TWI550566B - Travel road identification system and its method - Google Patents

Description

Driving section identification system and method thereof

The present invention relates to a traveling section identification system and method thereof, and more particularly to a driving section identification system based on cellular network data to determine whether a vehicle is traveling on a predetermined route and a method thereof.

In general, for a short distance, pedestrians or drivers can usually move toward the destination smoothly by remembering the familiar geography or map. However, for long distances or unfamiliar routes and environments, even with map references, most people may still go wrong or lose their way without knowing it. In addition, since passenger transporters or transport operators have arranged routes for driving to transport passengers to destinations or pick up passengers, however, although the route has been arranged, it may still deviate because of driving irregularities. The established driving route.

Based on the above problems, it is known that a technique for identifying road segments has been developed to determine whether there is an offset route by means of image recognition. However, it is not possible to set up a camera at each intersection, which may cause insufficient coverage. . There is also a conventional road segment identification technology that uses cell identification codes for identification. However, since one cell may cover multiple roads, and one road may be covered by multiple cells, it is only judged by the cell identification code. May cause errors.

In view of the above-mentioned problems of the prior art, the object of the present invention is to provide a traveling road segment identification system based on cell network data and a method thereof, and consider the cell identification code, sequence, length of stay time, and then develop algorithms and historical data. The comparison is made to obtain the most similar plural data, and the most corresponding road segments in the plurality of data are used as the road segments currently traveled by the vehicle, thereby improving the accuracy of the route deviation determination and avoiding the occurrence of misjudgment.

According to the purpose of the present invention, a driving road segment identification system is provided, which comprises: a cloud historical database, which stores a plurality of historical driving tracks, each historical driving track includes a historical vector set corresponding to each road segment; and an in-vehicle terminal device When the vehicle is traveling for a certain period of time, the vehicle terminal device collects an azimuth information, and periodically captures the cell network signal to collect a plurality of cell identification codes, and records a time point respectively when collecting each cell identification code. In order to record the time duration of one of the cell identification codes, when the vehicle terminal device determines that the azimuth information changes above a threshold, it will determine that the vehicle is steering, and the vehicle terminal device transmits the road segment before the vehicle is turned. a plurality of cell identification codes, a duration of each time, and a sequence of each cell identification code; and a cloud computing server connected to the cloud historical database and the vehicle terminal device to receive and analyze the plurality of transmitted by the terminal device Cell identification code, duration of each time, and sequence of each cell identification code, and thus Establish a set of vectors, and then the cloud computing server and a plurality history vector set vector set for comparison, in order to determine whether the vehicle deviates from the established route.

According to the purpose of the present invention, a driving road segment identification method is further provided, which comprises the steps of: collecting azimuth information by using one of the vehicle-mounted terminal devices on the vehicle when driving a road segment, and periodically capturing the cell network signal; Collecting a plurality of cell identification codes, and recording a time point corresponding to each cell identification code to record a time stay length of each cell identification code; determining whether the azimuth information change is higher than a threshold through the vehicle terminal device Value, if yes, determining that the vehicle is performing steering to transmit the sequence of multiple cell identification codes, time durations, and cell identification codes collected by the road segment before the vehicle is turned to a cloud computing server; using cloud computing The server analyzes the plurality of cell identification codes transmitted by the vehicle terminal device, the duration of each time interval, and the sequence of each cell identification code, thereby correspondingly establishing a vector set; and collecting the vector and the cloud history data by using the cloud computing server The plurality of historical travel trajectories stored in the library respectively correspond to each road segment History vector set for comparison, in order to determine whether the vehicle deviates from the established route.

Preferably, each historical driving track further includes the number and name of each road segment.

Preferably, when the in-vehicle terminal device is connected to a cell network base station, the cell identification code corresponding to the cell network base station is collected and corresponds to the time point of the record collection.

Preferably, the cloud computing server further sets the right to each received cell identification code. The weight value, in which the order of the collected or the longer the time length of the cell, the higher the weight value of the cell identification code.

Preferably, the cloud historical database further stores a segment number table corresponding to the predetermined route, wherein the cloud computing server further analyzes the number of the segment according to the comparison result of the vector set and the plurality of historical vector sets, and then analyzes the number of the segment. The analyzed road segment number is compared with the road segment number table to determine whether the vehicle is traveling on the established route.

According to the invention, the driving section identification system and the method thereof can have one or more of the following characteristics:

1. The present invention can effectively identify the driving route by analyzing the cell network data, and use the back-end cloud server to perform calculations to determine whether the driving deviates from the established route.

2. The present invention considers factors such as cell identification code, collection order, and length of stay time to determine the deviation of the route, and the performance is higher than that proposed by the prior art.

100‧‧‧Cloud History Database

200‧‧‧Vehicle terminal equipment

300‧‧‧Cloud computing server

S11~S15‧‧‧Steps

Cell 1~Cell 5‧‧‧ cells

Road 1, Road 2‧‧‧

Figure 1 is a block diagram of the driving segment identification system of the present invention.

Fig. 2 is a flow chart showing the method for identifying the driving section of the present invention.

Figure 3 is a schematic diagram of the traffic network and cell coverage.

Figure 4 is a schematic illustration of the system architecture of an embodiment of the travel segment identification system of the present invention.

The technical features, contents, and advantages of the present invention, as well as the advantages thereof, can be understood by the present inventors, and the present invention will be described in detail with reference to the accompanying drawings. The subject matter is only for the purpose of illustration and description. It is not intended to be a true proportion and precise configuration after the implementation of the present invention. Therefore, the scope and configuration relationship of the attached drawings should not be interpreted or limited. First described.

Please refer to FIG. 1 , which is a block diagram of the driving section identification system of the present invention. In the figure, the driving segment identification system includes a cloud history database 100, an in-vehicle terminal device 200, and a cloud computing server 300. The cloud computing server 300 is connected to the cloud history database 100 and the in-vehicle terminal device 200. The Cloud History Database 100 stores a plurality of historical travel trajectories, each calendar The historical travel track contains a set of historical vectors corresponding to each road segment, and each historical travel track further includes the number and name of each road segment. When the vehicle travels for a certain period of time, the vehicle terminal device 200 collects an azimuth information, and periodically extracts the cell network signal to collect a plurality of cell identification codes, and respectively records a time point when collecting each cell identification code. To record the time duration of one of the cell identification codes. The vehicle terminal device 200 is sequentially connected to each cell network base station covering the road segment to obtain the cell identification codes corresponding to the cell network base stations one by one, and correspondingly record when collecting each cell identification code. The point in time and order of acquisition. When the in-vehicle terminal device 200 determines that the azimuth information changes above a threshold, it will determine that the vehicle is performing steering. At this time, the in-vehicle terminal device 200 transmits a plurality of cell identification codes and time times collected by the road segment before the vehicle turns. The length of stay and the order of each cell identification code are sent to the cloud computing server 300. The cloud computing server 300 can be used to analyze the plurality of cell identification codes transmitted by the vehicle terminal device 200, the duration of each time interval, and the sequence of each cell identification code. During the analysis, the cloud computing server 300 identifies each received cell. The code sets a weight value, wherein the higher the weight of the cell identification code is, the higher the weight of the cell identification code is, and then the cell identification code, the time stay length and the sequence correspondence are used to establish a vector set, and then This vector set is compared with a plurality of historical vector sets to determine whether the vehicle deviates from the established route. The cloud history database 100 further stores a road segment number table corresponding to the predetermined route, and the cloud computing server 300 analyzes the number of the road segment according to the comparison result after comparing the vector set of the road segment, and analyzes the road segment number and The link number table is compared to determine whether the vehicle is driving on a predetermined route.

Please refer to FIG. 2 , which is a flow chart of the method for identifying a driving road segment according to the present invention, and the process steps are as follows:

Step S11: collecting azimuth information by using one of the vehicle terminal devices on the vehicle when driving the vehicle, and periodically collecting the cell network signal to collect a plurality of cell identification codes, and collecting each cell identification code. Recording a time point respectively to record the time duration of one of the cell identification codes. This step mainly collects cell identification codes, time points, and the like on the road through the vehicle terminal device.

Step S12: determining whether the azimuth information change is higher than a threshold value by the vehicle terminal device, and if yes, determining that the vehicle is performing steering to collect a plurality of cell identification codes, time durations, and time lengths collected by the road segment before the vehicle is turned The order of each cell identification code is sent to one Cloud computing server.

Step S13: The cloud computing server is used to analyze the plurality of cell identification codes transmitted by the vehicle wearing terminal device, the length of each time staying, and the order of each cell identification code, thereby correspondingly establishing a vector set. This step mainly converts the cell identification code, time point and other data into a vector space, and analyzes the connection order of the cell identification codes in the same road segment, and performs weighting according to the connection order, for example, the earlier the cell identification code is connected to the line. The higher the weight. And analyzing the length of staying time of each cell identification code in the same road segment, and weighting according to the length of staying time, for example, the longer the staying time, the higher the weight of the cell identification code.

Step S14: The cloud computing server compares the vector set with a historical vector set corresponding to each of the plurality of historical road tracks stored in the cloud historical database. This step mainly compares the vector set obtained by converting the current cell identification code and time point data collected by the in-vehicle terminal device 200 with the vector set of each section of the historical data, and extracts the most similar majority comparison result.

Step S15: The cloud computing server analyzes the number of the road segment according to the comparison result of the vector set and the plurality of historical vector sets, and then analyzes the road segment number corresponding to the established route stored in the cloud historical database. A road number table is compared to determine whether the vehicle is on a predetermined route. This step is based on the majority of the comparison results, in a majority decision to determine the section number of the vehicle currently traveling, and then compare the analyzed section number with the list of the number of the route of the established route, to determine whether the driving section is between Within the established route.

The following is a detailed description of the technical flow of the present invention.

The present invention mainly uses cell network data to determine the road segment of the vehicle, and each road will be covered by the transmission range of many cell network base stations. As shown in Fig. 3, it is a traffic network and cell coverage. Schematic diagram. As can be seen from Figure 3, there are five cells Cell 1~Cell 5, and there are two road sections Road 1, Road 2. When the paths are different, the cells they cover will also be different. Therefore, the network signal information of the mobile device is recorded, and the cell number is recorded according to the delivery order, and the mobile device can be judged to move on that path. For example, when the mobile device moves on the road section 1, the order of cell switching is likely to be Cell 1→Cell 2→Cell 4; if the mobile device moves on the road 2, the order of cell switching is likely to be Cell 1→Cell 2→Cell 4→Cell 5. In addition, when only the cell identification of the connection is judged, there may be no The method distinguishes the situation of the parallel road and the opposite road. Therefore, the present invention also considers the cell sequence and the length of the staying time, and analyzes the driving section by analyzing factors such as the cell identification code, the sequence, and the length of the staying time.

The present invention designs a traveling section identification system based on cellular network data, and the system architecture of the embodiment is as shown in FIG. The system will include a plurality of in-vehicle terminal devices 200, a cloud computing server 300, and a cloud history database. The azimuth information, cell network signal (including the cell ID (Cell ID)), will be periodically collected and recorded by the in-vehicle terminal device 200 having the azimuth sensor and the cell network module. Time point. And an azimuth threshold value (for example, 30 degrees) is set as the vehicle terminal device 200 steering judgment application. When the azimuth angle changes beyond the threshold value, the vehicle-mounted terminal device 200 performs steering, and the cell identification code and the time point of the connection collected before the steering are transmitted back to the cloud computing server 300. For example, the in-vehicle terminal device moving in the road segment in FIG. 3 collects and records the azimuth information, the cell identification code, and the time point every 1 second, and sequentially connects to the cells Cell 1, Cell 2, and Cell 3, respectively, which are connected thereto. The cell identification code and time point will be recorded. Table 1 is azimuth information, cell identification code, time point of the schematic data, is a sample per second, can get the cell identification code connected to each second and the current azimuth, and they are connected to the cell Cell 1, respectively Cell 2, Cell 3. When leaving the road section, the vehicle makes a turn, the direction angle changes by 90 degrees (as shown in Table No. 240), and is greater than the threshold value of 30 degrees. At this time, the cells collected by the road segment (No. 1~239) will be collected. The identification code and time point are passed back to the cloud computing server. When the cloud computing server receives the cell identification code and the time point from the vehicle terminal device, it is input into the driving road segment identification method, and compared with the data of the cloud historical database to determine whether it deviates from the predetermined route.

The invention designs a method for segment identification based on cell network data, which mainly comprises four steps: (1) collecting connection signals and handover signals of the cell network; (2) analyzing cell identification codes, sequences, The dwell time is converted into a vector space; (3) compared with each section of the historical data, the most similar number k is taken out, and the subsection is judged; and (4) whether the deviation from the established route is recorded and recorded. The above steps are described below.

(1) Collect the connection signal and handover signal of the cell network: the cell identification code of the connection will be collected by the vehicle terminal device and record the time point, then return to the cloud computing server and be driven by the cloud server. The operation of the road segment identification method.

(2) Analysis of cell identification code, order, dwell time, and conversion to vector space when the cloud server receives the cell identification code and time point from the in-vehicle terminal device, and assumes that all cells in the environment (including historical data) After a total of n, the cell identification code, the order, and the length of the dwell time are respectively converted into a vector space, as follows.

a. Cell identification code: setting the corresponding value of the connected cells in the data returned by the vehicle terminal device is 1, and the corresponding value of the unconnected cells is 0, so that the data can be expressed as C, such as the formula ( 1) shown. Taking Table 1 as an example, it is connected to cells Cell 1, Cell 2, and Cell 3, so c1, c2, and c3 are all 1, and all others are 0, which can be expressed as the vector space of formula (2).

C = {1,1,1,0,...,0}.........(2)

b. Cell sequence: In view of the judgment of the road, the present invention also considers the cell sequence, so that the cell to which the cell is first joined has a higher weight, and the cell weight to which the cell is attached is lower. And only consider the first x connected cells, set a vector A = { a ₁ , a ₂ , ..., a _x } with x weight values, so that the first connected cell weight is a1, the first The two linked cells have a weight of a2, and so on. The permissions for the x cells are not set to 0. Finally, according to the weight oi set of each cell, it is represented as O, as shown in formula (3). The value of x can be set according to the application context. In this embodiment, the value of x is set to 3, that is, only the first three connected cells are considered, and the weight vector is set to A = {1, 0.5, 0.25}. According to the data in Table 1, the first connected cell (ie, Cell 1) has a weight of 1, the second linked cell (ie, Cell 2) has a weight of 0.5, and the third linked cell (ie, Cell). 3) The weight is 0.25. This is represented as a vector set O, as shown in equation (4).

O = { o ₁ , o ₂ , o ₃ , o ₄ ,..., o _n }, where o _i = order weight corresponding to cell i ... (3)

O = {1.0,0.5,0.25,0,...,0}......(4)

c. Length of stay: In view of the judgment of parallel roads, since the length of staying time of each cell is not necessarily the same even for parallel roads, the present invention also considers the length of stay time to reduce the misjudgment of parallel roads. possibility. However, there may be signal interference conditions in the cellular network, which may cause swings back and forth between cells. As numbered 5 to 7 in Table 1, first change from cell 1 to cell 2, then cell 2 to cell 1. Therefore, in the present invention, the number of samples of the cell is multiplied by the sampling cycle time as the length of the residence time. Taking Table 1 as an example, there are 239 data in the data returned to the cloud computing server, in which the cell Cell 1 has 6 data, the cell Cell 2 has 153 data, and the cell Cell 3 has 80 data. Further, since the sampling period was 1 second, the residence time of the cell Cell 1 was 6 seconds, the residence time of the cell Cell 2 was 153 seconds, and the residence time of the cell Cell 3 was 80 seconds. And only consider the first y connected cells, set a vector B = { b ₁ , b ₂ ,..., b _y } with y weight values, so that the cell with the longest retention time has a weight of b1, the residence time The second long cell has a weight of b2, and so on. The permission for the y cells is not set to 0. Finally, the set of weights ti according to each cell is represented as T, as shown in equation (5). The y value can be set according to the application context. In this embodiment, the y value is set to 3, that is, only the first three connected cells are considered, and the weight vector is set to B = {1, 0.5, 0.25}. According to the data in Table 1, the cells with the longest residence time (ie Cell 2) have a weight of 1, the cells with the second longest residence time (ie Cell 3) have a weight of 0.5, and the cells with the third longest residence time (ie Cell 1) The weight is 0.25. This is represented as a vector set T as shown in equation (6).

T = { t ₁ , t ₂ , t ₃ , t ₄ ,..., t _n }, where t _i = cell time i corresponding dwell time weight... (5)

T = {0.25, 1.0, 0.5, 0, ..., 0}.........(6)

d. Comprehensive consideration of the above factors: The present invention will simultaneously consider factors such as cell identification code, cell order, length of residence time, and combine equations (1), (3), and (5) into a vector set R, such as formula (7). ) shown. And the data of Table 1 can be expressed as formula (8).

R = { C , O , T }={ c ₁ , c ₂ , c ₃ , c ₄ ,..., c _n , o ₁ , o ₂ , o ₃ , o ₄ ,..., o _n , t ₁ , t ₂ , t ₃ , t ₄ ,..., t _n }.........(7)

T = {1,1,1,0,...,0,0.25,1.0,0.5,0,...,0,0.25,1.0,0.5,0,...,0}.........(8 )

(3) Comparing with the historical sections of the historical data, taking out the most similar k-pen and judging the sub-section: the plurality of segments corresponding to the various segments in the cloud historical database of the present invention are in the formula (7) format. Historical vector data, in which a total of m pen data exists. The new data reported by the in-vehicle terminal device can be compared with the historical data set H = { h ₁ , h ₂ , ..., h _m }, and the distance of each piece of data hz is calculated by the vector distance formula, such as a formula ( 9) and (10). And take the nearest k-pen data from the historical data, observe the corresponding road segment number, and determine the road segment by majority decision.

_{H = {h 1, h 2} , ..., h m}, where h 1 = {C 1, O 1, T 1} = {c 1,1, c 2,1, c 3,1, c 4 _,1 ,..., c _{n ,1} , o _1,1 , o _2,1 , o _3,1 , o _4,1 ,..., o _{n ,1} , t _1,1 , t _2,1 , t _3,1 , t _4,1 ,..., t _{n ,1} }.........(9)

Assume that there are 5 cells (ie, n=5), and there are 6 data in historical data (ie m=6), historical data H = { h ₁ , h ₂ ,..., h _m } and new data R As shown in Table 2. The distance between R and h1~h6 can be calculated by using formula (10). The distances are 0, 0.71, 1.17, 1.06, 2.83, 2.94. In this example, let k be 3, that is, observe the closest 3 strokes, so the most recent data are h1, h2, and h4, and the corresponding road segment numbers are 1, 1 and 2, respectively, and the majority decision is 1. Therefore, it can be judged that the in-vehicle terminal device has traveled through the link number 1.

(4) Evaluate whether to deviate from the established route and record: The user can first set a plurality of specific routes, each of which contains a plurality of link numbers. When the step (3) determines the driving section number, it can compare with the section number in the established route to determine whether there is a correct line. Driving in a given route, if the road segment is not in the established route, it is a deviation from the route.

In summary, the system and method for detecting road segment identification based on cell network data disclose the cell identification code and time point of the connection by the vehicle terminal device through the previously established driving segment identification system. The cloud computing server performs driving segment identification. Then, the cloud computing server uses the driving segment identification method to analyze factors such as cell identification code, sequence, and length of stay, and converts it into a vector space, and then compares it with the data in the cloud history database. After taking out the most similar k-pen, judge the driving section of the driving in a majority decision. Finally, compare with the established route to determine whether it deviates from the route.

The above is intended to be illustrative only and not limiting. Any equivalent modifications or alterations to the spirit and scope of the invention are intended to be included in the scope of the appended claims.

S11~S15‧‧‧Steps

Claims (10)

A driving road segment identification system, comprising: a cloud historical database, storing a plurality of historical driving tracks, each of the historical driving tracks comprising a historical vector set corresponding to each road segment; and an in-vehicle terminal device driving the vehicle in a road segment The in-vehicle terminal device collects an azimuth information, and periodically extracts a cell network signal to collect a plurality of cell identification codes, and records a time point corresponding to each of the cell identification codes to record One of the cell identification codes has a time stay length. When the vehicle-mounted terminal device determines that the azimuth information changes above a threshold, it determines that the vehicle is steering, and the vehicle-mounted terminal device transmits the road segment before the vehicle turns. Collecting the plurality of cell identification codes, each of the time stay lengths, and the order of the cell identification codes; and a cloud computing server connecting the cloud history database and the vehicle terminal device to receive and analyze the vehicle Wearing the plurality of cell identification codes transmitted by the terminal device, each of the time stay lengths, and each of the cell identification codes Sequence, thereby associating a set of vectors, and the server computing cloud then the set of vectors and a plurality of the history vector set for comparison, thereby determining whether or not the vehicle deviates from a predetermined route. The driving road segment identification system according to claim 1, wherein each of the historical driving tracks further includes a number and a name of each road segment. The driving section identification system according to claim 1, wherein when the in-vehicle terminal device is connected to a cell network base station, the cell identification code corresponding to the cell network base station is collected and corresponding to the record collection. At that point in time. The traveling road segment identification system of claim 1, wherein the cloud computing server further sets a weight value for each of the received cell identification codes, wherein the order of the collected is earlier or the length of the time stays longer The weight value of the cell identification code is higher. The traveling road segment identification system according to claim 1, wherein the cloud historical database further stores a road segment number table corresponding to the predetermined route, and the cloud computing server compares the vector collection of the road segment The result is analyzed by the number of the road segment, and the analyzed road segment number is compared with the road segment number table to determine whether the vehicle is driving on the established route. A driving road segment identification method, comprising the following steps: collecting an azimuth angle by using one of the vehicle-mounted terminal devices on the vehicle when driving the vehicle And periodically collecting the cell network signals to collect a plurality of cell identification codes, and respectively recording the time points of each of the cell identification codes to record a time duration of each of the cell identification codes; Determining, by the in-vehicle terminal device, whether the azimuth information change is higher than a threshold value, and if so, determining that the vehicle is performing steering to capture the plurality of cell identification codes collected by the road segment before the vehicle is turned, and each time stays The length and the sequence of each cell identification code are sent to a cloud computing server; the cloud computing server is used to analyze the plurality of cell identification codes transmitted by the vehicle terminal device, each time duration and each cell identification code The sequence is further configured to establish a vector set; and the cloud computing server performs the vector set with a historical vector set corresponding to each of the plurality of historical trajectories stored in the cloud historical database. Compare to determine if the vehicle deviates from the established route. The method for identifying a traveling road segment according to claim 6, wherein each of the historical driving tracks further includes a number and a name of each road segment. The method for identifying a driving section according to claim 6, wherein when the in-vehicle terminal device is connected to a cell network base station, the cell identification code corresponding to the cell network base station is collected and corresponding to the record collection. At that point in time. The driving route segment identification method according to claim 6, wherein the cloud computing server further sets a weight value for each of the received cell identification codes, wherein the order of the collected is earlier or the time is longer. The longer the weight value of the cell identification code is. The driving road segment identification method according to claim 6, wherein the cloud computing server further analyzes the number of the road segment according to the comparison result of the vector set and the plurality of historical vector sets, and then The analyzed road segment number is compared with a road segment number table corresponding to the established route stored in the cloud historical database, so as to determine whether the vehicle is driving on the established route.