对该子区域所包含的各个交通道路,基于其所对应的各历史交通数据、以及步骤B中所获的各个分类变量T,构建道路安全量化子模型,即获得该子区域中各个交通道路分别所对应的道路安全量化子模型;
    基于该子区域中各个交通道路分别所对应的道路安全量化子模型、以及该子区域的历史交通数据,构建该子区域所对应的区域安全量化子模型,随后进入步骤D;
    步骤D、针对子区域,以该子区域所对应的区域安全量化子模型和该子区域中各个交通道路分别所对应的道路安全量化子模型构成的模型组作为该子区域对应的安全评价模型,并且模型组中各个子模型的输入量为其所对应的历史交通数据;
    步骤E、按照步骤A至步骤C中方法,基于子区域的实际交通数据、以及子区域内各个交通道路的实际交通数据,获得子区域所对应的区域安全量化子模型、以及各个道路安全量化子模型,随后进入步骤F;
    步骤F、针对该子区域,按照步骤D中方法应用安全评价模型,以约束函数为目标,对子区域对应的区域安全量化子模型、以及各个道路安全量化子模型进行求解,得到影响子区域道路安全的影响因素,根据影响因素对子区域以及该子区域内各个交通道路进行安全评价。
  2. 根据权利要求1所述的一种基于多维度影响因素的道路安全评价方法,其特征在于,周期获得限定区域范围内各个子区域在预设时长的历史交通数据,各子区域所对应的历史交通数据分别均包括:子区域的人口密度N、子区域的GDP、子区域内道路网密度K、子区域的机动车年平均日交通量AADT1、子区域绿化面积占比L1、子区域居住区占比L2、子区域非居住区占比L3、子区域道路面积占比L4、以及子区域内的平 均行车速度V;
    各子区域内的各个交通道路所对应的历史交通数据分别均包括:交通道路长度D、交通道路车道数J、交通道路宽度W、交通道路是否设置有专用车道Q、交通道路的机动车年平均日交通量AADT2、交通道路的交叉口密度A、以及交通道路等级D。
  3. 根据权利要求2所述的一种基于多维度影响因素的道路安全评价方法,其特征在于,所述步骤B中,基于子区域在预设时长内的历史交通数据、以及子区域内各个交通道路分别在预设时长内的历史交通数据,针对该子区域分别所对应的各个交通道路,根据以下公式:
    Figure PCTCN2022103535-appb-100001
    得到该子区域、以及对应各个交通道路的风险曝光量分别所对应的各个分类变量T,其中,AADT i为AADT1或AADT2,当AADT i=AADT1时,AADT i′为限定区域范围内所有子区域的机动车年平均日交通量的中位数,当AADT i=AADT2时,AADT i′为子区域内所有交通道路的机动车年平均日交通量的中位数。
  4. 根据权利要求3所述的一种基于多维度影响因素的道路安全评价方法,其特征在于,所述步骤C中,分别针对子区域所包含的各个交通道路,根据以下公式:
    Figure PCTCN2022103535-appb-100002
    获得各个交通道路分别所对应的各个道路安全量化子模型lnE2 n,其中,E2为交通道路在预设时间周期内的事故发生量,θ为交通道路的安全影响系数,ε n为道路安全量化子模型的误差项,n的取值范围为1至N,N为各个子区域中分别所包含的交通道路的总数,当交通道路设置有专用车道时Q=1,当交通道路设置有专用车道时Q=0,当道路等级为主干道时D=1,当道路等级为次干道时D=2,当道路等级为支路时D=3,其中,
    Figure PCTCN2022103535-appb-100003
    此时,AADT i′为子区域内所有交通道路的机动车年平均日交通量的中位数;
    分别针对限定区域范围内各个子区域,根据以下公式:
    Figure PCTCN2022103535-appb-100004
    获得限定区域范围内各个子区域所对应的各个区域安全量化子模型lnE1 m,其中,E1为子区域在预设时间周期内的事故发生量,β为子区域的安全影响系数,ε m为区域安全量化子模型的误差项,m的的取值范围为1至M,M为限定区域范围所包含各个子区域的总数,其中,
    Figure PCTCN2022103535-appb-100005
    此时,AADT i′为限定区域范围内所有子区域的机动车年平均日交通量的中位数。
  5. 根据权利要求4所述的一种基于多维度影响因素的道路安全评价方法,其特征在于,所述步骤F中约束函数如下:
    Figure PCTCN2022103535-appb-100006
    以该约束函数为训练目标,对安全评价模型进行训练,在约束条件下对子区域对应的区域安全量化子模型、以及各个道路安全量化子模型中的安全影响系数进行求解,获得安全影响系数在95%置信区间内的显著程度,当安全影响系数在95%置信区间内正向显著,则安全影响系数对应的交通数据会增加交通道路上交通事故的发生率,当安全影响系数在95%置信区间内负向显著,则安全影响系数对应的交通数据会降低交通道路上交通事故的发生率。
  6. 一种基于多维度影响因素的道路安全评价系统,其特征在于,包括:
    一个或多个处理器;
    存储器,存储可被执行的指令,所述指令在通过一个或多个处理器执行时,一个或多个处理器执行包括权利要求1-5中任意一项所述道路安全评价方法的过程。
  7. 一种存储软件的计算机可读取介质,其特征在于,所述软件包括能通过一个或多个计算机执行的指令,所述指令在被所述一个或多个计算机执行时,执行如权利要求1-5中任意一项所述道路安全评价方法的操作。
PCT/CN2022/103535 2021-12-03 2022-07-04 一种基于多维度影响因素的道路安全评价方法及系统 WO2023098076A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/925,748 US11887472B2 (en) 2021-12-03 2022-07-04 Method and system for evaluating road safety based on multi-dimensional influencing factors

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111465931.5 2021-12-03
CN202111465931.5A CN114241753B (zh) 2021-12-03 2021-12-03 一种基于多维度影响因素的道路安全评价方法及系统

Publications (1)

Publication Number Publication Date
WO2023098076A1 true WO2023098076A1 (zh) 2023-06-08

Family

ID=80752912

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/103535 WO2023098076A1 (zh) 2021-12-03 2022-07-04 一种基于多维度影响因素的道路安全评价方法及系统

Country Status (3)

Country Link
US (1) US11887472B2 (zh)
CN (1) CN114241753B (zh)
WO (1) WO2023098076A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116721550A (zh) * 2023-08-10 2023-09-08 济南高品伟业信息科技有限公司 一种基于网格计算的交通管理系统
CN117034435A (zh) * 2023-10-09 2023-11-10 中国公路工程咨询集团有限公司 一种基于bim技术的公路安全评价方法及系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114241753B (zh) * 2021-12-03 2022-11-01 东南大学 一种基于多维度影响因素的道路安全评价方法及系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016122881A1 (en) * 2015-01-28 2016-08-04 Allstate Insurance Company Road segment safety rating
CN107909247A (zh) * 2017-10-26 2018-04-13 东南大学 一种基于空间层次贝叶斯模型的城市宏观道路交通安全影响因素分析方法
CN108428340A (zh) * 2018-05-11 2018-08-21 深圳市图灵奇点智能科技有限公司 道路交通状况分析方法和系统
CN111738474A (zh) * 2019-03-25 2020-10-02 京东数字科技控股有限公司 交通状态预测方法和装置
CN112037513A (zh) * 2020-09-01 2020-12-04 清华大学 一种实时交通安全指数动态综合评价系统及其构建方法
CN112950946A (zh) * 2021-02-26 2021-06-11 中冶南方城市建设工程技术有限公司 一种市政道路施工期交通拥堵原因判别方法
CN113487863A (zh) * 2021-06-30 2021-10-08 中国科学技术大学先进技术研究院 一种多因素交通风险实时定量评估系统及方法
CN114241753A (zh) * 2021-12-03 2022-03-25 东南大学 一种基于多维度影响因素的道路安全评价方法及系统

Family Cites Families (30)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5798949A (en) * 1995-01-13 1998-08-25 Kaub; Alan Richard Traffic safety prediction model
US7912628B2 (en) * 2006-03-03 2011-03-22 Inrix, Inc. Determining road traffic conditions using data from multiple data sources
US9932033B2 (en) * 2007-05-10 2018-04-03 Allstate Insurance Company Route risk mitigation
US10157422B2 (en) * 2007-05-10 2018-12-18 Allstate Insurance Company Road segment safety rating
US20140074402A1 (en) * 2012-09-12 2014-03-13 Lexisnexis Risk Solutions Fl Inc. Systems and methods for determining risks associated with driving routes
US10262373B2 (en) * 2013-06-07 2019-04-16 State Farm Mutual Automobile Insurance Company Systems and methods for grid-based insurance rating
WO2016018936A1 (en) * 2014-07-28 2016-02-04 Econolite Group, Inc. Self-configuring traffic signal controller
US10024684B2 (en) * 2014-12-02 2018-07-17 Operr Technologies, Inc. Method and system for avoidance of accidents
US9747506B2 (en) * 2015-10-21 2017-08-29 Ford Global Technologies, Llc Perception-based speed limit estimation and learning
CA3014660C (en) * 2016-02-15 2021-08-17 Allstate Insurance Company Early notification of non-autonomous area
CN106530706B (zh) * 2016-12-02 2019-02-22 清华大学 一种基于连通性空间权重矩阵的区域交通安全评价方法
US10490078B1 (en) * 2017-07-20 2019-11-26 State Farm Mutual Automobile Insurance Company Technology for providing real-time route safety and risk feedback
US11183081B2 (en) * 2017-08-18 2021-11-23 Tourmaline Labs, Inc. System and methods for relative driver scoring using contextual analytics
CN109118773B (zh) * 2018-09-30 2019-10-01 中交第一公路勘察设计研究院有限公司 一种高速公路交通事故风险评估方法
CN109448369B (zh) * 2018-10-26 2021-08-03 中交第一公路勘察设计研究院有限公司 高速公路实时运行风险计算方法
CN109598929B (zh) * 2018-11-26 2020-06-16 北京交通大学 一种多类别交通事故起数预测方法
CN109714421B (zh) * 2018-12-28 2021-08-03 国汽(北京)智能网联汽车研究院有限公司 基于车路协同的智能网联汽车运行系统
CN115380196A (zh) * 2019-11-11 2022-11-22 御眼视觉技术有限公司 用于确定道路安全性的系统和方法
FR3104306B1 (fr) * 2019-12-09 2022-02-18 Ifp Energies Now Procédé de détermination d’émissions polluantes et/ou sonores et/ou de paramètres de sécurité routière sur une portion de réseau routier
US20210233395A1 (en) * 2020-01-23 2021-07-29 Lg Electronics Inc. Differentially private solution for traffic monitoring
CN111553517A (zh) * 2020-04-17 2020-08-18 平安科技(深圳)有限公司 道路优化方法、系统、终端及计算机可读存储介质
CN112100857B (zh) * 2020-09-17 2022-04-12 燕山大学 一种分心驾驶行为的风险评估方法
CN112447048A (zh) * 2020-11-19 2021-03-05 长安大学 一种基于模糊综合评价的城市道路风险分级系统及方法
US20220258739A1 (en) * 2021-02-17 2022-08-18 Qualcomm Incorporated Method and System for Generating a Confidence Value in a Position Overlap Check Using Vehicle Threshold Models
CN113313357A (zh) * 2021-04-30 2021-08-27 东南大学 一种基于高斯过程回归分析的交通道路安全评价方法
AU2021104800A4 (en) * 2021-08-02 2021-09-30 Wuhan University Urban Compactness Measurement, Evaluation and Optimization Method Based on Traffic Safety
US20230033672A1 (en) * 2021-08-02 2023-02-02 Here Global B.V. Determining traffic violation hotspots
US20230093601A1 (en) * 2021-09-14 2023-03-23 Motional Ad Llc Environmental limitation and sensor anomaly system and method
US20230204378A1 (en) * 2021-12-27 2023-06-29 Here Global B.V. Detecting and monitoring dangerous driving conditions
US20230245564A1 (en) * 2022-01-31 2023-08-03 Nissan North America, Inc. System and Method for Intersection Collision Avoidance

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016122881A1 (en) * 2015-01-28 2016-08-04 Allstate Insurance Company Road segment safety rating
CN107909247A (zh) * 2017-10-26 2018-04-13 东南大学 一种基于空间层次贝叶斯模型的城市宏观道路交通安全影响因素分析方法
CN108428340A (zh) * 2018-05-11 2018-08-21 深圳市图灵奇点智能科技有限公司 道路交通状况分析方法和系统
CN111738474A (zh) * 2019-03-25 2020-10-02 京东数字科技控股有限公司 交通状态预测方法和装置
CN112037513A (zh) * 2020-09-01 2020-12-04 清华大学 一种实时交通安全指数动态综合评价系统及其构建方法
CN112950946A (zh) * 2021-02-26 2021-06-11 中冶南方城市建设工程技术有限公司 一种市政道路施工期交通拥堵原因判别方法
CN113487863A (zh) * 2021-06-30 2021-10-08 中国科学技术大学先进技术研究院 一种多因素交通风险实时定量评估系统及方法
CN114241753A (zh) * 2021-12-03 2022-03-25 东南大学 一种基于多维度影响因素的道路安全评价方法及系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
WANG, XUSONG ET AL.: "Macro Level Safety Modeling and Impact Factor Analysis", JOURNAL OF TONGJI UNIVERSITY (NATURAL SCIENCE), vol. 40, no. 11, 15 November 2012 (2012-11-15), pages 1627 - 1633, ISSN: 0253-374x *
YANG, SHAOWEI ET AL.: "Review of Operation Analysis Methods of Expressway Weaving Area", JOURNAL OF HIGHWAY AND TRANSPORTATION RESEARCH AND DEVELOPMENT, vol. 35, no. 10, 15 October 2018 (2018-10-15), XP009546834, ISSN: 1002-0268 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116721550A (zh) * 2023-08-10 2023-09-08 济南高品伟业信息科技有限公司 一种基于网格计算的交通管理系统
CN116721550B (zh) * 2023-08-10 2023-10-24 济南高品伟业信息科技有限公司 一种基于网格计算的交通管理系统
CN117034435A (zh) * 2023-10-09 2023-11-10 中国公路工程咨询集团有限公司 一种基于bim技术的公路安全评价方法及系统

Also Published As

Publication number Publication date
US11887472B2 (en) 2024-01-30
CN114241753A (zh) 2022-03-25
US20230215270A1 (en) 2023-07-06
CN114241753B (zh) 2022-11-01

Similar Documents

Publication Publication Date Title
WO2023098076A1 (zh) 一种基于多维度影响因素的道路安全评价方法及系统
CN109035778B (zh) 拥堵成因分析方法、装置、计算机设备及存储介质
Lee et al. Quantifying effects of ramp metering on freeway safety
CN106127586A (zh) 大数据时代下车险费率辅助决策系统
CN109523786A (zh) 一种高速公路交通事故成因分析方法
CN110288825B (zh) 基于多源数据融合和snmf的交通控制子区聚类划分方法
CN110288826B (zh) 基于多源数据融合和milp的交通控制子区聚类划分方法
Chandra et al. Effect of traffic composition on capacity of multilane highways
Vasconcelos et al. Estimating the parameters of Cowan’s M3 headway distribution for roundabout capacity analyses
Chengye et al. Modelling motorway accidents using negative binomial regression
Chand et al. Application of Fractal theory for crash rate prediction: Insights from random parameters and latent class tobit models
Liu et al. Predicting real-time crash risk for urban expressways in China
Lyon et al. Empirical investigation of interactive highway safety design model accident prediction algorithm: Rural intersections
CN113313357A (zh) 一种基于高斯过程回归分析的交通道路安全评价方法
Budhkar et al. Multiple-leader vehicle-following behavior in heterogeneous weak lane discipline traffic
Kidando et al. An application of Bayesian multilevel model to evaluate variations in stochastic and dynamic transition of traffic conditions
CN113570852B (zh) 基于安全替代评价指标的高风险路段风险防控手段评估的方法
Kushchenko et al. Analysis of congestion occurrence cycles
Milligan et al. Road safety performance measures and AADT uncertainty from short-term counts
Gordon et al. Analysis of crash rates and surrogate events: unified approach
Zhang Research on risk projection of urban road traffic safety based on RS-SVM combination model.
Mirzahossein et al. Evaluating geometric design characteristics of two-lane two-way rural roads on crash frequencies occurrences
Pallela et al. Examining the lane-wise time headway and speed characteristics at curb-side bus stop on four-lane divided urban arterials
Wenfang et al. Traffic accident analysis with or without bus priority
Mousa et al. Modeling Speed-Density Relation for Highways in Developing Countries with No Lane Discipline: A Case Study in Egypt

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22899896

Country of ref document: EP

Kind code of ref document: A1