JP6302848B2 - Map generation system and map generation method - Google Patents

Description

  The present invention relates to a technique for generating map data.

  As a technology for detecting and estimating the shape of a newly used road based on position (latitude and longitude) and time information included in probe information from a navigation device or a smartphone, as a background art of this technical field, No. 150796 (Patent Document 1). This publication states that “in the navigation device 1, storage means for storing map information and updating new map information, and two or more types of position detection for detecting the current position of the navigation device 1 as a detection position. A road data presence / absence determining means for determining whether the detected position exists on the road of the map, and a trajectory connecting a plurality of detected positions when determined not to be on the road as a new detected path. New detection route detection means for detecting, new detection route determination means for determining by dividing the new detection route into one or more new detection sections based on whether the vehicle has traveled over a predetermined distance at a predetermined traveling speed or higher, and a new road And road data update means for updating the data in the storage means. "

  In Japanese Patent Application Laid-Open No. 2008-297872 (patent document 2), in detecting an abnormality for repairing a road, a variance and an average value of road surface property information when traveling on a certain road, which is one of probe information, are calculated. As a reference standard, means for collecting the dispersion and average value of road surface property information of other roads constructed in the same period and determining the necessity for repair is described.

JP 2009-150796 A JP 2008-297872 A

  Regarding the technology for detecting and estimating the shape of a newly used road based on the position and time information such as latitude and longitude included in the probe data from the navigation device or the smartphone, a part other than the road as in Patent Document 1 is used. A technique for discriminating is disclosed. However, with regard to the shape of the road, for example, if the movement locus of the in-vehicle terminal (that is, the vehicle equipped with the in-vehicle terminal) is recognized as a road as it is, the shape greatly deviates from the actual road due to GPS error or multipath. May be recognized as a road, the accuracy of the road shape may be reduced.

  The effects of errors such as these can be reduced by collecting and processing a large number of probe data by using the fact that the distribution approximates a normal distribution when collecting a large number of measured values, and improving the accuracy of road shape estimation. I can plan.

  However, it is one of the important product values to provide information on the shape of newly used roads in the navigation system at an early stage, and it will be necessary to provide information on the shape of new roads until a large amount of probe data can be collected. The refusal means a decline in product value.

  The present invention has been made in view of the above-described problems, and an object of the present invention is to estimate shape information of a new road with high accuracy based on a small number of probe data.

  In order to solve the above problems, an embodiment of the present invention is a map generation system including an interface connected to a network, a processor connected to the interface, and a storage device connected to the processor. The storage device includes map data including coordinate values indicating positions and shapes of a plurality of roads, and a vehicle position including coordinate values at each time of each vehicle acquired from a plurality of terminal devices mounted on the plurality of vehicles. And the processor specifies a coordinate value outside the range of the road included in the map data among the coordinate values included in the vehicle position data, and for each of a plurality of regions in the coordinate space , The position within the range of the surrounding road included in the vehicle position data with respect to the coordinate value of the position of the road around the area included in the map data. Based on the standard deviation of the values, the number of coordinate values necessary for estimating the new road with a predetermined accuracy is calculated, and the number of the specified coordinate values in each area is greater than the number of the necessary coordinate values. Determining whether the number of coordinate values is greater than the number calculated based on the standard deviation, and calculating the center of the distribution of the specified coordinate values in the region, A line connecting the centers of the distributions calculated for the region is estimated as a new road.

  According to the present invention, when estimating the shape of a newly used road based on position information or the like included in probe information from a navigation device or a smartphone, accurate road shape estimation is performed with a small number of samples. Can do. Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments.

It is a block diagram which shows the outline of the map data generation system of embodiment of this invention. It is explanatory drawing of the vehicle position data stored in the probe data storage part of embodiment of this invention. It is explanatory drawing of the example of the vehicle position data of embodiment of this invention. It is a flowchart which shows the new road data estimation process which the map server of embodiment of this invention performs. It is a flowchart which shows the 1st process which estimates the road shape included in the new road data estimation process which the map server of embodiment of this invention performs. It is explanatory drawing of the 1st example of the process which specifies the center of distribution of a coordinate value which the map server of embodiment of this invention performs. It is explanatory drawing of the 2nd example of the process which specifies the center of distribution of coordinate value which the map server of embodiment of this invention performs. It is a flowchart which shows the 2nd process which estimates the road shape contained in the new road data estimation process which the map server of embodiment of this invention performs. It is a flowchart which shows the 3rd process which estimates the road shape contained in the new road data estimation process which the map server of embodiment of this invention performs. It is a flowchart which shows the 4th process which estimates the road shape contained in the new road data estimation process which the map server of embodiment of this invention performs.

  An embodiment of the map data distribution system of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing an outline of a map data generation system according to an embodiment of the present invention.

  The map data generation system of this embodiment includes a map server 10 owned by a map generator, a navigation terminal 20, a network 40 and the Internet 41 connecting them. A radio communication device 30 and a GPS (Global Positioning System) device 32 are connected to the navigation terminal 20. The navigation terminal 20 is connected to the network 40 and the Internet 41 via the radio base station 31 by the radio communication device 30. The wireless communication device 30 may be built in the navigation terminal 20. The map server 10 stores the probe information received from the navigation terminal 20 in the probe data storage unit 11.

  In this embodiment, the navigation terminal 20 can be replaced with a mobile terminal such as a smartphone that does not perform navigation and only collects position information. Further, the position information measured by these terminals is not only stored in the map server 10 via radio as described above, but is also temporarily stored in the position data storage unit 26 of the navigation terminal 20, and other than radio or communication. You may store in the probe data memory | storage part 11 in the map server 10 offline by another means.

  The map server 10 includes a processor 17, a memory 18, a network interface (IF) unit 15, a probe data storage unit 11, an update data storage unit 12, and a map data storage unit 16. The memory 18 holds a plurality of programs. As functions realized by executing these programs on a computer, at least a probe data processing unit 13 for processing probe data and an update data management unit 14. There is. Processing realized by the processor 17 executing the program will be described later. In the following description, the processing executed by the probe data processing unit 13 or the like is actually executed by the processor 17 according to the program of the probe data processing unit 13 or the like.

  The probe data storage unit 11, the update data storage unit 12, and the map data storage unit 16 are storage devices such as a hard disk drive (HDD). The map data storage unit 16 stores link data of each link constituting a road on the map. An example of this link data is a map for car navigation. The link data includes information such as road interconnection information, road width, and road type. The map data includes coordinate values indicating the position and shape of map components (features) such as roads and buildings. The map data storage unit 16 of the present embodiment stores at least road data indicating the position, shape, connection relationship, and the like of roads as map data.

  The probe data storage unit 11 stores vehicle position data acquired from each navigation terminal 20. Details of the vehicle position data will be described later (see FIG. 2). The update data storage unit 12 stores data transmitted to the navigation terminal 20 in order to update the map data held by the navigation terminal 20. This data may include data related to the shape of the new road generated by the present embodiment.

  Each navigation terminal 20 is a mobile terminal mounted on a vehicle, and performs route search, output of search results, and the like in order to guide a user on the vehicle to a destination. Hereinafter, one typical configuration of the plurality of navigation terminals 20 will be described. Since the configuration of the other navigation terminals 20 may be the same as that shown in FIG. 1, illustration and description thereof are omitted.

  The function of the map server 10 may be realized by a plurality of computers. For example, each of the probe data storage unit 11 and the map data storage unit 16 is a storage device of another computer connected to the network 40, and the processor 17 of another computer further performs processing to be described later based on the data of these storage units. May be executed.

  The navigation terminal 20 includes a network IF unit 21 for connection to the network 40, a memory 29, a processor 28, a screen input / output IF 25 having a GUI function, and a storage device such as a hard disk drive (HDD) or a flash memory. And have. The memory 29 holds a plurality of programs, and functions realized by executing these programs on a computer include at least a vehicle position recording / transmission unit 61 and a navigation processing unit 62. Further, the storage device includes a position data storage unit 26 and a map data storage unit 27. The map data storage unit 27 holds map data for navigation.

  The vehicle position recording / transmission unit 61 transmits the vehicle position information acquired by the position information acquisition unit 24 to the map server 10 at a predetermined timing (for example, periodically). The acquired position information is recorded in the position data storage unit 26, and then the position information is collectively transmitted to the map server 10 by the vehicle position recording / transmission unit 61, or the recording medium storing the position information is mapped to the map. The map server 10 can also be referred to by a method such as reading by the server 10. The transmitted data is stored in the probe data storage unit 11 of the map server 10 (see FIG. 2).

  The position information acquisition unit 24 uses the GPS device 32 to acquire position information of the navigation terminal 20 (for example, a vehicle on which the navigation terminal 20 is mounted). The acquired position information includes at least two-dimensional coordinate values such as latitude and longitude.

  The navigation terminal 20 illustrated in FIG. 1 is an example of a terminal device, and a part or all of the plurality of navigation terminals 20 may be other types of terminal devices, such as a smartphone that can collect terminal position information. It may be replaced by a terminal device.

  FIG. 2 is an explanatory diagram of the vehicle position data 200 stored in the probe data storage unit 11 according to the embodiment of this invention.

  One record of the vehicle position data 200 includes the position information of the vehicle acquired by the position information acquisition unit 24 of the navigation terminal 20 of one vehicle at a certain time. Specifically, each record includes a device ID 201, a time stamp 202, a latitude 203, a longitude 204, and a speed 205.

  The device ID 201 is information for uniquely identifying each navigation terminal 20. The time stamp 202 is the time when the position information acquisition unit 24 of each navigation terminal 20 acquires the position information. Latitude 203 and longitude 204 are vehicle position information (that is, coordinate values) acquired by each navigation terminal 20 at each time. A speed 205 is a traveling speed of each vehicle at each time.

  For example, in the record 211, the latitude and longitude acquired by the position information acquisition unit 24 of the navigation terminal 20 identified by the identification information “001” at 12:34:00 on May 1, 2012 are 139.7337639 degrees and It is 35.66939167 degrees, indicating that the traveling speed of the vehicle equipped with the navigation terminal 20 at that time was 23 km / h.

  The map server 10 stores the vehicle position information received from each navigation terminal 20 in the probe data storage unit 11 so that one record includes information related to one time from one navigation terminal 20. When the received vehicle position information does not include the speed, the map server 10 calculates the speed of each vehicle at each time based on the time stamp 202, the latitude 203 and the longitude 204 acquired from each navigation terminal 20, and calculates the result. The speed 205 may be stored. For example, since the records 211, 214, and 217 include the latitude and longitude 204 of each time acquired from one navigation terminal 20, the speed of each vehicle of the vehicle on which the navigation terminal 20 is mounted is calculated based on them. be able to. Similarly, the speed of another vehicle can be calculated based on records 212, 215 and 218, and the speed of another vehicle can be calculated based on records 213, 216 and 219.

  Alternatively, the vehicle position data 200 may not include the speed 205, but in that case, the map server 10 determines that each vehicle at each time is based on the time stamp 202, the latitude 203, and the longitude 204 as necessary. Calculate speed.

  Each navigation terminal 20 may transmit the vehicle position information including the position information at that time to the map server 10 every time the position information acquisition unit 24 acquires the position information at each time. Vehicle position information including all coordinate values acquired (for example, one day) may be collectively transmitted to the map server 10 (transmission step 51 in FIG. 1). In the present embodiment, since map information including new road data is created based on position information collected from a plurality of vehicles over a certain period of time, immediacy is not required for transmission of vehicle position information.

  Hereinafter, an example of the vehicle position data 200 will be described with reference to FIG. In these examples, the “existing road” is a road in which the map data storage unit 16 holds information indicating its position, shape, connection relationship with other roads, and the like. On the other hand, the “new road” is a road for which the map data storage unit 16 does not yet hold information indicating its position, shape, connection relationship with other roads, and the like.

  FIG. 3 is an explanatory diagram of an example of the vehicle position data 200 according to the embodiment of this invention.

  In FIG. 3, as an example, existing roads 301A to 301E and a new road 302A connecting them are displayed. In this example, existing roads 301A and 301B are connected at an intersection 303A. The existing roads 301C and 301D are connected at an intersection 303B. An existing road 301E is further connected to the intersection 303B so as to be approximately orthogonal to the existing roads 301C and 301D. The new road 302A connects between the intersections 303A and 303B, and is provided so as to be substantially orthogonal to the existing roads 301A to 301D. For this reason, the new road 302A and the existing road 301E are connected so as to form a substantially straight line at the intersection 303B (in other words, the angle formed by them is approximately 0 degrees).

  Further, in FIG. 3, positions indicated by coordinate values acquired from a plurality of navigation terminals 20 of a plurality of vehicles traveling on the respective roads (that is, positions specified by latitude 203 and longitude 204) are X-type symbols. Is displayed. The coordinate values vary within the road range, and some coordinate values are out of the road range. The existence of such variations is the same in the examples described later.

  A method for estimating a new road based on the distribution of coordinate values and a method for determining whether the estimated new road or the like is connected to an existing road will be described later (see FIG. 5A, FIG. 5B, etc.). ).

  FIG. 4A is a flowchart illustrating new road data estimation processing executed by the map server 10 according to the embodiment of this invention.

  First, the probe data processing unit 13 of the map server 10 refers to the vehicle position data 200 stored in the probe data storage unit 11 and the map data (not shown) stored in the map data storage unit 16, thereby Among the records included in the data 200, a record having a coordinate value outside the range of the existing road is specified (step 401).

  Here, when the map data storage unit 16 stores information for identifying the center line of the road as data indicating the position and shape of the road, the range of the existing road is a predetermined center around the center line. It may be a range of widths. Or when the information which specifies the outline of a road is stored as data which shows the position and shape of a road, the range of the existing road may be the range inside the outline, and the inside of the outline It may be a range obtained by expanding the range by a predetermined amount. Such expansion takes into account that the coordinate value of the vehicle traveling on the road may deviate from the road due to the error of the coordinate value acquired by the position information acquisition unit 24.

  Next, the probe data processing unit 13 excludes the data specified in step 401 (in other words, the records stored in the vehicle position data 200 from which records having coordinate values within the range of the existing road are excluded. A new road is estimated based on the record data (step 402). Details of this estimation process will be described later with reference to FIG. 4B.

  FIG. 4B is a flowchart illustrating a first process for estimating a road shape included in the new road data estimation process executed by the map server 10 according to the embodiment of this invention.

  The process shown in FIG. 4B is executed in step 402 in FIG. 4A.

  First, the probe data processing unit 13 specifies the distribution of coordinate values in each of a plurality of ranges in the coordinate space for the coordinate values included in the record specified in step 401 (step 411), and A line connecting the specified centers of the range is estimated as a road shape (for example, a road center line) (step 412).

  A typical example of the processing executed in step 411 will be described in detail with reference to FIGS. 5A and 5B.

  FIG. 5A is an explanatory diagram of a first example of processing for specifying the center of the distribution of coordinate values, which is executed by the map server 10 according to the embodiment of this invention.

  As in FIG. 3, the position where the X-type symbol is plotted in FIG. 5A is the position indicated by the coordinate values acquired from the navigation terminals 20 of a plurality of vehicles.

  Ranges 501A and 501B shown in FIG. 5A are ranges separated by two straight lines that are parallel to the coordinate space in an arbitrary direction (the x-axis direction in the example of FIG. 5A) and have a predetermined interval. . Similarly, each of the ranges 501C and 501D includes two straight lines that are parallel to the coordinate space in one direction different from the above (in the example of FIG. 5A, the y-axis direction orthogonal to the x-axis) and have a predetermined interval. Is a range delimited by. For example, the x-axis direction may be a latitude direction and the y-axis direction may be a meridian direction.

  In step 411, the probe data processing unit 13 calculates frequency distributions 502A to 502D of coordinate values in the ranges 501A to 501D, respectively. Then, the probe data processing unit 13 calculates the center of the calculated frequency distribution whose variation (for example, the variance value) is equal to or less than a predetermined value. In the example of FIG. 5A, since the variance value of the frequency distribution 502B exceeds a predetermined value, the frequency distribution 502B is excluded from the estimation of the new road, and the probe data processing unit 13 , 503C and 503D. This calculation may be performed by any method. In the present embodiment, the median values of the respective Gaussian distributions calculated by the least square method for the Gaussian distribution are the centers 503A, 503C, and 503D.

  Next, in step 412, the probe data processing unit 13 estimates a line 504A connecting the calculated distribution centers 503A, 503C, and 503D as the shape of the new road (for example, the center line of the new road). Although only four ranges 501A to 501D are shown in FIG. 5A for the sake of simplicity of explanation, the shape of a new road can be improved to a higher resolution by actually setting more ranges and performing estimation based on them. Can be expected.

  FIG. 5B is an explanatory diagram of a second example of the process of specifying the center of the distribution of coordinate values executed by the map server 10 according to the embodiment of this invention.

  Similar to FIG. 3, FIG. 5B also shows coordinate values acquired from the navigation terminals 20 of a plurality of vehicles. Specifically, the position where the triangular symbol is plotted is the position indicated by the coordinate value acquired from one navigation terminal 20 of the plurality of vehicles, and the position where the X-type symbol is plotted is the other It is the position which the coordinate value acquired from the navigation terminal 20 of a plurality of vehicles shows. By connecting the position of the vehicle in order of the time when the respective coordinate values are acquired, the travel locus 505 of the vehicle is acquired. Each of the ranges 501E and 501F shown in FIG. 5B is a range having a predetermined width centered on a straight line orthogonal to the travel locus 505.

  In step 411, the probe data processing unit 13 calculates the frequency distributions 502E and 502F of the coordinate values in the ranges 501E and 501F, and calculates the center of those having a variance value equal to or less than a predetermined value. This calculation method may be the same as in FIG. 5A. In the example of FIG. 5B, since the variance values of the frequency distributions 502E and 502F are not more than a predetermined value, the respective centers 503E and 503F are calculated.

  Next, in step 412, the probe data processing unit 13 estimates a line 504B connecting the calculated distribution centers 503E and 503F as the shape of the new road (for example, the center line of the new road). Although only the two ranges 501E and 501F are shown in the example of FIG. 5B, it can be expected that the shape of the new road is estimated with high resolution by performing estimation based on more ranges.

  When the number of coordinate values included in the range set in step 411 is sufficiently large, it is expected that a new road with a certain degree of reliability can be estimated based on them. On the other hand, when the number of coordinate values is small, it is considered that those coordinate values are not reliable enough to be used for estimating a new road.

  From a statistical viewpoint, the distribution of the coordinate values of the probe data is considered to be a normal distribution with respect to the true position of the new road. Here, the true position of the road is an actual road position, for example, a center position of the actual road. The center of the road may be a geographical center of the road or a position where the probability that the vehicle actually travels on the road is the highest. Hereinafter, the true position of the road is also referred to as the road true value. The road true value of the existing road can be specified based on the map data, but the road true value of the new road is unknown. The deviation of the coordinate value of the probe data with respect to the road true value is caused by the deviation of the position where the vehicle actually travels from the road true value (for example, when the vehicle actually travels right or left from the center of the road). In addition to the above, there are some that are caused by errors in positioning by GPS, and the latter is generally dominant.

  In this case, the accuracy of the estimation of the position of the road, that is, the estimation of the population average depends on the number of samples, that is, the number of coordinate values and the number of probe data. For example, in the case of estimating the population deviation σ and estimating an error of 15 m or less from the road center (population average), that is, a confidence interval width of 15 m and a confidence interval of 95%, the number of necessary probe data samples is statistically Is equal to or greater than the square of ((2 × 1.96 × σ) / 15). Specifically, in order to estimate the position of a new road with the above accuracy, 62 samples are required when the population standard deviation is 30 m, and 28 samples are required when the population standard deviation is 20 m.

  In general, since the true value of a road is unknown for a new road, it is difficult to obtain the standard deviation of probe data there. Statistically, even if the population standard deviation is unknown, it is possible to estimate the population by setting the confidence interval and its width, but the population standard deviation is already known on the spot. It will be much more than in the case of. This means that the new road cannot be estimated with high accuracy until a considerable time has elapsed since the start of the new road. However, in consideration of the convenience of the navigation system, it is desirable that the estimation of a new road is performed earlier while maintaining the accuracy thereof, and navigation of the new road can be performed early. In the present embodiment, the number of samples necessary for performing population estimation with sufficient accuracy is estimated based on variations in probe data of existing roads around a new road to be estimated using the properties of GPS data. When the estimated number of samples of probe data has already been acquired, the population estimation is started to enable early navigation of the new road. An example of processing executed for this purpose will be described with reference to FIGS. These processes are executed in step 402 of FIG. 4A instead of the process of FIG. 4B.

  FIG. 6 is a flowchart showing a second process for estimating the road shape included in the new road data estimation process executed by the map server 10 according to the embodiment of the present invention.

  Considering the nature of the GPS data that is the source of the probe position data, the variance is considered to have the same nature on the new road to be obtained and the surrounding roads. This is because the accuracy of GPS data is affected by surrounding terrain, buildings, and the like, and thus GPS data acquired at geographically close positions is considered to contain similar errors. For this reason, the map server 10 of the present embodiment may determine the number of probe data samples necessary for estimating the shape of the new road based on the standard deviation of the position of the probe data on the surrounding roads whose road shape is known. Good.

  Specifically, the probe data processing unit 13 first calculates a standard deviation from the road true value of the probe coordinates of the surrounding road (step 611). The peripheral road is an existing road that is geographically close to the new road, for example, an existing road whose distance from the probe data that is not excluded in step 401 in FIG. 4A is a predetermined value or less.

  Next, the probe data processing unit 13 determines whether or not the standard deviation of the surrounding road is equal to or less than a certain reference value (step 612). When the standard deviation of the surrounding road is equal to or smaller than a certain reference value, the probe data processing unit 13 regards that the position data can be obtained with high accuracy around the new road, and the required number of samples using the standard deviation of the existing surrounding road Is determined, and it is determined whether or not the number of samples of probe data already acquired is larger than the number of necessary samples (step 613). Here, the necessary number of samples is the number of samples of probe data required for estimating the shape of the new road (for example, the position of the center of the new road) with a predetermined accuracy, as already described.

  When the number of samples of probe data already acquired is larger than the required number of samples, the probe data processing unit 13 performs road shape estimation. Specifically, the probe data processing unit 13 specifies the distribution of coordinate values in each of the plurality of ranges in the coordinate space (step 614), and draws a line connecting the specified centers of the plurality of ranges to the road shape (for example, It is estimated as a road centerline) (step 615). Details of these processes are as illustrated in FIG. 5A and FIG.

  On the other hand, if it is determined in step 612 that the standard deviation of the surrounding road is greater than or equal to a certain reference value, and if it is determined in step 613 that the number of probe data already acquired is less than the required number of samples, The probe data processing unit 13 does not perform the shape estimation of the new road section (step 616).

  Even if the standard deviation of the surrounding road is equal to or larger than a certain reference value in step 612, the probe data processing unit 13 calculates the necessary number of samples based on the standard deviation in step 613, and already acquired probe data. It may be determined whether or not the number of samples is larger than the required number of samples, and if the number is larger, road shape estimation may be performed. However, in that case, the number of samples required is larger than in the case where the standard deviation of the surrounding roads is equal to or less than a certain reference value, so that the time until the road shape estimation is started becomes longer.

  Here, a specific example of the processing of FIGS. 4 and 6 will be described with reference to FIG. When the probe data as shown in FIG. 3 is acquired, the probe data processing unit 13 excludes the probe data within the existing roads 301A to 301E in step 401 of FIG. 4A. Thereby, the probe data outside the range of the existing roads 301A to 301E is specified. In the example of FIG. 3, the specified probe data is obtained from a vehicle that has traveled on the new road 302A.

  Next, the probe data processing unit 13 divides the coordinate space into a plurality of ranges as shown in FIG. 5A or 5B, for example. As a result, probe data from a vehicle traveling on the new road 302A is divided into a plurality of ranges. Then, the probe data processing unit 13 identifies the surrounding roads of each range (for example, a portion of the existing roads 301A to 301E whose distance from each range is within a predetermined value), and is included in the map data. The standard deviation of the coordinate value of the plug data of the surrounding road is calculated with respect to the road true value of the surrounding road (step 611).

  Next, the probe data processing unit 13 determines whether or not the calculated standard deviation is equal to or less than a certain reference value (step 612), and when the calculated standard deviation exceeds the certain reference value, It is determined whether or not the number of samples of the probe data is larger than the necessary number of samples based on the standard deviation (step 613), and the number of samples of the probe data in each divided range is larger than the necessary number of samples based on the standard deviation. In this case, the center of the distribution of the coordinate values of the probe data in each range is calculated (step 614). Then, the probe data processing unit 13 estimates the position of the line connecting the distribution centers calculated in each range as the position of the new road 302A, and the shape of the line as the shape of the new road 302A (step 615). .

  According to the processing of FIG. 6 described above, by using the standard deviation of the probe data of the surrounding road as the standard deviation of the approximate new road probe data, the number of samples is small when the accuracy of the GPS data on the surrounding road is good. Thus, it is possible to accurately estimate the shape of a new road and use the result early.

  FIG. 7 is a flowchart showing a third process for estimating the road shape included in the new road data estimation process executed by the map server 10 according to the embodiment of the present invention.

  When the probe data is analyzed, there is a tendency to accurately match the road true value when there is no large direction change such as a right or left turn. When the vehicle makes a relatively large change in direction such as a right or left turn, there is a high probability that the probe data immediately after that will deviate from the true road value (that is, the error of the acquired coordinate value becomes large). For this reason, the probe data processing unit 13 excludes probe data from a vehicle that has entered with a relatively large change in direction due to a right or left turn or the like for a new road section, and recalculates the standard deviation in the remaining probe data. (Step 711).

  For example, in the example of FIG. 3, it is estimated that probe data that does not belong to any of the existing roads 301A to 301E is probe data from a vehicle that has traveled on the new road 302A. A vehicle that has traveled on the new road 302A makes a right turn from the existing road 301C (ie, with a change in direction of about 90 degrees) and a vehicle that has entered the new road 302A and turns left from the existing road 301D (ie, about 90 degrees). A vehicle that has entered the new road 302A (with a change in direction) and a straight road from the existing road 301E (ie, with a change in direction of about 0 degrees, in other words, with little or no change in direction) In step 711, the probe data processing unit 13 calculates the standard deviation using only the probe data on the existing road 301E from the vehicle that has gone straight on from the existing road 301E and entered the new road 302A. calculate.

  The amount of change in direction when entering the new road 302A can be calculated by various methods, but an example is as follows. The probe data processing unit 13 specifies the movement locus of the vehicle on which each navigation terminal 20 is mounted from the vehicle position data 200, and the vehicle after a predetermined time has passed since the locus deviated from the point where it deviated from the existing road. The angle between the line connecting the position and the existing road can be obtained as the amount of change in direction when entering the new road. For example, by setting a predetermined angle larger than 0 degree and smaller than 90 degrees as a threshold value of the amount of direction change and executing step 711 using the threshold value, a vehicle that has entered the new road 302A from the existing road 301E, that is, It is possible to calculate the standard deviation of probe data of a vehicle that has entered the new road 302A without a relatively large change in direction.

  Next, the probe data processing unit 13 determines whether or not the standard deviation calculated in step 711 is equal to or less than a certain reference value (step 712). When the standard deviation is less than or equal to a certain reference value, the probe data processing unit 13 sets the required number of samples using this standard deviation, and whether or not the number of samples of probe data already acquired is greater than the required number of samples. Is determined (step 713). At this time, the probe data processing unit 13 identifies probe data whose amount of change in direction when entering a new road is equal to or less than a predetermined threshold in the same manner as in step 711, and determines the number of samples as necessary. Compare with the number.

  Then, the probe data processing unit 13 performs road shape estimation when the number of already acquired probe data samples is larger than the required number of samples (steps 714 and 715). Since the road shape estimation is the same as steps 614 and 615 in FIG. 6 except that only the probe data whose direction change amount is equal to or less than a predetermined threshold is used, detailed description thereof is omitted.

  On the other hand, if it is determined in step 712 that the standard deviation of the surrounding road is greater than or equal to a certain reference value, and if it is determined in step 713 that the number of probe data already acquired is less than the required number of samples, The probe data processing unit 13 does not perform the shape estimation of the new road section (step 716).

  As a result, highly probable probe data including a large error is excluded, so that it is possible to estimate the road with high accuracy with a smaller number of probe data.

  FIG. 8 is a flowchart showing a fourth process for estimating the road shape included in the new road data estimation process executed by the map server 10 according to the embodiment of the present invention.

  Since the positional information of the probe data is based on GPS observation values, there may be a case where a certain offset shift from the true road value occurs depending on the arrangement of GPS satellites and the like. Such an offset also occurs in the surrounding roads where the arrangement of GPS satellites that can be seen and the surrounding environment are similar. Therefore, at the time of estimating the shape of the new road described above, the position of the GPS data of the new road extending in the same direction is corrected based on the method of deviation of the GPS data on the surrounding existing road.

  Specifically, the probe data processing unit 13 first executes Steps 811 to 813 and executes Step 817 when the determination result in Step 812 or 813 is NO. These are the same as steps 711 to 713 and 716 of FIG.

  When it is determined in step 813 that the number of samples of probe data already acquired is larger than the required number of samples, the probe data processing unit 13 sets the coordinate values of the probe data from the vehicle that has traveled on the new road to be substantially the same. Correction is performed based on the deviation of the coordinate value of the probe data from the vehicle traveling on the surrounding road at the time from the true value of the road (step 814).

  Here, substantially the same time is a time within a predetermined time range including, for example, an acquisition time of probe data from a vehicle traveling on a new road. This predetermined time range is preferably a time range in which the arrangement of GPS satellites in which the navigation terminal 20 can receive signals is similar. Since the arrangement of GPS satellites is periodic and the same arrangement appears at a predetermined interval, the almost same time is the same GPS satellite arrangement as the probe data acquisition time from a vehicle traveling on a new road. It may be a time within a predetermined time range including the time of appearance.

  The above is an example of a method for specifying a time range in which the arrangement of GPS satellites is similar, and may be specified by other methods, but simple calculation by utilizing the periodicity of the arrangement of GPS satellites as described above. To specify the time range.

  In the present embodiment, GPS is shown as an example of a positioning system, but the present invention can also be applied to a positioning system (for example, a quasi-zenith satellite system, Galileo, GLONASS, etc.) using radio signals from other artificial satellites. it can. In this case, the same processing as described above is performed for the arrangement of the artificial satellites used in each system.

  In addition, when the probe data processing unit 13 corrects the coordinate value of the probe data on the existing existing road by known map matching, for example, the correction amount is the amount of deviation from the true value of the coordinate value of the probe data. It may be estimated that there is.

  Next, the probe data processing unit 13 executes Steps 815 and 816. Since these are the same as steps 714 and 715 in FIG. 7 except that the coordinate values corrected in step 814 are the target, description thereof will be omitted.

  The correction of the coordinate values as described above can be combined with the process of FIG. For example, the probe data processing unit 13 may execute Step 814 after Step 613 in FIG. 6 and execute Steps 614 and 615 for the coordinate values corrected in Step 814.

  As described above, it is possible to estimate the road shape with higher accuracy by correcting the coordinate value of the new road on the basis of the offset amount of the surrounding road in the time zone where the arrangement of GPS satellites is similar.

  In addition, this invention is not limited to above-described embodiment, Various modifications are included. For example, the above-described embodiment has been described in detail for easy understanding of the present invention, and is not necessarily limited to one having all the configurations described. Further, a part of the configuration of an embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of an embodiment. In addition, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Further, each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function is stored in a non-volatile semiconductor memory, a hard disk drive, a storage device such as an SSD (Solid State Drive), or a computer-readable non-readable information such as an IC card, an SD card, or a DVD. It can be stored on a temporary data storage medium.

  Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

10 Map Server 11 Probe Data Storage Unit 12 Update Data Storage Unit 13 Probe Data Processing Unit 14 Update Data Management Units 15 and 21 Network Interface (I / F)
16, 27 Map data storage unit 17, 28 Processor 18, 29 Memory 20 Navigation terminal 24 Position information acquisition unit 25 Screen input / output I / F
26 position data storage unit 30 wireless communication device 31 wireless base station 40 network 41 internet 61 vehicle position recording / transmission unit 62 navigation processing unit

Claims (10)

A map generation system comprising an interface connected to a network, a processor connected to the interface, and a storage device connected to the processor,
The storage device includes map data including coordinate values indicating positions and shapes of a plurality of roads, vehicle position data including coordinate values at each time of each vehicle acquired from a plurality of terminal devices mounted on the plurality of vehicles, and Hold
The processor is
Among the coordinate values included in the vehicle position data, specify a coordinate value outside the range of the road included in the map data,
For each of a plurality of areas in the coordinate space, a standard of coordinate values within the range of the surrounding roads included in the vehicle position data with respect to the coordinate values of the road positions around the respective areas included in the map data Based on the deviation, the number of coordinate values necessary for estimating a new road with a predetermined accuracy is calculated, and whether or not the number of the specified coordinate values in each region is greater than the number of the required coordinate values Determine whether
If the number of identified coordinate values is greater than the number calculated based on the standard deviation, calculate the center of the distribution of the identified coordinate values in the region;
A map generation system, wherein a line connecting the centers of the distributions calculated for a plurality of the regions is estimated as a new road. The map generation system according to claim 1,
The processor determines whether the standard deviation is less than or equal to a predetermined value, and if the standard deviation exceeds the predetermined value, the number of the specified coordinate values is equal to the required coordinate value. A map generation system characterized by not determining whether or not the number is greater than the number. The map generation system according to claim 1,
The processor calculates the amount of change in direction of each vehicle that has entered the new road based on the vehicle position data, and targets the coordinate value of the vehicle whose amount of change in direction is within a predetermined value. Calculating a standard deviation, determining whether the number of the specified coordinate values in each region is greater than the number of the required coordinate values, and calculating a center of the distribution of the specified coordinate values. A featured map generation system. The map generation system according to claim 1,
The coordinate value included in the vehicle position data is a coordinate value acquired based on a radio signal from an artificial satellite,
The processor includes, in the map data, coordinate values within a range of roads around each area acquired at a time when the specified coordinate value is acquired and a time when the artificial satellite is located in a similar manner. A map generation system, wherein the specified coordinate value is corrected based on a difference from a coordinate value of a position of the surrounding road. The map generation system according to claim 4,
Based on the periodicity of the arrangement of the artificial satellites, the processor has an arrangement of the artificial satellites within a predetermined time range including the time when the specified coordinate value is acquired and the same time as the time. A map generation system characterized in that a time within the predetermined time range including a time is a time at which the position of the artificial satellite is similar to the time at which the specified coordinate value is acquired. A map generation method executed by a computer system having an interface connected to a network, a processor connected to the interface, and a storage device connected to the processor,
The storage device includes map data including coordinate values indicating positions and shapes of a plurality of roads, vehicle position data including coordinate values at each time of each vehicle acquired from a plurality of terminal devices mounted on the plurality of vehicles, and Hold
The map generation method includes:
A first procedure for specifying a coordinate value outside the range of the road included in the map data among the coordinate values included in the vehicle position data;
For each of a plurality of areas in the coordinate space, a standard of coordinate values within the range of the surrounding roads included in the vehicle position data with respect to the coordinate values of the road positions around the respective areas included in the map data Based on the deviation, the number of coordinate values necessary for estimating a new road with a predetermined accuracy is calculated, and whether or not the number of the specified coordinate values in each region is greater than the number of the required coordinate values A second procedure for determining whether
A third step of calculating a center of the distribution of the specified coordinate values in the region when the number of the specified coordinate values is greater than the number calculated based on the standard deviation;
And a fourth procedure for estimating a line connecting the distribution centers calculated for a plurality of the regions as a new road. The map generation method according to claim 6,
Further comprising determining whether the standard deviation is less than or equal to a predetermined value;
The map generation method, wherein the second procedure is not executed when the standard deviation exceeds the predetermined value. The map generation method according to claim 6,
Further comprising calculating a direction change amount of each vehicle entering the new road based on the vehicle position data;
The map generating method, wherein the second procedure to the fourth procedure are executed for a coordinate value of a vehicle whose amount of change in direction is within a predetermined value. The map generation method according to claim 6,
The coordinate value included in the vehicle position data is a coordinate value acquired based on a radio signal from an artificial satellite,
The coordinate values in the range of the roads around each area acquired at the time when the specified coordinate values were acquired and the time when the arrangement of the artificial satellite is similar, and the roads in the surroundings included in the map data A map generation method, further comprising a step of correcting the specified coordinate value based on a difference from the coordinate value of the position of. The map generation method according to claim 9,
The time at which the specified coordinate value is acquired and the time at which the artificial satellite arrangement is similar are within a predetermined time range including the time at which the specified coordinate value is acquired, and the arrangement of the artificial satellite. A map generation method comprising: a time within the predetermined time range including a time at which the same arrangement of the artificial satellites appears as the time based on periodicity.

Images