JPH0561410A - Vehicle position detecting device - Google Patents

Abstract

PURPOSE:To detect the accurate position of a vehicle even when road data used for map matching arithmetic operation include an error in partial shape. CONSTITUTION:A map range selecting means 105 estimates the error quantity of the road data and uses a part which is small in error quantity as an evaluation range. A map matching arithmetic means 106 performing the matching arithmetic operation between a travel track found from the outputs of an azimuth sensor 101 and a distance sensor 102 and the road data by using only the evaluation range of the road data which is small in error quantity to calculate the current position of the vehicle calculates the present position of the vehicle.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a navigation system for guiding and guiding a route from a starting point to a destination,
In particular, the present invention relates to a vehicle position detection device that accurately obtains the current position of a moving body such as an automobile running on a road.

[0002]

2. Description of the Related Art In a conventional vehicle position detecting device, a moving distance and a traveling direction of a vehicle are obtained from a distance sensor and an azimuth sensor, respectively, and a running locus is calculated by self-contained navigation, and a CD-RO.
There has been devised a method for estimating the position of the vehicle while correlating with map data stored in a memory such as M. As a conventional vehicle position detecting device, for example, there is one disclosed in Japanese Patent Laid-Open No. 2-130415. FIG. 26 is a block diagram showing the configuration of the conventional technique.

The distance sensor 2601 outputs a pulse signal for each unit traveling distance according to the rotation of the tire, and the traveling distance of the vehicle can be known by counting the number of pulses. Direction sensor 2602 is the turning angular velocity of the vehicle
(Yaw rate) is detected and a signal proportional to the turning angle of the vehicle is output. The output signals of both sensors are the signal processing device 260.
3, and the position of the vehicle is obtained by sequential calculation on the XY coordinates. The position information calculated each time the vehicle travels a unit distance is stored in the travel locus storage device 2604. From the road data stored in the map information storage medium 2605, only the necessary portion is selectively read out by the storage medium reproducing device 2606. Signal processing device 2603
The current position calculated by is displayed on the display device 2607 so as to be superimposed on the surrounding road map. Operating device 260
Reference numeral 8 is an instruction to change the scale factor of the map to be displayed and the display direction.

According to this conventional example, the traveling locus of the vehicle is approximated by a straight line with a constant distance, while the estimated current position of the vehicle on the road currently traveling and on all roads branched from the road. Are set respectively, and the road is approximated to a polygonal line like the trajectory. Next, the locus and the road are matched so that the deviation between the polygonal line vector forming the locus and the polygonal line vector forming the road is minimized. This process is performed for all roads, and the estimated current position on the road with the smallest vector deviation is set as the current position. If the roads are complicated and complicated, the number of roads to be extracted increases. Therefore, the distance between the current position deviated from the road and the estimated current position on the road is calculated. By selecting, the estimated current position on the road, which is a candidate for the current position, is reduced to reduce the calculation load.

On the other hand, there is also a method of calculating the position of a vehicle by receiving radio waves from a satellite. This is generally called radio navigation, and is usually the Global Positioning System.
m, abbreviated as GPS) is used, but radio waves from at least three satellites are received, the distances to the satellites are obtained, and positioning is performed by trilateration. Conventionally, these two methods are used alone or in combination to calculate the current position of the vehicle, and to CR the current position together with the obtained map information around the vehicle position.
It was displayed on a display device such as T.

In order to calculate the position of the vehicle with high accuracy, it is necessary that the outputs of the respective sensors are stable. Especially, the distance sensor has a long-term change due to factors such as changes in tire air pressure. There is a problem in that there is a difference between the mileage obtained from the output and the actual mileage. In order to solve these problems, it has been considered to use a positioning position by GPS, for example, Japanese Patent Laid-Open No. 1-1424.
In 12, it was considered to correct the distance conversion constant of the distance sensor output from the history of the position measured by the GPS receiver.

[0007]

In the vehicle position detecting device according to the conventional technique as described above, highly accurate road data is required, but a great deal of labor is required to improve the accuracy of the road data. Since new roads are created every day, it is impossible to obtain road data without error. For example, when a newly opened road is passed and the road is not described in the road data, the estimated current position cannot be created at the correct position. Further, the curve may be simplified to reduce the amount of data, and the intersection may be different from the actual shape. Further, the square calculation is required to calculate the distance, and the calculation requires a lot of time. Further, in the conventional technique, since the pattern matching of the road data and the locus is performed unconditionally, correct position detection cannot be performed if there is an error in the shape of the reference road data. Thus, when there is a portion where the shape of the road data is different from the actual shape of the road, there is a problem that accurate position detection cannot be performed.

When GPS is used, particularly when positioning is performed using GPS in urban areas, the satellite arrangement is GD.
Even when the value of the geometric accuracy reduction rate called OP is small, the positioning accuracy often decreases due to the influence of indirect waves reflected by obstacles such as buildings around the road. When the actual traveling distance is obtained by integrating the distances, the accumulation of errors cannot be ignored, and there is a problem that the calibration accuracy of the distance conversion constant of the distance sensor output decreases.

In the present invention, in consideration of such points, the influence of map data error is small. An object of the present invention is to provide a position detection device that can perform map matching with high calculation speed and accuracy. Further, the present invention provides a vehicle position detecting device that corrects a distance sensor output while taking into account the positioning accuracy by an absolute position calculating means such as GPS and a sign post that directly obtains the latitude and longitude of the vehicle, and calculates the vehicle position with high accuracy. The purpose is to

[0010]

In order to achieve the above object, the present invention provides a direction sensor for detecting a traveling direction of a vehicle, a distance sensor for detecting a traveling distance of the vehicle, a traveling direction from the direction sensor, and A current position estimating means for estimating the current position of the vehicle with respect to the reference position using the traveling distance from the distance sensor, a map storing means for storing road data, and an error amount of the road data stored in the map storing means Then, a map range selecting means for determining an evaluation range, a map matching calculation is performed using the current position calculated by the current position estimating means and the evaluation range selected by the map range selecting means, and the current position on the map road is calculated. A vehicle position detecting device comprising: a map matching calculation means for correcting a position; and an output means for outputting the current position of the vehicle corrected by the map matching calculation means.

Further, an azimuth correcting means is provided, and the azimuth correcting means respectively obtains the amount of change in the heading of the vehicle and the heading of the road data before and after the turn which is matched by the map matching calculation means.
The vehicle position detecting device is characterized in that the traveling direction is corrected when the difference is smaller than a predetermined value.

Further, a bend road determining means is provided, and the bend road determining means determines whether or not the vehicle is traveling on the bend road, and when it is determined that the vehicle is traveling on the bend road, the allowable error amount in the map matching calculation is increased. Is a vehicle position detecting device.

When the number of roads in the road data within a predetermined range including the current position is less than or equal to a predetermined value and the movement locus of the vehicle calculated by the current position estimating means is bending, It is a vehicle position detecting device characterized by increasing an allowable error amount at the time of map matching calculation on the assumption that the vehicle is currently traveling on a curved road.

Further, the curved road determination means is provided when the number of intersections in the road data within a predetermined range including the current position is equal to or less than a predetermined value and the movement locus of the vehicle calculated by the current position estimation means is curved. The vehicle position detecting device is characterized in that the vehicle is currently traveling on a curved road.

Further, an azimuth sensor for detecting a traveling azimuth of the vehicle, a distance sensor for detecting a traveling distance of the vehicle, an output of the azimuth sensor, and a current position estimating means for estimating a current position from the output of the distance sensor. A map storage unit that stores road network data; a map matching calculation unit that matches the current position estimated by the current position estimation unit with a point on the road network stored by the map storage unit; The temporary position used when the calculating means matches the current position, the temporary position calculating means for calculating using the road network data, and the output means for outputting the current position matched by the map matching calculating means, The map matching calculation means recognizes the temporary position calculated by the temporary position calculation means as the current position when approaching the curve or when exiting the curve. To calculate the travel direction of the Teru point,
Vehicle position detection characterized by matching the current position by selecting a provisional position and a point recognized as the current position that is most suitable as the current position by comparing with the output of the direction sensor It is a device.

Azimuth calculating means for calculating the traveling azimuth of the vehicle, a distance calculating means for calculating the moving distance of the vehicle, a map storing means for storing map data, the azimuth calculating means, the distance calculating means, and the map storing. A current position estimating means for estimating the position of the vehicle using the output of the means, an output means for outputting the vehicle position obtained by the current position estimating means, and an absolute position calculating means for calculating the vehicle position by latitude and longitude. A position verification means for determining the possible existence range of the vehicle based on the absolute position calculated by the absolute position calculation means and for verifying the inclusive relationship between the estimated position and the possible existence range calculated by the current position estimation means; By the means, it is verified that the estimated position is included in the possible range at least at the start point of the distance calibration section and not at the end point, and the distance calculation error is made from the estimated position at the end point and the possible range. Calculated, a vehicle position detecting apparatus being characterized in that the distance calibration section length and the distance calculation error with a distance constant correcting means for correcting the distance conversion constant of the distance computing means output.

Further, there is provided a road length calculating means for calculating a moving distance of the vehicle as an integrated value of road lengths using the map data of the map storing means, and the distance constant correcting means is calculated by the road length calculating means. The vehicle position detecting device is characterized by correcting the distance conversion constant of the output of the distance calculating means by using the road length.

Further, the absolute position calculating means uses a global positioning system, and instead of the position checking means, satellite combination judging means for judging the combination of satellites used by the absolute position calculating means to calculate the absolute position. And a straightness determining means for determining the straight traveling state of the vehicle, wherein the distance constant correcting means is a satellite combination determining means for maintaining a fixed satellite combination,
And correcting the distance conversion constant from the distance between the absolute positions of the vehicle determined by the absolute position calculation means when the straightness determination means determines that the vehicle is in the straight traveling state and the integrated output value of the distance calculation means. Is a vehicle position detecting device.

Further, the vehicle is provided with a vehicle speed calculating means for calculating a moving speed of the vehicle, and the distance constant correcting means converts the distance of the output of the distance calculating means in a distance calibration section in which the vehicle speed obtained by the vehicle speed calculating means is a certain value or more. It is a vehicle position detecting device characterized by correcting a constant.

Further, there is provided road type discriminating means for identifying the road type of the traveling road by using the map storing means, and the distance constant correcting means is a distance calibration discriminated as a highway or a toll road by the road type discriminating means. In the vehicle position detecting device, the distance conversion constant of the output of the distance calculating means is corrected in a section.

[0021]

With the above structure, the error amount of the road data is estimated, and the pattern matching with the locus is performed by using the evaluation range selected from the portion having the small error amount. It is possible to detect various positions.

Further, since the azimuth of the azimuth sensor is corrected from the azimuth of the road data by using the selected road data having a small error amount, accurate azimuth correction can be performed, and the position detection accuracy can be improved.

Further, the error of the road data is particularly large,
Accurate position detection can be performed in order to detect a mountain road or a curved road where the error of the sensor data becomes large due to the road gradient and enhance the pulling of the map matching to the road.

In addition, in the curve, since the temporary position is set and the most appropriate current position is selected from the temporary position and the current position to update the current position, the map matching in the curve can be more accurately performed. You can do it without any discomfort.

Further, whether the absolute position calculated by the absolute position calculating unit in the position verification unit is treated as a possible existence range considering the error range, not as a point, and whether the position calculated by the current position estimation unit is included in the range. In order to correct the distance conversion constant by the distance constant correction means by calculating the distance calculation error from the position obtained by the current position estimation means and the possible range of the vehicle after traveling to some extent from the verified position as the starting point, It is possible to calculate the position of the vehicle while calibrating the output of the distance calculating unit with high accuracy without accumulating the positioning error by the absolute position calculating unit.

Further, since the moving distance of the vehicle is calculated by adding the road length of the map data using the road length calculating means, the distance calculation accuracy is high and the distance constant correcting means corrects the distance conversion constant from the value. Therefore, it is possible to calculate the position of the vehicle while calibrating the output of the distance calculating unit with high accuracy without accumulating the positioning error by the absolute position calculating unit.

From the position obtained by the absolute position calculating means when the straightness judging means judges that the vehicle is traveling straight and the satellite combination judging means judges that the combination of satellites used for positioning is constant. To calculate the travel distance of the vehicle,
It is possible to calculate the position of the vehicle while calibrating the output of the distance calculating means with high accuracy and less error accumulation in calculating the moving distance of the vehicle as compared with the case where the combination of satellites is not limited.

[0028]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 4 is a hardware configuration diagram of a vehicle position detecting device applied to the first to third embodiments of the present invention. Reference numeral 401 denotes a high-accuracy direction sensor, which uses an optical fiber gyro (hereinafter referred to as an optical gyro) in this embodiment. Other than this, for example, a vibration rate gyro or a gas rate gyro may be used. A distance sensor 402 outputs a pulse signal for each unit traveling distance according to the rotation of the tire. A road data storage device 403 is, for example, a CD-ROM that stores road data and a CD that reads the data.
A ROM player is used. Reference numeral 404 denotes a microcomputer having an I / O for reading sensor data and road data by an arithmetic processing unit. Reference numeral 405 is a display device such as a display.

FIG. 1 is a block diagram of a vehicle position detecting device applied to a first embodiment of the present invention.

In FIG. 1, reference numeral 101 denotes a high-accuracy direction sensor, and an optical gyro is used in this embodiment. A wheel speed sensor and an axle sensor 102 are used as a distance sensor. 1
Reference numeral 03 is a current position estimating means, which calculates the current position of the vehicle with respect to the reference position based on the outputs from the azimuth sensor 101 and the distance sensor 102. A map storage unit 104 stores road data described on the map. 10
Reference numeral 5 denotes a map range selection means, which selects a map range to be used for map matching calculation from a portion having a small error in road data. Reference numeral 106 denotes a map matching calculation means, which is a trajectory formed by the current position of the vehicle calculated by the current position estimation means 103 and the current position from the past, and the map range selection means 1
Pattern matching is performed with the road data within the map range obtained in 05 to correct the current position of the vehicle on the road. An output unit 107 displays the current position of the vehicle calculated by the map matching calculation unit 106 on a display or the like mounted on the vehicle.

The operation of the vehicle position detecting device of the first embodiment constructed as described above will be described below.
Although the present invention can be realized by hardware, in the present embodiment, a case of processing by software using a microcomputer or the like will be described.

In the first embodiment, the error amount of the road data is estimated, the map range is selected from the portion having a small error amount, and it is not affected by the error of the road data in order to match with the locus.
The purpose is to perform highly accurate position detection.

FIG. 5 is a flow chart showing the procedure of position detection in the first embodiment, and the operation will be described according to this. It is necessary to set the initial position when installing the device,
The coordinates of the position of the vehicle are manually input or set by external information such as radio navigation, but once set, it is not necessary to re-set by storing the position where the vehicle stopped last time. , Normally, setting the initial position is unnecessary. Similarly, when using a sensor that detects the turning angle of the vehicle such as an optical gyro or a vibration rate gyro as the azimuth sensor, once the azimuth is set manually or by external information, the absolute azimuth setting is usually unnecessary. ..

In step 501, the traveling direction and traveling distance of the vehicle are detected every time the vehicle travels a unit distance (for example, 5 m). When an optical gyro is used for the direction sensor, the traveling direction D
Is D = D '+ Ta (1). D'is the absolute azimuth required by the previous time, T
a is a turning angle detected by the optical gyro while traveling a unit distance.

In the next step 502, the estimated position of the vehicle is calculated by the following equation using the estimated position of the vehicle obtained up to the previous time as the reference position.

X = X ′ + LcosD (2) Y = Y ′ + LsinD (3) where X, Y is the estimated position coordinate of the vehicle X ′, Y ′ The estimated position coordinate of the previous vehicle L is the traveled distance D is the heading. In step 503, the estimated position coordinates are sequentially stored, and a straight line is drawn between the respective coordinates to create a traveling locus. As will be described later in detail, since the road data is obtained by linearly approximating the shape of the road, the straight part of the road is generally more accurate than the curved part. Therefore, a straight traveling portion is required in the traveling locus to match the road data, and it is determined in step 504 whether the vehicle is in the straight traveling state. For example, when the change angle of the heading in the section of 50 m is 5 ° or less, the straight traveling state is set. If the vehicle is in the straight traveling state, the process proceeds to step 505 and thereafter, and if it is not in the straight traveling state, the estimated position of the vehicle is displayed in step 513, and the one-time process is ended. Step 505
Then, the road data used by the map matching calculation means 106 is read. The road data is La (for example, 5
It should be within 0 m) and closest to the estimated position. In the map matching calculation means 106, 1 is placed between the two straight sections.
When there are more than one turn, the main processing is to calculate the correlation between the running trajectory and the road data and set the position on the road.
If the vehicle is simply going straight, the current position on the road is advanced. Therefore, in step 506, it is determined whether or not a turn has been made within a certain distance in the past (for example, 200 m). This is judged by, for example, whether the change angle of the traveling direction in the section is 10 ° or less. If there is a bend, move to step 507,
If not, the current position on the road currently being traveled is moved by the travel distance in step 512, and one processing is ended.

In the present embodiment, the map matching calculation is performed using the points on the road data, and in step 507, the process of selecting the map evaluation point from the map range, which is the portion of the road data with few errors, is performed. First, the error of the road data will be described.

In the case of Japan, the road data is based on a topographic map issued by the Geospatial Information Authority of Japan, and the shape of the road shown on the map is approximated to a straight line, and each straight line is expressed in the form of a vector C
It is stored in a map storage medium such as a D-ROM.

There are various storage methods, such as using a vector as a line segment and storing the start point and end point of the vector, but all are the same in that the road is approximated by a straight line. In addition, even if it is a straight line approximation, it can be approximated to an accurate shape by approximating it with a very short straight line, but since the data amount increases, the data amount and accuracy are actually compromised with a practical line. .. A problem that occurs when a road is linearly approximated will be described with reference to FIG. For example, when the road has a shape as shown in FIG. 6A, the road data may be as shown in FIG. 6B because it is created from the center line of the road. On an actual road, when it is possible to reach from point A to point E with one gentle right turn, the traveling locus of the vehicle and the shape of the road data greatly differ between B and C of the road data. The error adversely affects the accuracy and stability of the map matching process. For example, since the method shown in the conventional example considers a plurality of current position candidates, if there is an error in the correct road, the order of the candidates is erroneously determined, and the calculation of the current position is erroneous.
Therefore, in step 507, it is determined that the error amount of the road data that satisfies the following two conditions is small.

However, one piece of road vector data is called a road segment. 1) The length of the road line segment is longer than Ld (for example, 100 m) 2) The difference between the direction of the target road line segment and the direction of the road line segments connected before and after it is less than a predetermined value (for example, 20 °) Since the coordinates and orientation of the midpoint of the line segment are considered to be the highest among the road line segments, the midpoint of the road segment that satisfies the above conditions is used as the map evaluation point. Next, in step 508, it is determined whether two map evaluation points have been obtained before and after the turn. If it is obtained, the map matching calculation is performed in step 509, and if it is not obtained, the estimated position is output in step 513, and the one-time process is ended.

The map matching calculation method in step 509 will be described below with reference to FIG. FIG. 7A is an example of road data, and the line segment connecting points is the road segment described above. FIG. 7B shows a travel locus when the vehicle travels from A to F on such road data (S1, S).
(2 is a straight section before and after the turn) At this time, in the conventional method, as shown in FIG.
The error between CD of (a) becomes a problem. Step 50
The map evaluation points obtained in 7 are M1 and M2 in FIG. 7C, and the azimuth at M1 is θ. Now, a point on the straight-ahead portion S1 before the turn of the traveling path is made to coincide with the coordinate of M1. If the error in the traveling direction is large, the running locus may be matched with M1, and at the same time, the track may be rotated so that the azimuth at the point of matching with M1 is θ. This makes it possible to determine the absolute coordinates of each point on the travel locus. The shortest distance d between M2 and the locus at this time is obtained (see FIG. 7D). The point of the straight part S1 to be matched with M1 is within a predetermined range SA (for example, 200 m)
The distance is shifted by a predetermined distance (for example, 10 m), and d is obtained again. Then, when the minimum d is given, the coordinate of M1 is given to the point that coincides with M1 to determine the tip of the traveling locus, that is, the coordinate of the current position.

At step 510, the azimuth at M2 is compared with the azimuth of the running locus separated by d from M2, and when the difference in azimuth is less than Da (for example, 10 °), at step 511, M2 is moved from the tip of the running locus. A perpendicular line is drawn to the road segment that includes or connects to M2, and the foot is displayed as the current position. When the difference in azimuth is equal to or larger than the predetermined value, the estimated position is displayed in step 513, and the one-time process ends.

As described above, according to this embodiment, since the portion of the road data having a large error amount is not used for the matching calculation, it is possible to stably detect the current position.

Next, a second embodiment of the present invention will be described. FIG. 2 shows a block diagram of this embodiment. In this figure, the same parts as those in FIG. 1 are designated by the same reference numerals. In this embodiment, in addition to the first embodiment, the azimuth correcting means 108 is provided, and the absolute azimuth is corrected from the result of the matching calculation by the map matching calculating means to eliminate the accumulation of the azimuth error, and highly accurate azimuth calibration. The purpose is to realize. Next, the operation of this embodiment will be described with reference to FIG.

Since the optical gyro has a very small drift and a small scale error, the relative azimuth in a section of several 100 m can be accurately detected, but the absolute azimuth is calculated because the turning angle of the vehicle is integrated to calculate the absolute azimuth. Accumulation of error is unavoidable. Although it is possible to set the absolute azimuth using the azimuth of the road data, the accuracy of the turning angle by the optical gyro is very high, so if you set the absolute azimuth from the azimuth of the road data that contains a large azimuth error. Conversely, the azimuth accuracy will be reduced. Therefore, by estimating the error amount of the road data and using only the data with a small error amount, it is possible to set the absolute azimuth with high accuracy. Here, the relative azimuth between the evaluation points on the road data and the relative azimuth by the optical gyro are compared, and when the two agree well, the absolute azimuth is corrected using the azimuth of the road data.

Since the flow of processing is the same as that of the first embodiment, it will be described with reference to FIG. The correction of the azimuth is performed when the map matching calculation is performed in step 509 used in the description of the first embodiment. The operations in other steps are the same as those in the first embodiment. FIG. 8A is an example of road data,
(B) is a traveling locus when traveling on the road. Points on the locus that match the map evaluation points M1 and M2 before and after the turn by the above-described matching calculation are set to m1 and m2. In this embodiment, the right and left turns are made three times between the evaluation points. However, even if only one turn is made, the azimuth can be corrected by the same process. When the trajectory and the road data are thus matched, the azimuth difference θmap at the evaluation points M1 and M2 is calculated as θmap = θM1−θM2 (4). Similarly, regarding the traveling locus, m1 and m
The azimuth difference θloc at 2 is θloc = θm1−θm2 (5). Here, | θmap−θloc | <Ea (6) However, Ea corrects the current traveling azimuth D by the following formula when the condition of the threshold (for example, 5 °) of the azimuth correction angle is satisfied.

D = D '+ α * (θM2-θm2) (7) where α is a correction coefficient, and is a number from 0 to 1, for example.
From 0.5 onward, the corrected absolute azimuth D is used to perform traveling locus creation, map matching, etc. according to the flow described in the first embodiment, and position detection is performed.

As described above, according to the present embodiment, after selecting a portion having a small error amount of road data, a change amount of the direction of the road data before and after the turn and a change amount of the direction of the trajectory are respectively obtained, and the difference is small. Since the azimuth is corrected sometimes, the azimuth can be corrected with high accuracy.

In this embodiment, the azimuth is corrected every time the map matching is performed. However, the azimuth error is estimated from the lapse time since the previous azimuth correction and the total of the turning angles, which leads to the absolute azimuth error. However, the azimuth correction described in this embodiment may be performed only when the estimated error value exceeds a predetermined value (for example, 10 °).

Next, a third embodiment of the present invention will be described. FIG. 3 shows a block diagram of this embodiment. In this figure, the same parts as those in FIG. 2 are designated by the same reference numerals. In this embodiment, in addition to the second embodiment, a bend road judging means 109 is provided, and it is judged from the current position by the current position estimating means 103 and the road data of the map storing means 104 that the error amount is particularly large, Map matching calculation means 10
By changing the error allowable amount at the time of matching calculation in 6 above, it is an object to perform highly accurate position detection even when the error amount of road data increases. Next, the operation of this embodiment will be described.

As described above, the road data is created by the straight-ahead approximation, and the shape of the curve is greatly changed in order to suppress an increase in the amount of data. This will be described with reference to FIG. FIG. 9A shows the actual shape of the road on a curved road.
(B) is an example of road data representing the road. On a road with few straight lines, the actual road shape and the shape of the road data are very different. When traveling on such a road, a large deviation occurs between the traveling locus and the position and direction of the road data, so that it is difficult to align and the position detection accuracy when aligned is lower than that of a general road. Further, in the first embodiment, since the matching calculation is performed using the straight line portion of the road data which is estimated to have a small error in the road data, there is little opportunity to perform the matching calculation on a curved road or a mountain road with many curves. further,
On mountain roads, due to the slope of the road and the rolling of the car body due to sudden turns, the traveling direction due to the optical gyro causes an error that is several times that during normal driving. Normal map matching cannot be performed because the amount of error between the road data and the running locus increases, but at this time it is considered that it would not make the user feel uncomfortable if the current position is obtained on the road. Further, if the same matching calculation as usual is performed, the current position is obtained at a point far away from the correct position due to the error of the traveling direction obtained from the optical gyro. Therefore, it is necessary to strengthen the pulling in to the road by, for example, increasing the allowable error amount at the time of matching calculation.

Since the flow of processing is the same as that of the first embodiment, it will be described with reference to FIG. Step 5 for determining the bend
It is performed immediately before reading the road data of 05, and when it is determined that the vehicle is currently traveling on a curved road, by changing each constant described later, it is possible to strengthen the pulling into the road by the same processing as usual. That is, the operations in other steps are the same as those in the first embodiment. A method for determining whether the road currently running is a bend road or a mountain road will be described below.

In step 505, road data existing within a certain range (for example, 200 m) including the current position calculated by the current position estimating means is obtained from the map storing means. FIG. 10 shows an example of road data included in the range.
10A is an example of road data in an urban area, and FIG. 10B is an example of road data on a curved road. D1 to D7 are road names
(Or number), and the black circle is the end point of a simple road segment
(Connection point), and the white circle is the end point of the road line segment indicating the intersection. D1 → D6 → D2 → D7 → D3 on the road of FIG.
The traveling locus when traveling on → D6 → D4 is assumed to be exactly the same as traveling on the road in FIG. 10 (b).
Therefore, since it is impossible to determine the curved road based on the traveling locus, the curved road is determined as follows using the road data.

In the case of road data in which a road line segment group forming one road is stored as one road, the number of roads (the number of D) within the above range is checked, and the number of roads is a predetermined value ( It is detected that the vehicle is traveling on a curved road or a mountain road when the sum of absolute values of the azimuth change amount of the traveling locus is equal to or more than a predetermined value (for example, 360 °).

Alternatively, if the number of intersections in the road data within the above range is less than or equal to a predetermined value (for example, 4 points), and the total sum of absolute values of the azimuth change amount of the traveling locus is at least a predetermined value (for example, 360 °). It is detected that the vehicle is traveling on a curved road or a mountain road at a certain time.

When the bend road or the mountain road can be detected as described above, the pulling into the road is enhanced by the following method.
The values of the respective constants before the change are those shown in the first embodiment. (1) Expansion of La when searching for road data around the estimated vehicle position
(50m → 150m) (2) Shorten Ld of the condition to select the evaluation point
(100m → 50m) (3) Enlargement of drawing direction error Da during matching calculation
(10 ° → 20 °) After changing each constant like this, read the road data.
(Step 505), evaluation point selection (step 507), and map matching (step 509) are processed in the same flow as in the first embodiment, so that the pulling of the current position onto the road can be strengthened. it can.

As described above, according to this embodiment, it is determined that the error amount of the road data is particularly large on the curved road or the mountain road, and the error allowable amount at the time of the matching calculation is changed to the road. Since the pull-in is strengthened, the stable current position can be detected even if the error in the road data and the heading increases.

In this embodiment, the fact that the sum of the absolute values of the change angles of the traveling azimuth detected by the azimuth sensor mounted on the vehicle is equal to or greater than a predetermined value is a part of the detection conditions for the curved road or the mountain road. However, the sum of absolute values of the direction change amounts of road line segments used for map matching may be used instead.

Next, a fourth embodiment of the present invention will be described. When performing map matching, a particular problem is where there are many curves such as mountain roads. In the present embodiment, the error in the curve is taken into consideration, and the map matching without any discomfort in the curve is realized. FIG. 11 shows a block diagram of this embodiment. Also in this figure, the same parts as those in the above embodiment are designated by the same reference numerals. In this embodiment, in addition to the first embodiment, the temporary position calculating means 11
0 is provided, and the map matching calculation unit 106 calculates a temporary position by the temporary position calculation unit 106 when approaching a curve or when exiting from a curve, and advances between this temporary position and a point recognized as the current position. By matching the current position by comparing the azimuths, it is possible to realize map matching without any discomfort even in a curve.
The present embodiment will be described in detail below. FIG. 13 is a flowchart showing the operation of this embodiment. In FIG. 13, the operation from step 1301 to step 1305 is exactly the same as that of the above-mentioned embodiment, so the description thereof is omitted here. First, in step 1306, it is determined whether a temporary position exists. If the temporary position exists, the process proceeds to step 1312, and if the temporary position does not exist, the process proceeds to step 1307. In step 1307, it is determined whether there is a curve ahead. This judgment is made using the data in the map storage means 104. The map matching calculation unit 106 can detect the presence of a curve in advance because the road data up to 50 m ahead is read in advance.
In this embodiment, it is determined whether or not there is a curve within 50 m ahead. Further, in the present embodiment, the determination as to whether the curve is a curve is made based on whether or not the azimuth of the road segment ahead is turning by 60 degrees or more. The value of the turning angle may be changed depending on, for example, the accuracy of the map. If there is a curve that turns 60 degrees or more within 50 m in the front in step 1307, proceed to step 1308,
Create a temporary position. The temporary position will be described with reference to FIG. In FIG. 14 (a), A is a point that is map-matched as the current position, and indicates that the curve is about to be reached. Points a and b existing before and after A are temporary positions. In the present embodiment, this temporary position is created at a point 15 m before and after the current point A. The position where the temporary position is created may also be changed depending on the accuracy of the map, for example. Further, the number of temporary positions is not limited to two and may be one or three or more.

When the temporary position is created in step 1308,
In step 1312, evaluation values are calculated for each of the created temporary positions a and b and the current position A. This evaluation value represents how much the direction of the point and the direction of the actual output of the direction sensor are deviated.
It is calculated using the correlation function shown in. FIG. 12 is an example of a function for determining the evaluation value from the orientation error. This function may be another function as long as it satisfies the following properties.

A monotonically increasing function.・ Ignore the slight azimuth error.

To enhance the sensitivity near the azimuth error which is actually large. When the evaluation values can be calculated for each of the provisional positions a and b and the current position A by integrating the values obtained by the function having the above properties with each evaluation value, these three evaluation values are normalized. The normalization method may be, for example, normalization based on the smallest one of the three evaluation values (the smallest one is set to 1). Next, the process proceeds to step 1313, and the evaluation values are too different (for example, 1
(0 or more) is deleted. Next, the process proceeds to step 1314, and the provisional position a, b and the current position A, whichever is closest to the output of the azimuth sensor, is newly updated as the current position. The state of this updating will be described with reference to FIGS. 14, 15 and 16. In these figures, the solid line shows the road data and the broken line shows the actual road.
FIG. 14 shows a case where the map data matches the actual road, and FIGS. 15 and 16 show a case where the map data includes an error and deviates from the actual road.
Further, in each drawing, (a) represents a state in which the vehicle is approaching a curve and a temporary position is generated.
(B) shows a state in which the vehicle has actually started to turn a curve, and (c) shows a state in which the vehicle has finished turning a curve.

In FIG. 14, when the vehicle approaches a curve, temporary positions a and b are generated before and after the current position A, as shown in (a). Then, the evaluation value e is calculated for each of the current position A and the temporary positions a and b. The change in the evaluation value with respect to the distance traveled after the curve starts is shown at the bottom of each of FIGS. 14, 15 and 16. In FIG. 14, the road data stored in the map storage means 104 does not include a fatal error and is equal to the actual road, so that the current position A is always displayed from the time when the curve is approaching until the exit from the curve is completed. The evaluation value is smaller than the other temporary positions. Therefore, in FIG. 14, the replacement of the temporary position and the current position does not occur. In FIG.
The data stored in the map storage unit 104 includes an error and is stored as a curve having a diameter larger than that of the actual road. When the vehicle travels along a curve in FIG. 15 as well, as can be seen from the graph of the evaluation value, the azimuth indicated by the current point A is actually at the stage of approaching the curve (state b in the figure). The evaluation value deviates from the azimuth output by the azimuth sensor. This is because the road data contains an error. Further, in this case, since the temporary position a has a smaller evaluation value than the current position A, the temporary position a is replaced as the current position. The same applies to the case shown in FIG. In FIG. 16 as well, similar to FIG. 15, the data stored in the map storage means 104 includes an error and is stored as a curve having a smaller diameter than the actual road. When the vehicle travels along a curve in the same manner in FIG. 16, the temporary position b has a smaller evaluation value than the current position A, so that the temporary position is replaced with the temporary position b. When the temporary positions are replaced in step 1314, it is determined in step 1315 whether or not it is time to integrate the temporary positions. Here, the integration of temporary positions will be described. As described above, in steps 1307 and 1308, since temporary positions are generated each time the curve is reached, temporary positions are created one after another when the curves are continuous, such as on a mountain road. There will be more temporary positions than. Therefore, it is necessary to erase the temporary position generated every time the vehicle travels to some extent. The operation required for this is integration. In this embodiment, the integration of temporary positions is 2
It is supposed to be integrated every time the vehicle travels 00m.

If it is determined in step 1314 that it is the integration timing, the temporary positions are integrated in step 1316, and the flow advances to step 1310. In step 1310, it is determined whether or not the map matching is successful, and if successful, the process proceeds to step 1311 to output the position on the road stored in the map storage means 104. If the map matching is unsuccessful, the process proceeds to step 1309, and the estimated position calculated by the current position estimating unit 103 is output to the output unit.

This completes the series of operations, and then proceeds to step 1309 to repeat the above operations to perform map matching.

As described above, according to this embodiment, the temporary positions a and b are set on the curve, and the most appropriate one of the temporary positions a and b and the current position A is selected as the current position. Since the current position is updated, map matching on a curve can be performed more accurately and comfortably.

In this embodiment, the temporary position is generated before the curve is reached, but the timing for generating the temporary position may be after the curve is actually reached or when the vehicle exits the curve. You may make it generate | occur | produce a temporary position.

Next, the present invention and the fifth embodiment will be described. FIG. 20 shows a basic structure of a vehicle position detecting device common to the fifth to the fifth embodiments of the present invention. In FIG. 20, a distance sensor 2001 is, for example, an axle sensor or a wheel speed sensor, and is used for each unit distance travel (for example, 50
pulse every cm. The moving distance is calculated by integrating the number of pulses and multiplying by a pulse-distance conversion constant (hereinafter referred to as a distance constant). Reference numeral 2002 is an azimuth sensor.
This may be, for example, a geomagnetic direction sensor that detects the horizontal component force of the geomagnetism to calculate the absolute direction of the vehicle, or it may calculate the angular velocity of the vehicle, calculate the turning angle from the integrated value, and add it to the reference direction. A rate sensor (optical gyro, gas gyro, vibration gyro, etc.) that determines 200
Reference numeral 3 denotes a map storage means, for example, a CD-ROM disk as a storage medium for storing map data including road information,
A CD-ROM player is used as a driving device. 20
Reference numeral 04 is a GPS antenna, and 2005 is a GPS receiver. Reference numeral 2006 denotes an arithmetic processing unit, which is a microcomputer system including a microprocessor, a memory, an I / O and the like. Reference numeral 2007 is a display device, and an LCD, CRT or the like is used.

FIG. 17 shows a block diagram of the vehicle position detecting device of the fifth embodiment of the present invention. In FIG. 17, 1701
Is a distance calculation means, which is composed of a distance sensor, calculates the moving distance of the vehicle from the number of pulses, converts it into distance information, and outputs it. Reference numeral 1702 denotes an azimuth calculation means, which is composed of an azimuth sensor and outputs the traveling azimuth of the vehicle as azimuth information. 1703
Is a map storage means, 1704 is a present position estimation means, 170
5 is output means, 1706 is absolute position calculation means, 1707
Is a position verification means, and 1708 is a distance constant correction means.

The operation of the vehicle position detecting device of the present embodiment constructed as above will be described below. Although the present embodiment can be configured by hardware, a case where it is realized by software using a microprocessor will be described here. The purpose of this embodiment is to detect the vehicle position while correcting the distance conversion constant of the distance sensor with high accuracy by eliminating the influence of the positioning error of the vehicle position calculated by the absolute position calculating means as much as possible.

In the present embodiment, the traveling locus of the vehicle is obtained from the moving distance / traveling direction of the vehicle obtained by the distance sensor / azimuth sensor with reference to the initially set position, and from the correlation between the traveling locus and the map data. The current position shall be estimated. Various methods have been devised for estimating the current position. For example, the method disclosed in JP-A-61-56910 may be used. The position calculation is performed at predetermined intervals (for example, every 2 m running), and the calculated own vehicle position is sent to the output means.
The calculation and output of the vehicle position are preferentially executed, but in addition to this, automatic calibration processing of the distance sensor is also executed in parallel.

FIG. 21 is a flowchart showing the procedure for automatically calibrating the distance sensor. First, in step 2101, initialization processing of each variable used for automatic calibration is performed. Next, in step 2102, it is determined whether GPS positioning is possible. If the positioning cannot be performed, or if the positioning is successful but the satellite arrangement is bad (PDOP value is large, for example, 3 or more), the process returns to step 2101. In other cases, the positioning is possible and the process proceeds to step 2103. Step 210
In Fig. 3, the possible existence range expected by GPS positioning is calculated, but the positioning performance by GPS is 100 m when various error factors are combined when using a C / A code that can be normally used.
Since it is considered to be a degree, in the present embodiment, a circular area having a radius of 100 m with the positioning position as the center is set as the existence possible range by GPS positioning. In the next step 2104, it is verified whether the estimated position of the vehicle is included in the range obtained in step 2103. If it is included, the estimated position is set as the starting point of the distance calibration section, but if it is not included, the process returns to step 2101 and the process is repeated from the beginning.

In the next step 2105, the data of the direction sensor / distance sensor is collected, and in step 2106, the estimated position of the vehicle is calculated and stored. And step 2107
Then, the number of pulses L corresponding to the moving distance when the estimated position verified in step 2104 is used as the starting point is obtained by integration. In step 2108, L is compared with a predetermined value L ₀ . If the comparison result is L <L ₀ , the processing from step 2105 is repeated, and if the condition of L ≧ L ₀ is satisfied, the processing proceeds to step 2106. L ₀ is determined by the target calibration accuracy of the distance sensor and the GPS positioning accuracy, but here L ₀ = 20 km. If this value, step 2
The GPS positioning error radius of 100 m assumed in 103 is 0.5
This corresponds to an error amount of%.

When it is confirmed in step 2108 that the vehicle has traveled the predetermined distance L ₀ or more, in steps 2109 to 2111 the latest estimated position of the vehicle is compared with the possible existence range by GPS positioning. The method is step 2102.
From step 2104 is the same. However, if GPS positioning is not possible, the processing from step 2105 is repeated, and if the estimated position is included in the possible existence range by GPS positioning, it is determined that the accuracy of the distance sensor is practically sufficient, and step 2101 is performed. Return to. Only when the estimated position of the vehicle (that is, the end point of the distance calibration section) is outside the GPS positioning range, it is considered that the distance constant has an error, and the calibration process is started.

First, in step 2112, the calculation error due to the distance constant used so far is obtained. FIG. 22 is an explanatory diagram thereof. In FIG. 22, M1 is the start point of the distance calibration section obtained by the current position calculation means, and is verified in the possible range (G1) centered on the GPS positioning point.
If the distance constant is smaller than the correct value, the end point M2 of the distance calibration section obtained by the current position calculation means is (a).
Existence possible range (G2) centered on GPS positioning point like
And the distance constant is large, (b)
It goes beyond G2 and goes too far. Therefore, step 211
In step 2, first, the history of estimated positions (that is, the locus) obtained in step 2106 is used to test whether or not there is a position included in G2 within a predetermined distance range (for example, 500 m). When it is included, the distance between the boundary of G2 and the intersection of the traveling road and M2 is set as the distance error ΔL (<0) as shown in FIG. 6B. On the other hand, when it is not included, the distance from the start point of M2 until reaching G2 is set to ΔL (> 0). Then, in step 513, the distances L and ΔL between M1 and M2 and the distance constant K
The distance constant is modified from d'as follows.

Kd = Kd ′ × (L + ΔL) / L (8) If a new distance constant Kd is calculated in step 2113, the process returns to step 2101 to repeat the processing.

As described above, according to the fifth embodiment, the positioning position obtained by the absolute position calculating means is used not as a point but as an area including an error, and the start and end points of the calibration section are calibrated for the distance sensor calibration. Since it is used only for position verification, it is possible to eliminate the accumulation of positioning error as compared with the case where the moving distance of the vehicle is calculated directly from the distance between positioning positions, and the distance conversion constant of the distance sensor can be calculated with high accuracy. The vehicle position can be detected while calibrating.

Although the GPS is used as the absolute position calculating means in the fifth embodiment, a sign post may be used instead. Also, although only the length condition was added to the road used for distance calibration, the moving speed of the vehicle is calculated using the output of the distance sensor, and the condition that the vehicle is traveling at a certain speed or more is added to meander the vehicle. The influence of error factors such as rotation at a small turning radius may be eliminated.

Next explained is the sixth embodiment of the invention. FIG. 18 shows a block diagram of this embodiment. As can be seen from the comparison between FIG. 17 and FIG. 18, all the constituent elements included in FIG. 17 are included in FIG. 18, and are given the same numbers. In FIG. 18, in addition to that, road length calculation means 1709
Is added. The operation of the vehicle position detecting device of this embodiment configured as described above will be described below. It should be noted that this embodiment also describes a case where it is realized by software as in the fifth embodiment. The purpose of the present embodiment is to obtain a moving distance calculation error of the vehicle due to a change in the output characteristic of the distance sensor from the map data using the road length calculating means, and calculate the vehicle position while calibrating the output of the distance sensor with high accuracy. That is.

FIG. 23 is a flowchart showing the procedure for automatically calibrating the distance sensor. Steps 2301 to 2305 and steps 2310 to 2314 are
Steps 2101 to 2105 of FIG. 21, respectively,
Which is equivalent to steps 2107 to 2111,
The calculation of the moving distance of the vehicle and the modification of the distance constant are different from those of the fifth embodiment. In this embodiment, the sensor data is acquired in step 2305 and the estimated position is calculated in step 2306 to match the map data. If the road can be identified as a result of the matching, steps 2307 to 2
Move to 308, but if not identifiable, step 230
Return to 1. In step 2308, the continuity of the road is determined, and if the newly calculated road is connected to the immediately preceding road, the process proceeds to step 2309, and if it is not connected, the process returns to step 2301. Then, in step 2309, the road length LM from the starting point is calculated as an integrated value of the values obtained from the map data. This will be described below with reference to FIG. When there is a road network as shown in FIG. 24 (a), the road is treated as a set of road line segments approximated by straight lines, and on the map data, for example, intersections / road bending points are shown as shown in FIG. 24 (b). A road is represented by the position of a node (for example, latitude / longitude coordinates) and the connection between the nodes centering on the node. Therefore, if the estimated position is obtained and the road line segment can be specified, the road length (road line segment length) can be calculated as the distance between the nodes, and the connection relation can also be referred to from the map data ( In addition to this, other information such as road types can be included in the map data).

Since the end point can be specified up to step 2314, the distance error is calculated in the next step 2315. The method of obtaining the error is almost the same as that of the first embodiment, except that the distance error is calculated by the road length obtained from the map data. Up to step 2315, the integrated value P of the distance pulse in the distance calibration section, the road length integrated value LM obtained from the map data, and the distance calculation error (but obtained from the road length) ΔLM are obtained. The distance constant Kd is modified as follows.

Kd = (LM + ΔLM) / P (9) If a new distance constant Kd is calculated in step 2315, the process returns to step 2301 to repeat the processing.

As described above, according to the sixth embodiment, the positioning position obtained by the absolute position calculating means is used not as a point but as a region including an error, and the distance sensor is calibrated by starting and ending the calibration section. Since it is used only for the position verification of and the travel distance of the vehicle is obtained from the map data using the road length calculation means, it is possible to eliminate the accumulation of positioning error when calculating the travel distance of the vehicle. The vehicle position can be detected while accurately calibrating the distance conversion constant of the distance sensor.

In order to further improve the accuracy of calculating the road length in the sixth embodiment, a road type discriminating means for discriminating the road type of the traveling road from the map data is added so that the road for which the distance is to be calibrated is a highway. , It may be limited to toll roads.

Next, a seventh embodiment of the present invention will be described. FIG. 19 shows a block diagram of this embodiment. As can be seen from FIGS. 17 and 19, the position verification means 170 of FIG.
Instead of 7, the satellite combination determination means 1710 and the straightness determination means 1711 are added.

The operation of the vehicle position detecting device of the present embodiment constructed as above will be described below. Note that this embodiment will describe a case where it is realized by software as in the case of the fifth embodiment. The purpose of this example is to
That is, the moving distance of the vehicle is accurately obtained by selecting and using the positions obtained by the absolute position calculating means, and the position of the vehicle is calculated while calibrating the output of the distance sensor with high accuracy.

FIG. 25 is a flow chart showing the procedure for automatically calibrating the distance sensor. Steps 2501 and 2502 are similar to steps 2101, 2102 in the fifth embodiment, and the combination of satellites used for GPS positioning (combination of SV numbers of GPS satellites) is stored in the next step 2503. In step 2504, sensor data is acquired, and in step 2505, the traveling direction is calculated. The first heading of the vehicle after the GPS positioning calculated here is used as a reference heading of whether or not the vehicle will travel straight thereafter. In step 2506, it is judged whether or not the vehicle is traveling straight ahead by comparing the traveling direction calculated in step 2505 with the reference direction. If the vehicle is not traveling straight ahead, step 2
If it is straight ahead, the process proceeds to step 2507 (if it is the first time, the process proceeds to step 2507). In step 2507, starting from the GPS positioning point in step 2503, the pulse output of the distance sensor thereafter is integrated (the integrated value is P). Steps 2508 and 2509 are rejection conditions, and when the elapsed time after GPS positioning in step 2502 is longer than a predetermined value T ₀ (for example, 2 minutes), the process returns to step 2501 and the moving distance from the start point is changed. If it is shorter than the predetermined value L ₀ (for example, 2 km), the process proceeds to step 2504.

If the condition that the elapsed time from the GPS positioning is within the predetermined time and the moving distance is equal to or more than the predetermined distance is satisfied, it is determined in step 2510 whether GPS positioning is possible. If possible, the process proceeds to step 2504. In step 2511, the combination of satellites used for GPS positioning in step 2503 is compared with the combination of satellites used in positioning in step 2511. And if there is no change in the satellite combination, step 2
The process moves to 512, and if there is a change, the process returns to step 2501. In step 2512, the moving distance L of the vehicle is calculated as the linear distance between the GPS positioning positions in steps 2502 and 2510. Then, in step 2513, the distance constant Kd is calculated from the moving distance L of the vehicle and the pulse integrated value P in the meantime.
Is corrected as follows.

Kd = L / P (10) If a new distance constant Kd is calculated in step 2513, the process returns to step 2501 to repeat the process.

As described above, according to the seventh embodiment, the traveling distance of the vehicle is calculated using the position where the vehicle is in a straight traveling state and the position is determined by the same combination of satellites. In GPS positioning, if the combination of satellites used for positioning is not limited, an error of about 100 m occurs even if the satellite arrangement is good, but if the combination of satellites is limited, the relative accuracy of the positioning position is improved by about one digit. .. Therefore, it is possible to invalidate the positioning error and calculate the moving distance of the vehicle, and it is possible to detect the vehicle position while calibrating the distance conversion constant of the distance sensor with high accuracy.

In the seventh embodiment, the output of the azimuth sensor is used to determine the straightness of the vehicle. However, the straightness may be determined from the variation in the loci of GPS positioning positions, and the steering angle may be determined. The determination may be made using another turning angle sensor such as a sensor. Further, in this embodiment, the condition that the combination of satellites is constant is set as a condition, but this is added to the condition that positioning can be continuously performed and the combination of satellites is constant to further reduce the relative positioning error. May be.

[0093]

As described above, according to the present invention, the error amount of road data is estimated, and the error amount of road data is estimated by the map range selection means that extracts the evaluation range from the portion with a small error amount. Since the pattern matching with the locus is performed by using the evaluation range selected from the portion having a small error amount, accurate position detection can be performed without being affected by the error of the road data.

In addition, the azimuth correcting means for correcting the azimuth of the azimuth sensor based on the result of the pattern matching with the data before and after passing the curve is selected from the azimuth of the azimuth sensor by using the road data having a small error amount selected. Therefore, the correct azimuth correction can be performed, and the position detection accuracy can be improved.

Further, the road data error is particularly large and the road gradient is large due to the bend road judging means for detecting that the vehicle is traveling on a bend road or the like in which the road data error is particularly large and changing the allowable amount of the map matching error. Therefore, accurate position detection can be performed in order to detect a mountain road or a curved road where the error of the sensor data becomes large and to strengthen the pulling of the map matching to the road.

In addition, since the temporary position is generated during the map matching on the curve and the temporary position and the current position are exchanged, even if the road data includes an error, it is more accurate. This makes it possible to detect the position without any discomfort.

Further, since the distance conversion constant of the distance sensor output, which changes due to a change in tire air pressure while the vehicle is running, is automatically calibrated using an absolute position calculating means such as GPS, the output of the distance sensor is It is possible to maintain high accuracy, and it is possible to improve the vehicle position detection accuracy.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a vehicle position detection device according to a first embodiment.

FIG. 2 is a block configuration diagram of a vehicle position detection device according to a second embodiment.

FIG. 3 is a block diagram of a vehicle position detecting device according to a third embodiment.

FIG. 4 is a hardware configuration diagram of the vehicle position detection device according to the first to third embodiments.

FIG. 5 is a flowchart illustrating the operation of the first embodiment.

FIG. 6 is a diagram showing an example of creating road data.

FIG. 7 is an explanatory diagram of a map matching calculation according to the first embodiment.

FIG. 8 is an explanatory diagram of direction correction processing according to the second embodiment.

FIG. 9 is a diagram showing an example of creating road data on a curved road.

FIG. 10 is a diagram showing an example of creating road data in an urban area and a curved road.

FIG. 11 is a block configuration diagram of a vehicle position detection device according to a fourth embodiment.

FIG. 12 is a block diagram of a vehicle position detecting device according to a fourth embodiment.

FIG. 13 is a flowchart for explaining the operation of the vehicle position detecting device according to the fourth embodiment.

FIG. 14 is an explanatory diagram of the operation of the fourth embodiment.

FIG. 15 is an explanatory diagram of the operation of the fourth embodiment.

FIG. 16 is an explanatory diagram of the operation of the fourth embodiment.

FIG. 17 is a block diagram of a fifth embodiment.

FIG. 18 is a block diagram of a sixth embodiment.

FIG. 19 is a block configuration diagram of a seventh embodiment.

FIG. 20 is a hardware configuration diagram of fifth to seventh embodiments.

FIG. 21 is a flowchart showing the operation of the fifth embodiment.

FIG. 22 is an explanatory diagram of the operation of the fifth embodiment.

FIG. 23 is a flowchart showing the operation of the sixth embodiment.

FIG. 24 is an explanatory diagram of the operation of the sixth embodiment.

FIG. 25 is a flowchart showing the operation of the seventh embodiment.

FIG. 26 is a block configuration diagram of a conventional vehicle position detection device.

[Explanation of symbols]

 101 Orientation Sensor 102 Distance Sensor 103 Current Position Estimating Means 104 Map Storage Means 105 Map Range Selecting Means 106 Map Matching Calculating Means 107 Output Means 108 Orientation Correcting Means 109 Bending Road Determining Means 110 Temporary Position Calculating Means 1701 Distance Calculating Means 1702 Orientation Calculating Means 1703 Map storage means 1704 Current position estimation means 1705 Output means 1706 Absolute position calculation means 1707 Position detection means 1708 Distance constant correction means 1709 Road length calculation means 1710 Satellite combination determination means 1711 Straightness determination means

Claims (11)

[Claims] 1. A direction sensor for detecting a traveling direction of a vehicle,
A distance sensor for detecting a traveling distance of the vehicle, a current position estimating means for estimating a current position of the vehicle with respect to a reference position by using a traveling direction from the direction sensor and a traveling distance from the distance sensor, and road data is stored. Map storage means,
A map range selecting means for estimating an error amount of the road data stored in the map storage means and determining a range (hereinafter, referred to as an evaluation range) used for a map matching calculation from a portion having a small error amount, and the current position. A map matching calculation unit that performs a map matching calculation using the current position calculated by the estimating unit and the evaluation range selected by the map range selecting unit and corrects the current position on the map road, and the map matching calculating unit. A vehicle position detecting device having an output means for outputting a corrected current position of the vehicle. 2. An azimuth correcting means is provided, and the azimuth correcting means obtains the amount of change in the azimuth of the vehicle and the azimuth of the road data before and after the turn, which is matched by the map matching calculating means, and the difference is a predetermined value. The vehicle position detecting device according to claim 1, wherein the traveling azimuth is corrected when it is small. 3. A bend road determination means is provided, and the bend road determination means determines whether or not the vehicle is traveling on a bend road. When it is determined that the vehicle is traveling on a bend road, the allowable error amount in map matching calculation is increased. The vehicle position detecting device according to claim 1 or 2. 4. The curved road determination means is provided when the number of roads in the road data within a predetermined range including the current position is less than or equal to a predetermined value and the movement locus of the vehicle calculated by the current position estimation means is curved. 4. The vehicle position detecting device according to claim 3, wherein a permissible error amount at the time of map matching calculation is increased on the assumption that the vehicle is currently traveling on a curved road. 5. The curved road determination means is such that the number of intersections in road data within a predetermined range including the current position is less than or equal to a predetermined value,
4. The vehicle position detecting device according to claim 3, wherein the vehicle is currently traveling on a curved road when the movement locus of the vehicle calculated by the current position estimating means is curved. 6. A direction sensor for detecting a traveling direction of a vehicle,
A distance sensor that detects the traveling distance of the vehicle, an output of the azimuth sensor, a current position estimation unit that estimates the current position from the output of the distance sensor, a map storage unit that stores road network data, and the current position estimation A map matching calculation means for matching the current position estimated by the means with a point on the road network stored in the map storage means; and a temporary position used when the map matching calculation means matches the current position, It comprises a temporary position calculation means for calculating using the road network data and an output means for outputting the current position matched by the map matching calculation means, and the map matching calculation means, when approaching a curve, the temporary position. By calculating the traveling azimuth between the temporary position calculated by the calculation means and the point recognized as the current position, and comparing it with the output of the azimuth sensor, the temporary position and the current position are calculated. Vehicle position detecting apparatus characterized by matching the current position to select one to meet the most current position from the point which is recognized as a location. 7. An azimuth calculating means for calculating a traveling azimuth of the vehicle, a distance calculating means for calculating a moving distance of the vehicle, a map storing means for storing map data, the azimuth calculating means, the distance calculating means and the Current position estimation means for estimating the position of the vehicle using the output of the map storage means, output means for outputting the vehicle position obtained by the current position estimation means, and absolute position calculation for calculating the vehicle position in latitude and longitude Means and position verification means for determining an existence possible range of the vehicle based on the absolute position calculated by the absolute position calculation means, and for verifying an inclusive relationship between the estimated position and the existence possible range obtained by the current position estimation means, The position verification means verifies that the estimated position is included in the possible range at least at the start point of the distance calibration section and not at the end point, and the distance calculation error is made from the estimated position at the end point and the possible range. Calculated, the distance the vehicle position detecting apparatus according to claim from calibration section length and the distance calculation error, further comprising a distance constant correcting means for correcting the distance conversion constant of the distance computing means output. 8. A road length calculating means for calculating a moving distance of a vehicle as an integrated value of road lengths using the map data of the map storing means, wherein the distance constant correcting means is calculated by the road length calculating means. 8. The vehicle position detecting device according to claim 7, wherein the distance conversion constant output from the distance calculating means is corrected using the road length. 9. The absolute position calculating means uses a global positioning system, and satellite combination judging means for judging the combination of satellites used for calculating the absolute position by the absolute position calculating means instead of the position checking means. And straightness determining means for determining the straight traveling state of the vehicle, wherein the distance constant correcting means is
The distance between the absolute positions of the vehicle determined by the absolute position calculating means when the satellite combination determining means determines that the combination of satellites is constant and the straightness determining means determines that the vehicle is in the straight traveling state, and the distance calculating means. 8. The vehicle position detecting device according to claim 7, wherein the distance conversion constant is corrected from the output integrated value of. 10. A vehicle speed calculation means for calculating a moving speed of a vehicle, wherein the distance constant correction means converts the distance output of the distance calculation means in a distance calibration section in which the vehicle speed calculated by the vehicle speed calculation means is a certain value or more. 9. The vehicle position detecting device according to claim 7, wherein the constant is modified. 11. A road type discriminating means for identifying a road type of a traveling road using a map storing means, and the distance constant correcting means is a distance calibration discriminated as a highway or a toll road by the road type discriminating means. 9. The vehicle position detecting device according to claim 7, wherein the distance conversion constant of the output of the distance calculating means is corrected in a section.