TW202403537A - Apparatus and method for analyzing traffic status and computer program product implementing the method - Google Patents
Apparatus and method for analyzing traffic status and computer program product implementing the method Download PDFInfo
- Publication number
- TW202403537A TW202403537A TW111126500A TW111126500A TW202403537A TW 202403537 A TW202403537 A TW 202403537A TW 111126500 A TW111126500 A TW 111126500A TW 111126500 A TW111126500 A TW 111126500A TW 202403537 A TW202403537 A TW 202403537A
- Authority
- TW
- Taiwan
- Prior art keywords
- speed
- sample size
- road section
- data
- sample
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000004590 computer program Methods 0.000 title claims abstract description 12
- 230000011664 signaling Effects 0.000 claims abstract description 79
- 238000001914 filtration Methods 0.000 claims abstract description 53
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 230000002776 aggregation Effects 0.000 claims abstract description 16
- 238000004220 aggregation Methods 0.000 claims abstract description 16
- 238000001514 detection method Methods 0.000 claims description 22
- 238000004364 calculation method Methods 0.000 claims description 21
- 238000000513 principal component analysis Methods 0.000 claims description 14
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 239000000523 sample Substances 0.000 description 120
- 238000005516 engineering process Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000005520 cutting process Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000009825 accumulation Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012847 principal component analysis method Methods 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
本案係關於一種交通路況分析技術,詳言之,係關於利用行動信令資料分析壅塞路況之設備、方法及電腦程式產品。 This case is about a traffic condition analysis technology. Specifically, it is about equipment, methods and computer program products that use mobile signaling data to analyze congested road conditions.
近年來,以行動裝置基地站為基礎之車輛探偵(Cellular-Based Vehicle Probe;CVP)技術,因其涵蓋率廣和成本低等特色,廣泛使用於道路交通資訊之偵測上。一般而言,CVP技術主要是應用手機信令,即藉由CVP信令資料之時間差和位置差,來計算路段上之平均車速及旅行時間。 In recent years, Cellular-Based Vehicle Probe (CVP) technology based on mobile device base stations has been widely used in the detection of road traffic information due to its wide coverage and low cost. Generally speaking, CVP technology mainly uses mobile phone signaling, that is, using the time difference and location difference of CVP signaling data to calculate the average vehicle speed and travel time on the road section.
然而,由於行動信令路況分析技術是透過行動信令點位的時間與位置來分析路況,故會受到信令飄移和運具不同之影響。例如,信令飄移往往受限於基地站定位技術限制而難以避免,惟此問題卻會影響路況分析的正確性,尤其在壅塞時,信令的位置將更容易飄移不穩定,造成路況分析不準確。另一方面,運具通常需使用異質資料進行勾稽,如大眾運具班表、票證資訊、影像辨識資訊等,徒增系統或運算上的複雜度。 However, since mobile signaling traffic analysis technology analyzes traffic conditions through the time and location of mobile signaling points, it will be affected by signaling drift and differences in vehicles. For example, signaling drift is often limited by base station positioning technology and is unavoidable. However, this problem will affect the accuracy of traffic analysis. Especially during congestion, the signaling position will be more likely to drift and become unstable, resulting in inaccurate traffic analysis. Accurate. On the other hand, transport vehicles usually need to use heterogeneous data for verification, such as mass transport vehicle schedules, ticket information, image recognition information, etc., which only increases system or computational complexity.
因此,如何利用行動信令資料來快速且準確地分析壅塞路況,為目前待解決之議題。 Therefore, how to use mobile signaling data to quickly and accurately analyze traffic jams is an issue that needs to be solved.
為解決上述問題及其他問題,本案揭示一種分析路段狀況之設備、方法、電腦程式產品及電腦可讀取記錄媒體。 In order to solve the above problems and other problems, this case discloses a device, method, computer program product and computer-readable recording medium for analyzing road section conditions.
本案所揭之分析路段狀況之設備係包括:資料接收模組,用於接收複數個行動裝置的複數個行動信令資料;批次過濾模組,對該複數個行動裝置的該複數個行動信令資料進行批次分析過濾,以產生複數個行動信令位置資料;路段對應模組,將該複數個行動信令位置資料對應至一路段,以計算該路段的總樣本量、複數個區間樣本量、及各該區間樣本量所對應的時速;以及動態壅塞門檻計算模組,係根據該路段的歷史資料,計算該路段的壅塞門檻值;動態壅塞偵測模組,係根據該路段的該壅塞門檻值與該路段的該總樣本量,決定該總樣本量中被選取的樣本群;以及時速聚合模組,對該選取的樣本群中各該區間樣本量及其對應的時速進行時速聚合,俾產生該路段的即時時速資訊。 The equipment disclosed in this case for analyzing road section conditions includes: a data receiving module, which is used to receive a plurality of mobile signaling data from a plurality of mobile devices; a batch filtering module, which is used to receive the plurality of mobile signaling data from a plurality of mobile devices. The data is subjected to batch analysis and filtering to generate a plurality of mobile signaling location data; the road segment corresponding module maps the plurality of mobile signaling location data to a segment to calculate the total sample size and multiple interval samples of the road segment. and the speed corresponding to the sample size of each section; and the dynamic congestion threshold calculation module calculates the congestion threshold of the road section based on the historical data of the road section; the dynamic congestion detection module calculates the congestion threshold of the road section based on the The congestion threshold and the total sample size of the road section determine the sample group selected from the total sample size; and the speed aggregation module performs speed aggregation of the sample size and corresponding speed of each section in the selected sample group. , in order to generate real-time speed information for this road section.
於一實施例中,本案之批次過濾模組係依時間順序批次切割該複數個行動信令資料,以對各批次的行動信令資料進行主成分分析、及方向過濾和距離過濾,使得各該批次的該行動信令資料經主成分分析、及方向過濾和距離過濾之後,產生各批次的行動信令位置資料,以構成該複數個行動信令位置資料。 In one embodiment, the batch filtering module of this case cuts the plurality of mobile signaling data in batches in chronological order to perform principal component analysis, direction filtering and distance filtering on each batch of mobile signaling data. After each batch of the mobile signaling data is subjected to principal component analysis, direction filtering and distance filtering, mobile signaling position data of each batch is generated to form the plurality of mobile signaling position data.
於一實施例中,本案之動態壅塞偵測模組係比較該總樣本量與該壅塞門檻值,以於該總樣本量未大於該壅塞門檻值時選取時速較快的樣本群,而於該總樣本量大於該壅塞門檻值時,判斷該路段的低速占比是否大於根據該路段的該歷史資料所計算出的該路段的低速占比門檻值,以於該低速占比未大於該低速占比門檻值時選取時速較快的樣本群,而於該低速占比大於該低速占比門檻值時選取數量較多的樣本群。 In one embodiment, the dynamic congestion detection module of this case compares the total sample size with the congestion threshold, so as to select a sample group with a faster speed when the total sample size is not greater than the congestion threshold, and when the total sample size is not greater than the congestion threshold, When the total sample size is greater than the congestion threshold, it is determined whether the low-speed proportion of the road section is greater than the low-speed proportion threshold of the road section calculated based on the historical data of the road section, so that when the low-speed proportion is not greater than the low-speed proportion When the low-speed ratio is greater than the threshold, a sample group with a faster speed is selected, and when the low-speed ratio is greater than the low-speed ratio threshold, a larger sample group is selected.
於一實施例中,本案之時速聚合模組係對該選取的樣本群中各區間的時速低值與時速機率值進行加權平均,藉以產生該路段的即時時速資訊。 In one embodiment, the speed aggregation module of this case performs a weighted average of the low speed values and speed probability values of each section in the selected sample group to generate real-time speed information for the road section.
本案所揭之分析路段狀況之方法係包括:對所接收之複數個行動裝置的複數個行動信令資料執行批次分析過濾,以產生複數個行動信令位置資料;將該複數個行動信令位置資料對應至一路段,以計算該路段的總樣本量、複數個區間樣本量、及各該區間樣本量所對應的時速;根據該路段的該總樣本量與根據該路段的歷史資料所計算出之該路段的壅塞門檻值,決定該總樣本量中被選取的樣本群;以及對該選取的樣本群中各區間樣本量及其對應的時速進行時速聚合,俾產生該路段的即時時速資訊。 The method disclosed in this case for analyzing road segment conditions includes: performing batch analysis and filtering on a plurality of mobile signaling data received from a plurality of mobile devices to generate a plurality of mobile signaling location data; The location data is mapped to a road segment to calculate the total sample size of the road segment, the sample sizes of multiple intervals, and the speed corresponding to the sample size of each section; calculated based on the total sample size of the road segment and the historical data of the road segment Based on the congestion threshold of the road section, the sample group selected from the total sample size is determined; and the sample size and corresponding speed of each section in the selected sample group are aggregated to generate real-time speed information of the road section. .
於一實施例中,本案所述對所接收之複數個行動裝置的複數個行動信令資料進行批次分析過濾以產生複數個行動信令位置資料之步驟係包括:依時間順序批次切割該複數個行動信令資料;對各批次的行動信令資料進行主成分分析、及方向過濾和距離過濾;以及各該批次的該行動信令資料經主成分分析、及方向過濾和距離過濾之後,產生各批次的行動信令位置資料,以構成該複數個行動信令位置資料。 In one embodiment, the step described in this case of batch analyzing and filtering a plurality of mobile signaling data received from a plurality of mobile devices to generate a plurality of mobile signaling location data includes: cutting the mobile signaling data in batches in chronological order. A plurality of mobile signaling data; performing principal component analysis, direction filtering and distance filtering on each batch of mobile signaling data; and performing principal component analysis, direction filtering and distance filtering on each batch of mobile signaling data Then, each batch of mobile signaling location data is generated to constitute the plurality of mobile signaling location data.
於一實施例中,本案所述根據該路段的該總樣本量與根據該路段的歷史資料所計算出之該路段的壅塞門檻值,決定該總樣本量中被選取的樣本群之步驟係包括:比較該總樣本量與該壅塞門檻值,其中,於該總樣本量未大於該壅塞門檻值時,選取時速較快的樣本群,而於該總樣本量大於該壅塞門檻值時,判斷該路段的低速占比是否大於根據該路段的該歷史資料所計算出之該路段的低速占比門檻值,其中,於該低速占比未大於該低速占比門檻值時,選取時速較快的樣本群,而於該低速占比大於該低速占比門檻值時,選取數量較多的樣本群。 In one embodiment, the step of determining the sample group selected from the total sample size based on the total sample size of the road segment and the congestion threshold value of the road segment calculated based on the historical data of the road segment includes : Compare the total sample size with the congestion threshold, where, when the total sample size is not greater than the congestion threshold, select the faster sample group, and when the total sample size is greater than the congestion threshold, determine the Whether the low-speed ratio of the road section is greater than the low-speed ratio threshold of the road section calculated based on the historical data of the road section. When the low-speed ratio is not greater than the low-speed ratio threshold, samples with faster speeds are selected. group, and when the low-speed proportion is greater than the low-speed proportion threshold, a larger sample group is selected.
於一實施例中,本案所述選取時速較快的樣本群之步驟係包括:依時速快到慢計算前N個區間樣本量/該總樣本量以及N/總區間數量;以及於該前N個區間樣本量/該總樣本量係大於或等於該N/總區間數量,選取具有該前N個區間樣本量的群為時速較快的樣本群。 In one embodiment, the step of selecting a sample group with a faster speed described in this case includes: calculating the sample size of the first N intervals/the total sample size and N/the total number of intervals in order from fast to slow; and in the first N The sample size of intervals/the total sample size is greater than or equal to N/the total number of intervals, and the group with the sample size of the first N intervals is selected as the faster sample group.
於一實施例中,本案所述選取數量較多的樣本群之步驟係包括:以相鄰M個區間分為一群,以計算各群的樣本量加總量;以及選取具有最多該樣本加總量的群為該數量較多的樣本群。 In one embodiment, the steps of selecting a large number of sample groups described in this case include: dividing M adjacent intervals into one group to calculate the total sample size of each group; and selecting the sample group with the largest total number. The quantitative group is the sample group with a larger number.
於一實施例中,本案所述對該選取的樣本群中各區間樣本量及其對應的時速進行時速聚合以產生該路段的時速資訊之步驟係包括:對該選取的樣本群中各區間的時速低值與時速機率值進行加權平均,以產生該路段的即時時速資訊。 In one embodiment, the step described in this case of aggregating the sample size of each section in the selected sample group and its corresponding speed to generate the speed information of the road section includes: The low speed value and the speed probability value are weighted and averaged to generate real-time speed information for the road section.
本案所揭之分析路段狀況之電腦程式產品,經電腦下載以執行上揭分析路段狀況之方法。 The computer program product disclosed in this case for analyzing road section conditions can be downloaded from a computer to execute the method disclosed above for analyzing road section conditions.
本案所揭之分析路段狀況之電腦可讀取記錄媒體,經電腦下載以執行上揭分析路段狀況之方法。 The computer for analyzing the condition of the road section disclosed in this case can read the recording medium and download it to the computer to execute the method of analyzing the condition of the road section disclosed above.
藉由本案所揭之分析路段狀況之設備、方法、電腦程式產品及電腦可讀取記錄媒體,執行批次分析過濾以濾除飄移點位,之後將行動信令位置資料對應至路段上,再執行壅塞門檻值計算及動態壅塞偵測技術,以追蹤時速隨時序的變化,避免與實際路況不符的情形,故無需額外勾稽異質資料,即可快速且準確地分析路況為壅塞或順暢,藉此提升路況分析準確性。 Through the equipment, methods, computer program products and computer-readable recording media disclosed in this case for analyzing road section conditions, batch analysis and filtering is performed to filter out drift points, and then the mobile signaling position data is mapped to the road section, and then Implement congestion threshold calculation and dynamic congestion detection technology to track changes in speed over time to avoid situations that are inconsistent with actual road conditions. Therefore, there is no need to additionally check heterogeneous data to quickly and accurately analyze whether the road conditions are congested or smooth. Improve the accuracy of traffic analysis.
10:基地站 10: Base station
2:設備 2: Equipment
21:資料接收模組 21: Data receiving module
22:批次過濾模組 22:Batch filter module
23:路段對應模組 23:Road section corresponding module
24:動態壅塞門檻計算模組 24:Dynamic congestion threshold calculation module
25:動態壅塞偵測模組 25:Dynamic congestion detection module
26:時速聚合模組 26: Speed aggregation module
30:資料庫 30:Database
P1~P12:點位 P1~P12: Points
S201~S206:步驟 S201~S206: steps
S301~S304:步驟 S301~S304: steps
S401~S406:步驟 S401~S406: steps
圖1係為根據本案之分析路段狀況之設備之實施例的架構示意圖。 Figure 1 is a schematic structural diagram of an embodiment of a device for analyzing road section conditions according to this case.
圖2係為根據本案之分析路段狀況之方法之實施例的流程示意圖。 Figure 2 is a schematic flow chart of an embodiment of a method for analyzing road section conditions according to this case.
圖3A係為本案之分析路段狀況之方法中批次分析過濾之實施例示意圖。 Figure 3A is a schematic diagram of an embodiment of batch analysis and filtering in the method of analyzing road section conditions in this case.
圖3B係為本案之分析路段狀況之方法中主成分分析之實施例示意圖。 Figure 3B is a schematic diagram of an embodiment of principal component analysis in the method of analyzing road section conditions in this case.
圖4係為本案之分析路段狀況之方法中取群之實施例示意圖。 Figure 4 is a schematic diagram of an embodiment of clustering in the method of analyzing road section conditions in this case.
圖5係為本案之分析路段狀況之方法中區間樣本量及其所對應的時速之實施例示意圖。 Figure 5 is a schematic diagram of an example of the interval sample size and the corresponding hourly speed in the method of analyzing road section conditions in this case.
圖6係為本案之分析路段狀況之方法中選取快群之實施例的示意圖。 Figure 6 is a schematic diagram of an embodiment of selecting a fast group in the method of analyzing road section conditions in this case.
圖7係為本案之分析路段狀況之方法中選取大群之實施例的示意圖。 Figure 7 is a schematic diagram of an embodiment of selecting large groups in the method of analyzing road section conditions in this case.
以下藉由特定的實施例說明本案之實施方式,熟習此項技藝之人士可由本文所揭示之內容輕易地瞭解本案之其他優點及功效。本說明書所附圖式所繪示之結構、比值、大小等均僅用於配合說明書所揭示之內容,以供熟悉此技藝之人士之瞭解與閱讀,非用於限定本案可實施之限定條件,故任何修飾、改變或調整,在不影響本案所能產生之功效及所能達成之目的下,均應仍落在本案所揭示之技術內容得能涵蓋之範圍內。 The following uses specific examples to illustrate the implementation of the present invention. People familiar with this art can easily understand other advantages and effects of the present invention from the content disclosed in this article. The structures, ratios, sizes, etc. shown in the drawings attached to this manual are only used to coordinate with the content disclosed in the manual for the understanding and reading of those familiar with this art. They are not used to limit the conditions for the implementation of this case. Therefore, any modification, change or adjustment, without affecting the effectiveness and purpose of this case, should still fall within the scope of the technical content disclosed in this case.
請參閱圖1,本案之設備2包括資料接收模組21、批次過濾模組22、路段對應模組23、動態壅塞門檻計算模組24、動態壅塞偵測模組25、以及時速聚合模組26。
Please refer to Figure 1. The
在一實施例中,設備2係為一遠端或雲端之伺服器。在另一實施例中,圖1中的各模組均可為軟體、硬體或韌體;若為硬體,則可為具有資料處理與運算能力之處理單元、處理器、電腦;若為軟體或韌體,則可包括處理單元、處理器、電腦可執行之指令。
In one embodiment, the
資料接收模組21自基地站10接收複數個行動裝置的複數個行動信令資料。具體而言,行動裝置可例如為具有SIM(用户身份模組;Subscriber Identity Module)卡之智慧手機、智慧手錶、平板電腦等,其與
基地站10之間傳遞有行動信令資料,包括例如行動裝置之位置、時間與速度等資料。於一具體實施例中,行動信令資料例如為CVP信令資料。
The
批次過濾模組22對資料接收模組21所接收之複數個行動裝置的複數個行動信令資料,進行批次、分析、過濾之處理,以產生複數個行動信令位置資料。詳言之,對於一個行動裝置,批次過濾模組22依時間順序累積複數個行動信令資料,累積至一定數量後切割為一批次,接著對該批次使用主成分分析(principal components analysis,PCA)法以分析該批點位之主成分,再進行方向過濾與距離過濾,進而將與主成分移動方向相反或與主成分垂直距離過遠的點位濾除。接著,過濾後之點位的其中一部分傳入路段對應模組23,另保留一部分作為下一批次的前幾個點位,以繼續對該下一批次使用主成分分析法及方向過濾與距離過濾,以此類推。並且,批次過濾模組22對各個行動裝置的複數個行動信令資料執行上述依時間順序累積、批次切割、主成分分析、方向過濾和距離過濾等處理,而各批次的行動信令資料經處理之後傳至路段對應模組23者即為所述複數個行動信令位置資料。
The
路段對應模組23將批次過濾模組22所產生之複數個行動信令位置資料對應之一路段,以計算出該路段的總樣本量、複數個區間樣本量及其所對應的時速。詳言之,根據複數個行動信令可計算出複數個用戶裝置的所在位置,結合路網資訊可獲得該路段上所有用戶裝置的數量(即,總樣本量),根據該路段的時速限制切割出複數個時速區間,而每個時速區間有其對應的用戶裝置數量(即,複數個區間樣本量及其所對應的時速)。須說明的是,由於批次過濾模組22不間斷地執行依時間順序累積、批次切
割、主成分分析、方向過濾和距離過濾等處理,爾後將複數個行動信令位置資料傳輸至路段對應模組23,因而路段對應模組23所計算出之該路段的總樣本量、複數個區間樣本量及其所對應的時速也是隨著時序而變化。
The road
動態壅塞門檻計算模組24根據資料庫30中該路段的歷史資料計算該路段的壅塞門檻值。具體言之,資料庫30儲存有路網資訊、道路容量資料、和歷史資料等,其中歷史資料例如該路段上同一時段歷史上(例如過去七天)的每隔一段時間(例如每五分鐘)的歷史總樣本量,動態壅塞門檻計算模組24選取總樣本量最大的約十個樣本量,平均之後的大約六成作為壅塞門檻值。另外,若無過去幾天的樣本資訊,則動態壅塞門檻計算模組24可以將資料庫30中該路段的道路容量設作為壅塞門檻值。另外,動態壅塞門檻計算模組24根據該路段的歷史資料計算該路段的低速占比門檻值,同樣可選取該路段上同一時段過去七天每五分鐘的平均時速的大約八成作為低速門檻,然後計算低於該低速門檻的樣本量相對於該歷史總樣本量的占比,作為低速占比門檻值。
The dynamic congestion
動態壅塞偵測模組25根據該路段的該壅塞門檻值與該路段的該總樣本量,或根據該路段的該低速占比門檻值與該路段的該低速占比,決定該總樣本量中被選取的樣本群。於一實施例中,動態壅塞偵測模組25比較該路段的該總樣本量與該路段的該壅塞門檻值,若路段對應模組23所計算出的該路段的總樣本量不大於(即小於或等於)動態壅塞門檻計算模組24所計算出之壅塞門檻值,則動態壅塞偵測模組25.判定為車少或順暢,因而動態壅塞偵測模組25決定自該總樣本量中選取時速較快的樣本群。於另一實施例中,若路段對應模組23所計算出的該路段的總樣本量大於動態
壅塞門檻計算模組24所計算出之壅塞門檻值,則動態壅塞偵測模組25判定為車多或壅塞,動態壅塞偵測模組25接著比較該路段的低速占比與該路段的低速占比門檻值。若路段對應模組23所計算出的該路段的低速占比不大於(即小於或等於)動態壅塞門檻計算模組24所計算出的該路段的低速占比門檻值,則可能車多但車速夠快,仍未達壅塞條件,因而動態壅塞偵測模組25決定自該總樣本量中選取時速較快的樣本群。若路段對應模組23所計算出的該路段的低速占比大於動態壅塞門檻計算模組24所計算出的該路段的低速占比門檻值,則可能車多且速度慢,應視為滿足壅塞條件,因而動態壅塞偵測模組25決定自該總樣本量中選取數量較多的樣本群。
The dynamic
時速聚合模組26對動態壅塞偵測模組25所選取的樣本群中各該區間樣本量及其對應的時速進行時速聚合,即對該選取的樣本群中各區間的時速低值與時速機率值進行加權平均,藉以產生該路段的即時時速資訊,其中時速機率值係指一區間樣本量相比於該選取的樣本群的樣本量。另外,所產生之該段的即時時速資訊也會傳輸至資料庫30,以作為後續計算壅塞門檻值時之更新依據。
The
因此,藉由本案之包括資料接收模組21、批次過濾模組22、路段對應模組23、動態壅塞門檻計算模組24、動態壅塞偵測模組25及時速聚合模組26之設備2,接收來自基地站10的行動信令資料進行批次分析過濾以濾除飄移點位,疊代追蹤時速隨時序的變化,並且依據資料庫30中對應路段的歷史資料計算動態門檻值以藉此進行動態壅塞偵測,於車流量偏低的交通情況下,選取時速較快的樣本群以降低取樣到少數慢車所造成時速演算結果偏慢的影響,而於車流量偏中高的交通情況下,進一步考
慮車速資料以增加壅塞路況的判斷條件,另導入路段壅塞門檻動態更新機制,避免因採用路段基本固定資料而導致與實際路況不符的情形,以提升路況分析正確性。
Therefore, through the
於一實施例中,本案上述各模組所產生之資料可供多種交通服務進行後續應用,透過APP或網頁呈現出來,以供提前道路改向或是發布交通措施以避免道路壅塞,藉此節省旅行時間。於另一實施例中,本案之上述模組亦可單獨拆開,例如批次過濾模組所執行的處理,其中可抓出飄移之點位數量、比例等,後續亦能應用於停駐點、事故偵測等技術;而動態壅塞偵測模組所執行之處理,其中壅塞路況判斷則可用於道路易壅塞程度貼標,將貼標資料作為屬性之一,應用於其他交通分析服務。 In one embodiment, the data generated by each of the above modules in this case can be used by various traffic services for subsequent applications and presented through APPs or web pages for early road diversion or issuing traffic measures to avoid road congestion, thereby saving money. Travel time. In another embodiment, the above-mentioned modules in this case can also be disassembled separately, such as the processing performed by the batch filtering module, in which the number and proportion of drifting points can be captured, and subsequently applied to the stop points. , accident detection and other technologies; and the processing performed by the dynamic congestion detection module, among which the congestion road condition judgment can be used to label the degree of road congestion, using the labeling data as one of the attributes, can be used in other traffic analysis services.
請參閱圖2,本案之分析路段狀況之方法主要包括步驟S201~S206,主要由圖1的設備2所執行。
Please refer to Figure 2. The method of analyzing road section conditions in this case mainly includes steps S201~S206, which are mainly executed by the
於步驟S201中,自基地站接收複數個行動裝置的複數個行動信令資料,接著進至步驟S202。 In step S201, a plurality of mobile signaling data of a plurality of mobile devices are received from the base station, and then proceed to step S202.
於步驟S202中,對複數個行動裝置的複數個行動信令資料進行批次分析過濾,以產生複數個行動信令位置資料。所述批次分析過濾之步驟詳如圖3A所示,包括步驟S301~S304。 In step S202, batch analysis and filtering is performed on a plurality of mobile signaling data of a plurality of mobile devices to generate a plurality of mobile signaling location data. The details of the batch analysis and filtration steps are shown in Figure 3A, including steps S301 to S304.
於步驟S301中,執行批次切割。詳言之,對各個行動裝置的多個行動信令資料依時間順序累積,累計至一定數量後切割為一批次,如批次1,接著進至步驟S302。 In step S301, batch cutting is performed. Specifically, multiple mobile signaling data of each mobile device are accumulated in chronological order. After accumulating to a certain number, they are cut into batches, such as batch 1, and then proceed to step S302.
於步驟S302中,執行主成分分析。詳言之,對批次1的多個點位(也就是行動信令資料)執行主成分分析,參閱圖3B,圖中一批次包含 12個點位P1~P12,點位編號代表點位時間的先後順序,以主成分分析法找出之主成分,如箭頭所示,接著進至步驟S303。 In step S302, principal component analysis is performed. In detail, principal component analysis is performed on multiple points of batch 1 (that is, mobile signaling data). Refer to Figure 3B. The batch in the figure contains There are 12 points P1~P12, and the point numbers represent the order of point time. The principal components are found using the principal component analysis method, as shown by the arrows, and then proceed to step S303.
於步驟S303中,執行方向過濾及距離過濾。如圖3B中所示,12個點位中,可以看到點位P5、P6的平行移動方向與主成分不符,而點位P2、P4、P9、P10與主成分的垂直距離過大,因此過濾後之點位僅剩下點位P1、P3、P7、P8、P11、P12共6個,接著進至步驟S304。 In step S303, direction filtering and distance filtering are performed. As shown in Figure 3B, among the 12 points, it can be seen that the parallel movement directions of points P5 and P6 do not match the principal components, while the vertical distances of points P2, P4, P9, and P10 from the principal components are too large, so filtering After that, there are only 6 points left: P1, P3, P7, P8, P11, and P12, and then proceed to step S304.
於步驟S304中,留下過濾後的點位群,另保留部分點位至下一批次。具體言之,批次1的最後三成點位,也就是點位P11、P12將被保留為下一批次的前面兩者,故批次2的點位係為P11、P12、P13、P14、…,批次3也是以此類推而產生。因此,過濾後點位群1中包括點位P1、P3、P7、P8四個,即作為行動信令位置資料,過濾後點位群2和3也是以此類推而產生故所包括的點位即作為行動信令位置資料,接著進至步驟S203。
In step S304, the filtered point group is retained, and some points are reserved for the next batch. Specifically, the last 30% of the points of batch 1, that is, points P11 and P12, will be retained as the first two of the next batch, so the points of
於步驟S203中,將複數個行動信令位置資料對應至路段,接著進至步驟S204。 In step S203, a plurality of mobile signaling location data are mapped to road segments, and then step S204 is entered.
於步驟S204中,計算出該路段的總樣本量、複數個區間樣本量及其所對應的時速,接著進至步驟S205。於一具體實施例中,請參閱圖5,以一速限80之路段時速樣本分布為例,時速允許到速限+20,因而以(80+20)/5去切割時速區間,故由以樣本量為縱軸而以時速區間為橫軸之圖5可知,該路段當下時間的總樣本量、複數個區間樣本量及其所對應的時速。 In step S204, the total sample size of the road section, the plurality of interval sample sizes and their corresponding hourly speeds are calculated, and then proceed to step S205. In a specific embodiment, please refer to Figure 5. Taking the speed sample distribution of a road section with a speed limit of 80 as an example, the speed is allowed up to the speed limit + 20, so (80+20)/5 is used to cut the speed interval. Therefore, Figure 5 shows the sample size on the vertical axis and the speed interval on the horizontal axis. It can be seen that the total sample size of the road section at the current time, the sample size of multiple intervals and their corresponding speeds.
於步驟S205中,根據該路段的總樣本量與根據該路段的歷史資料所計算出的壅塞門檻值,決定總樣本量中被選取的樣本群。具體實 施請參閱圖4,本案之分析路段狀況之方法中取群方法包括步驟S401~S406。 In step S205, the sample group selected from the total sample size is determined based on the total sample size of the road section and the congestion threshold calculated based on the historical data of the road section. concrete Please refer to Figure 4. The method of analyzing road section conditions in this case includes steps S401~S406.
於步驟S401中,計算該路段的壅塞門檻值。於一實施例中,壅塞門檻值的計算式如下: In step S401, the congestion threshold of the road segment is calculated. In one embodiment, the congestion threshold is calculated as follows:
壅塞門檻值=AVG(樣本量峰值)*0.6 Congestion threshold = AVG (peak sample size) * 0.6
於一實施例中,樣本量峰值:以當下時間為基準,往回推七天內的每個五分鐘樣本量,取樣本量最大的十個樣本量。因此,壅塞門檻值即為近期七天內樣本峰值平均之6成。若無七天內之樣本資訊,則以該路段的道路容量為壅塞門檻值。 In one embodiment, the peak sample size: based on the current time, push back each five-minute sample size within seven days, and select the ten sample sizes with the largest sample size. Therefore, the congestion threshold is 60% of the average sample peak value in the recent seven days. If there is no sample information within seven days, the road capacity of the road section will be used as the congestion threshold.
於步驟S402中,判斷該路段的總樣本量是否大於該路段的壅塞門檻值。若總樣本量不大於壅塞門檻值,則可視為車少或順暢,進至步驟S403。若總樣本量大於壅塞門檻值,則可視為車多,進至步驟S405。 In step S402, it is determined whether the total sample size of the road segment is greater than the congestion threshold of the road segment. If the total sample size is not greater than the congestion threshold, it can be considered that there are few vehicles or the traffic is smooth, and the process proceeds to step S403. If the total sample size is greater than the congestion threshold, it can be considered that there are too many vehicles and the process proceeds to step S405.
於步驟S403中,選擇時速較快的樣本群。於一實施例中,假設該路段的總樣本量為407,則快群取法如下。 In step S403, a sample group with a faster speed is selected. In one embodiment, assuming that the total sample size of the road section is 407, the fast group selection method is as follows.
(由快到慢前N個區間樣本量/總樣本量)>=N/區間數量。 (Sample size of the first N intervals from fast to slow/total sample size)>=N/number of intervals.
請參閱圖6。 See Figure 6.
N=1,4/407<1/20 N=1,4/407<1/20
N=2,(4+8)/407<2/20 N=2,(4+8)/407<2/20
N=3,(4+8+7)/407<3/20 N=3,(4+8+7)/407<3/20
… …
N=12,(4+8+7+10+10+8+16+16+41+38+57+31)/407>12/20 N=12,(4+8+7+10+10+8+16+16+41+38+57+31)/407>12/20
換言之,N=1~12的區間即為時速較快的樣本群,如圖6中的方框所示即為被選取的樣本群,此作為後續計算該路段的即時時速資訊的依據。 In other words, the interval between N=1~12 is the sample group with faster speeds. The box in Figure 6 shows the selected sample group, which serves as the basis for subsequent calculation of the real-time speed information of this road section.
於步驟S405中,判斷該路段的低速占比是否大於該路段的低速占比門檻值。在此之前,於步驟S405中,計算該路段的低速占比門檻值。 In step S405, it is determined whether the low-speed ratio of the road section is greater than the low-speed ratio threshold of the road section. Prior to this, in step S405, the low speed ratio threshold of the road section is calculated.
低速占比門檻值的計算如下: The low-speed ratio threshold is calculated as follows:
於一實施例中,假設該路段過去7天平均時速為70,則低速門檻為70*0.8=56,則以該路當下時間為基準,過去七天低速占比門檻值=過去七天低速樣本量/過去七天總樣本量。 In one embodiment, assuming that the average speed of the road section in the past seven days is 70, the low-speed threshold is 70*0.8=56. Based on the current time of the road, the low-speed proportion threshold in the past seven days = low-speed sample size in the past seven days/ Total sample size over the past seven days.
另外,如圖7所示,假設該例中,總樣本量為407中,時速低於56kph之樣本有292個,而圖7顯示時速0-55區間的樣本量有287個,則該路段當下的低速占比為287/407(約0.7)。 In addition, as shown in Figure 7, assuming that in this example, the total sample size is 407, there are 292 samples with speeds below 56kph, and Figure 7 shows that there are 287 samples in the 0-55 speed range, then the road section is currently The low speed ratio is 287/407 (about 0.7).
若低速占比不大於低速占比門檻值,則可視為車雖多但車速不慢,進至步驟S403。若低速占比大於低速占比門檻值,則可視為車多又慢,滿足壅塞條件,進至步驟S406。 If the low-speed ratio is not greater than the low-speed ratio threshold, it can be considered that although there are many vehicles, the vehicle speed is not slow, and the process proceeds to step S403. If the low-speed ratio is greater than the low-speed ratio threshold, it can be considered that there are many and slow vehicles, and the congestion condition is met, and the process proceeds to step S406.
於步驟S406中,選擇數量較多的樣本群。於一實施例中,假設該路段的總樣本量為407,則大群取法為:M個相鄰區間分為一群,具有時速區間樣本量加總最高的一群即為大群。若不只一群的樣本量加總最高,則選取較時速較低的群。 In step S406, a larger sample group is selected. In one embodiment, assuming that the total sample size of the road section is 407, the large group selection method is: M adjacent sections are divided into one group, and the group with the highest sum of sample sizes in speed sections is the large group. If more than one group has the highest total sample size, the group with the lower speed is selected.
如圖7所示,於一實施例中,三個相鄰區間分為一群,則可知由時速區間35-40、40-45、45-50所構成之群的區間樣本量加總最高, 此為大群,如圖7中的方框所示即為被選取的樣本群,此作為後續計算該路段的即時時速資訊的依據。 As shown in Figure 7, in one embodiment, three adjacent intervals are divided into a group. It can be seen that the group composed of speed intervals 35-40, 40-45, and 45-50 has the highest total sample size. This is a large group, as shown in the box in Figure 7, which is the selected sample group, which is used as the basis for subsequent calculation of the real-time speed information of this road section.
完成上述步驟S205(包括步驟S401~S406)的選取樣本群之後,進至步驟S206。 After completing the sample group selection in step S205 (including steps S401 to S406), proceed to step S206.
於步驟S206中,對所選取的樣本群中各區間樣本量及其所對應的時速進行時速聚合,俾快速且精準地產生該路段的即時時速資訊。具體言之,對該選取的樣本群中各區間的時速低值與時速機率值進行加權平均,以產生該路段的即時時速資訊,其中: In step S206, speed aggregation is performed on the sample size of each section in the selected sample group and its corresponding speed, so as to quickly and accurately generate real-time speed information of the road section. Specifically, the weighted average of the low speed value and speed probability value of each section in the selected sample group is used to generate real-time speed information of the road section, where:
於一實施例中,若選取快群,參閱圖6,如方框所示之快群,12個區間共246個樣本量,時速聚合如下: In one embodiment, if the fast group is selected, refer to Figure 6. The fast group shown in the box has a total of 246 samples in 12 intervals, and the hourly aggregation is as follows:
(4/246)*95+(8/246)*90+(7/246)*85+(10/246)*80+(10/246)*75+(8/246)*70+(16/246)*65+(16/246)*60+(41/246)*55+(38/246)*50+(57/246)*45+(31/246)*40=55.96 (4/246)*95+(8/246)*90+(7/246)*85+(10/246)*80+(10/246)*75+(8/246)*70+(16 /246)*65+(16/246)*60+(41/246)*55+(38/246)*50+(57/246)*45+(31/246)*40=55.96
因此,在此實施例中,該路段當下時段由於總樣本量少於壅塞門檻值,或總樣本量雖然多於壅塞門檻值但低速占比仍少於低速門檻值,故選取快群,計算所得之該時段的即時時速為55.96kph。 Therefore, in this embodiment, since the total sample size of the road section during the current period is less than the congestion threshold, or the total sample size is more than the congestion threshold, the proportion of low speeds is still less than the low speed threshold, so the fast group is selected and calculated. The real-time speed during this period was 55.96kph.
於另一實施例中,若選取大群,參閱圖6,如方框所示之大群,3個區間共136樣本量,時速聚合如下: In another embodiment, if a large group is selected, refer to Figure 6. The large group shown in the box has a total sample size of 136 in 3 intervals, and the hourly aggregation is as follows:
(48/136)*35+(31/136)*40+(57/136)*45=40.33 (48/136)*35+(31/136)*40+(57/136)*45=40.33
因此,在此實施例中,該路段當下時段由於總樣本量多於壅塞門檻值且低速占比亦多於低速門檻值,故選取大群,計算所得之該時段的即時時速為40.33kph。 Therefore, in this embodiment, since the total sample size of the road section during the current period is more than the congestion threshold and the low-speed ratio is also greater than the low-speed threshold, a large group is selected, and the calculated real-time speed of the period is 40.33kph.
須說明的是,本發明之方法可執行在例如伺服器、電腦或其他具有資料處理、運算、儲存、網路通聯等功能的一個單獨或多個集合之設備中,其中,該伺服器、電腦或設備包括中央處理器、硬碟、記憶體等。 It should be noted that the method of the present invention can be executed in, for example, a server, a computer, or other equipment with functions of data processing, computing, storage, network communication, etc., in a single or multiple sets, wherein the server, computer Or equipment includes central processing unit, hard disk, memory, etc.
另外,本發明之電腦程式產品係經由電腦載入程式後執行該方法,而本發明之電腦可讀取紀錄媒體(例如硬碟、軟碟、光碟、USB隨身碟)係儲存有該電腦程式產品。另外,電腦程式產品亦可在網路上直接傳輸提供,故電腦程式產品係為載有電腦可讀取之程式且不限外在形式之物。 In addition, the computer program product of the present invention executes the method after loading the program through the computer, and the computer-readable recording medium (such as hard disk, floppy disk, optical disk, USB flash drive) of the present invention stores the computer program product . In addition, computer program products can also be directly transmitted and provided over the Internet. Therefore, computer program products are things that contain computer-readable programs and are not limited to external forms.
另外,本發明還提供一種電腦可讀取記錄媒體,係應用於具有處理器及/或記憶體之計算設備或電腦中,且電腦可讀取記錄媒體儲存有指令,並可利用計算設備或電腦透過處理器及/或記憶體執行電腦可讀取記錄媒體,以於執行電腦可讀取記錄媒體時執行上述方法及/或內容。 In addition, the present invention also provides a computer-readable recording medium, which is used in a computing device or computer with a processor and/or memory, and the computer-readable recording medium stores instructions, and can be used by the computing device or computer The computer-readable recording medium is executed by the processor and/or memory to execute the above method and/or content when the computer-readable recording medium is executed.
綜上所述,本案揭示一種分析路段狀況之設備、方法及電腦程式產品,利用行動信令資料分析壅塞路況,依序進行信令接收、飄移信令過濾、路段對應、時速計算等步驟,藉由主成分過濾法透過批次點位的移動方向找出向量,不符合該向量方向性的點位即視為飄移點位。而在減低運具影響層面,在路段對應後而開始計算路段時速前,透過歷史資料信令樣本數、車速資料或道路容量等,判別當下車流與時速是否符合壅塞特徵,故無需使用異質資料進行勾稽,透過行動信令本身之移動特性即可將信令進行分群和取群,進而調整用於最終時速計算所使用的樣本群,從而提升路況分析準確性。 In summary, this case discloses a device, method and computer program product for analyzing road section conditions, which uses mobile signaling data to analyze congested road conditions, and sequentially performs steps such as signaling reception, drift signaling filtering, road section correspondence, and speed calculation. The principal component filtering method is used to find the vector through the movement direction of the batch points. Points that do not conform to the directionality of the vector are regarded as drift points. In terms of reducing the impact of transportation vehicles, after the road segment is mapped and before the road segment speed is calculated, the number of historical data signaling samples, vehicle speed data or road capacity are used to determine whether the current traffic flow and speed meet the congestion characteristics, so there is no need to use heterogeneous data Through cross-checking, the signaling can be grouped and grouped through the mobile characteristics of mobile signaling itself, and then the sample group used for final speed calculation can be adjusted to improve the accuracy of road condition analysis.
上述實施例僅例示性說明本案之功效,而非用於限制本案,任何熟習此項技藝之人士均可在不違背本案之精神及範疇下對上述該些實 施態樣進行修飾與改變。因此本案之權利保護範圍,應如後述之申請專利範圍所列。 The above-mentioned embodiments are only illustrative of the effects of this case, and are not used to limit this case. Anyone familiar with this technology can perform the above-mentioned practical actions without violating the spirit and scope of this case. Make modifications and changes to your appearance. Therefore, the scope of rights protection in this case should be as listed in the patent application scope described below.
10:基地站 10: Base station
2:設備 2: Equipment
21:資料接收模組 21:Data receiving module
22:批次過濾模組 22:Batch filter module
23:路段對應模組 23:Road section corresponding module
24:動態壅塞門檻計算模組 24:Dynamic congestion threshold calculation module
25:動態壅塞偵測模組 25:Dynamic congestion detection module
26:時速聚合模組 26: Speed aggregation module
30:資料庫 30:Database
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111126500A TWI799318B (en) | 2022-07-14 | 2022-07-14 | Apparatus and method for analyzing traffic status and computer program product implementing the method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW111126500A TWI799318B (en) | 2022-07-14 | 2022-07-14 | Apparatus and method for analyzing traffic status and computer program product implementing the method |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI799318B TWI799318B (en) | 2023-04-11 |
TW202403537A true TW202403537A (en) | 2024-01-16 |
Family
ID=86948734
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW111126500A TWI799318B (en) | 2022-07-14 | 2022-07-14 | Apparatus and method for analyzing traffic status and computer program product implementing the method |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI799318B (en) |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100070128A1 (en) * | 2008-09-15 | 2010-03-18 | Microsoft Corporation | vehicle operation by leveraging traffic related data |
TWI512690B (en) * | 2013-09-16 | 2015-12-11 | Chunghwa Telecom Co Ltd | Application of vehicle population movement mode changes to detect road congestion |
CN105574895A (en) * | 2016-01-05 | 2016-05-11 | 浙江博天科技有限公司 | Congestion detection method during the dynamic driving process of vehicle |
CN112562334A (en) * | 2020-12-08 | 2021-03-26 | 绍兴数鸿科技有限公司 | Method, device and medium for calculating real-time speed of curved road section based on floating car data |
-
2022
- 2022-07-14 TW TW111126500A patent/TWI799318B/en active
Also Published As
Publication number | Publication date |
---|---|
TWI799318B (en) | 2023-04-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10339800B2 (en) | Method and device for processing traffic road information | |
US20170300919A1 (en) | Transaction risk detection method and apparatus | |
WO2019205020A1 (en) | Road condition recognition method, apparatus and device | |
WO2022142042A1 (en) | Abnormal data detection method and apparatus, computer device and storage medium | |
US9424743B2 (en) | Real-time traffic detection | |
CN106323301A (en) | Road information obtaining method and device | |
CN111737377B (en) | Method and device for identifying drift trajectory, computing equipment and storage medium | |
CN110689804B (en) | Method and apparatus for outputting information | |
CN114579657B (en) | Vehicle-road cooperation-based v2x edge cloud control method and system | |
WO2018001122A1 (en) | Method and apparatus for determining driving speed of bus | |
US20210406564A1 (en) | Perception data detection method and apparatus | |
CN112770265B (en) | Pedestrian identity information acquisition method, system, server and storage medium | |
CN112562330A (en) | Method and device for evaluating road operation index, electronic equipment and storage medium | |
CN107545318B (en) | Bus line priority determination and bus transfer line sequencing method and device | |
CN112185108B (en) | Urban road network congestion mode identification method, equipment and medium based on space-time characteristics | |
CN112530163A (en) | Traffic accident prediction method, traffic accident prediction device, electronic device, and storage medium | |
CN115359471A (en) | Image processing and joint detection model training method, device, equipment and storage medium | |
CN110046179B (en) | Mining method, device and equipment for alarm dimension | |
TW202403537A (en) | Apparatus and method for analyzing traffic status and computer program product implementing the method | |
EP4109431A1 (en) | Method, apparatus and storage medium of determining state of intersection | |
CN107270919B (en) | Bus route grading method and device and bus route navigation method and device | |
CN111709357B (en) | Method and device for identifying target area, electronic equipment and road side equipment | |
CN111797181B (en) | Positioning method, device, control equipment and storage medium for user location | |
CN108010319B (en) | Road state identification method and device | |
CN111708769A (en) | Method, system, computer storage medium and terminal for realizing information processing |