JP2001296138A - Radius of curvature calculator for running route - Google Patents

Abstract

PROBLEM TO BE SOLVED: To calculate the radius of curvature calculating based on the values obtained by a plurality of methods and also considering the reliability of the calculated values as a whole integrating the values obtained above. SOLUTION: This apparatus is provided with an image pickup means A for recognizing section lines of the running course, a road data means C, a detection means D for detecting the turning state of own vehicle, a receiving means E of a GPS satellite, a first radius of curvature calculating means F for determining the radius of curvature of the running course based on the results of image pickup by the image pickup means A, a section line recognition state judging means G to judge the recognition state of the running course section lines by the image pickup means A, a second radius of curvature calculating means H to calculate the radius of curvature of the running course according to road data, a reception state judging means I to judge the propriety of the reception state of the receiving means E, a third radius of curvature calculating means J for estimating the radius of curvature of the running course, based on the results of detection by the detection means D and a fourth radius of curvature calculating means K for calculating the radius of curvature integrated data P of the running route by combining the first, second and third radius of curvature calculating means F, H and J with the results of the outputs of the section line recognition state judging means G and the reception state judging means I.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for calculating a curvature of a traveling road, and more particularly, to an apparatus for calculating a curvature of a traveling path which can perform the calculation with high accuracy.

[0002]

2. Description of the Related Art Conventionally, a traveling path used for various controls by receiving radio waves from GPS satellites, processing map information obtained from a navigation device, calculating the curvature of the traveling path, and using the curvature data. Is known. In this type of device, it is important how accurately the curvature of the road can be calculated. For example, JP-A-11-
Japanese Patent No. 2528 discloses a technique in which point data from a navigation device is used as it is without complementation, thereby simplifying the calculation and increasing the processing speed.

[0003]

However, in the above prior art, the curvature radius of the road is calculated based on the digital map information, and the curvature is corrected based on the information of the white line obtained by the camera. The calculation result greatly depends on the point data from the navigation device. For example, when a large error occurs in positioning by a GPS satellite, there is a problem that the reliability of the curvature calculation data cannot be ensured. Therefore, the present invention
The curvature of the traveling path, which can calculate the curvature of the traveling path by a plurality of methods and calculate the curvature in consideration of the reliability of each calculated value and the reliability of the calculated value as a whole obtained from each calculated value. A calculation device is provided.

[0004]

In order to solve the above-mentioned problems, the invention described in claim 1 is based on an image pickup means A (for example, a white line 2 in the embodiment) for recognizing a road lane marking (for example, a white line 2 in the embodiment). An image device 1) according to the embodiment, a road data storage unit C that stores road map data B in advance,
Detecting means D (for example, YAW sensor 11, distance sensor 1 in the embodiment) for detecting a value indicating the turning state of the vehicle.
2) and receiving means E for receiving signals from positioning satellites
(For example, the GPS receiver 10 in the embodiment) and the curvature of the traveling road (for example, the right white line curvature data X,
First curvature calculating means F for obtaining left white line curvature data Y)
A lane marking recognition state determining means G for judging a recognition state of a lane marking by the image capturing means; and a curvature of the lane based on the road data stored in the road data storage means (for example, map curvature data in the embodiment). Z), a second curvature calculating means H, and a receiving state determining means I for determining whether the receiving state is good or bad based on the result of reception by the receiving means.
A third curvature calculating means J for estimating a curvature of a traveling road (for example, YAW curvature data Q in the embodiment) based on a detection result of the detecting means, and the first, second and third curvature calculating means. And a fourth curvature calculating means K for calculating the curvature of the traveling road (for example, the curvature fusion data P in the embodiment) by integrating the output results of the lane marking recognition state determining means and the receiving state determining means. And With such a configuration, the curvature data obtained by the first, second, and third curvature calculation means is converted into a state in which reliability is added based on the output results of the lane marking recognition state determination means and the reception state determination means. It can be used with.

According to a second aspect of the present invention, there is provided a curvature changing speed calculating means L for obtaining a changing speed of the curvature based on a past calculation result of the fourth curvature calculating means, and the fourth changing means calculates the curvature changing speed based on the changing speed of the curvature. A fourth aspect of the invention is characterized in that the curvature calculating means calculates the curvature of the traveling path. With this configuration,
In addition to the output results of the lane marking recognition state determination unit and the reception state determination unit, it is possible to use the curvature data in a state in which reliability is added by the curvature change speed calculated by the curvature change speed calculation unit. Become.

According to a third aspect of the present invention, the lane marking recognition state determining means determines the number of detected lane markings in an image area (for example, image area 3 in the embodiment) obtained by the imaging means. The state of the lane marking is determined based on reliability data (for example, reliability data a and b in the embodiment) obtained by multiplying a value obtained by dividing all candidate points by a coefficient. With this configuration, it is possible to represent the appropriateness of the image capturing by the image capturing unit by a numerical value.

According to a fourth aspect of the present invention, the receiving device receives a radio wave from a GPS satellite, and the receiving state determining means determines the GPS reliability data obtained from the number of the GPS receiving satellites and the DOP value of the GPS. It is characterized in that the receiving state of radio waves from GPS satellites is determined based on (for example, the GPS reliability data c in the embodiment). With this configuration, it is possible to represent the appropriateness of the reception state of the radio wave from the GPS satellite by a numerical value.

According to a fifth aspect of the present invention, the curvature change speed calculating means obtains the curvature stability data (for example, the curvature stability data d in the embodiment), and the third curvature is calculated based on the curvature stability data. It is characterized in that the calculating means estimates the curvature of the traveling path. With this configuration,
The curvature stability data makes it possible to increase the calculation accuracy of the third curvature calculation means.

The invention described in claim 6 is the invention according to claim 3.
The reliability data described in the above, and the GP described in the above claim 4.
The overall reliability is determined based on the S reliability data and the curvature stability data according to the fifth aspect. With this configuration, it is possible to use the output result of the fourth curvature calculation data in a state in which the reliability data, the GPS reliability data, and the curvature stability data are added.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a functional block diagram corresponding to claims 1 and 2, and the travel road curvature calculating device according to the present invention has the following components.

An image pickup means A for recognizing a lane marking on a road surface.
Road data storage means C for storing road map data B in advance; and detection means D for detecting a value indicating the turning state of the vehicle.
Receiving means E for receiving a signal from a positioning satellite; first curvature calculating means F for calculating the curvature of the running path based on recognition of the lane marking by the imaging means A; A lane marking recognition state determining means G for determining a recognition state of a road lane marking; a second curvature calculating means H for calculating a curvature of a traveling road based on the road data stored in the road data storage means C; E: a reception state determination unit I for determining whether the reception state is good or bad based on a reception result by E; a third curvature calculation unit J for estimating a curvature of a traveling road based on a value representing a turning state of the own vehicle; A fourth curvature calculating means K which integrates the output results of the second and third curvature calculating means F, H, J, the lane marking recognition state determining means G and the reception state determining means I to calculate the curvature of the traveling road. And A curvature changing speed calculating unit for calculating a curvature changing speed based on a past calculation result of the fourth curvature calculating unit; and a fourth curvature calculating unit configured to calculate a curvature of the traveling road based on the curvature changing speed. Is calculated.

More specifically, in FIG. 2, while the vehicle is running, an image device (imaging means) 1 such as a CCD camera photographs left and right white lines (traveling line dividing lines) 2 shown in FIG. . Through this image processing, the right white line curvature data (curvature) X and the left white line curvature data Y are obtained. As a method of calculating the curvature of the white line by the image apparatus 1, road curvature radius data is calculated by the least square method of a polynomial (quadratic). The digital map data is stored in the device and can be read. The above polynomial may be a cubic equation depending on the road conditions. Here, as shown in FIG. 3, the image processing apparatus 1 performs image processing on the image area 3 using a plurality of columns 4 corresponding to pixels. We take the ratio of candidate columns and use this as the basis for reliability data.

That is, the number of right white line detection points and the number of left white line detection points are obtained from the image device, and the number of right white line detection points is converted into reliability data a expressed by the following equation. a = (number of detected points / number of all candidate points) × coefficient Further, the number of left white line detected points is converted into reliability data b represented by the following equation. b = (number of detected points / total number of candidate points) × coefficient The reliability data a is input to the check unit 6 of the data check unit 5, and the reliability data b is input to the check unit 6 of the data check unit 5. In addition, in FIG.
Are represented by (1) to (50).

The right white line curvature data X is input to the curvature fusion unit 7 as data for curvature fusion, and
The comparison unit 8 of the data check unit 5 compares the value with the previous value P of the curvature fusion data P described later. When the right white line curvature data X has a large difference from the previous value P, the comparison unit 8 sets the reliability data a input to the check unit 6 to “0”, and sets the reliability data a Is input to the curvature fusion unit 7. When there is no large difference compared with the previous value P in the comparing unit 8 (when the difference between them is smaller than the threshold value), the reliability data a is input to the curvature fusion unit 7 as it is.

Similarly, the left white line curvature data (curvature) Y is input to the curvature fusion unit 7 as data for curvature fusion, and the comparison unit 8 of the data check unit 5.
Is compared with the previous value P, and if the left white line curvature data Y is too different from the previous value P in the comparing unit 8 (if the difference between the two is greater than or equal to the threshold value), the checking unit 6
Is set to "0" in the reliability data b input to the curvature fusion unit 7, and the reliability data b is input to the curvature fusion unit 7. Comparison section 8
At the left white line curvature data Y is compared with the previous value P,
When there is no large difference (when the difference is smaller than the threshold)
, The reliability data b is input to the curvature fusion unit 7 as it is. As described above, when the difference between the right white line curvature data X and the left white line curvature data Y is large compared to the previous value P (when the difference between them is equal to or larger than the threshold value), the change is too large to be reliable. , So that it is not taken into account during fusion.

Therefore, when the vehicle pitches greatly due to the unevenness of the road on which the vehicle is traveling and the white line 2 cannot be detected, "0" is set in the reliability data a and b. It is not adopted as data for calculating the curvature, and the accuracy of the curvature calculation is not degraded.

The map curvature calculation device 9 obtains the digital map road coordinate data ahead based on the position of the own vehicle calculated from the GPS data, and calculates the map curvature data (curvature) Z by the least square method of a polynomial (quadratic). Is calculated. Note that the above polynomial may be a cubic equation depending on the road condition in some cases. The calculated map curvature data Z is input to the curvature fusion unit 7 as data for curvature fusion, and is compared with the previous value P by the comparison unit 8.

On the other hand, from the GPS receiver (receiving means E) 10, the number of GPS receiving satellites and the GPS DOP value (H-DOP
Value and P-DOP value), the number of GPS receiving satellites and GPS
The GPS reliability data c is created from the DOP value data.
Here, the DOP value indicates an accuracy index of the GPS position based on the relative position of the GPS satellite, and P-DOP (P
position Deduction of Preci
area) is a value related to the measurement of level and altitude, and H
-DOP (Horizontal Deletion)
of Precision) is a value related to horizontal measurement. It is said that the smaller the value, the better the accuracy.

If the comparison unit 8 compares the map curvature data Z with the previous value P too much (when the difference between them is equal to or greater than a threshold value), the data is input to the check unit 6 of the data check unit 5. "0" in the GPS reliability data c
Is set, and the GPS reliability data c is
Is input to When the map curvature data Z is compared with the previous value P by the comparison unit 8 of the data check unit 5 and there is no large difference (when the difference between the two is smaller than the threshold value), G
The PS reliability data c is input to the curvature fusion unit 7 as it is. Here, the GPS reliability data c is given by the following equation. c = ((1 / H-DOP) × coefficient + (1 / P-DOP)
X coefficient + (1 / received satellites) x coefficient) x coefficient

As described above, the right white line curvature data X
Similarly to the curvature data Y on the left and the left, if the difference between the map curvature data Z and the previous value P is large (if the difference between them is equal to or greater than a threshold value), the change is too large and the data is unreliable. , So that it is not taken into account during fusion. Accordingly, when the GPS satellites are arranged close to each other and the P-DOP value is large and accurate positioning cannot be performed, "0" is set in the GPS reliability data c, and this data is used as the curvature. It is not adopted as data for calculation, and the accuracy of curvature calculation is not deteriorated.

The curvature is converted based on the direction of the vehicle obtained by the YAW sensor (detection means D) 11 and the distance obtained by the distance sensor (V and time Δt by the vehicle speed sensor) 12 as the detection means D. The obtained YAW curvature data Q is input to the curvature fusion unit 7 and is compared with the previous value P in the comparison unit 8 of the data check unit 5. On the other hand, as described later, the curvature fusion data (curvature) P obtained by the curvature fusion unit 7 is stored in the curvature storage device 13 for each process. Based on this, the curvature speed change is calculated and further converted into curvature stability data d. The curvature stability data d is given by the following equation. d = (1 / curvature change speed) × coefficient Here, the curvature change speed is given by (PP) / (P calculation time−P calculation time).

If the YAW curvature data (curvature) Q is too different from the previous value P in the comparison unit 8 of the data check unit 5 (if the difference between the two is greater than a threshold value), the data check unit 5 is set to “0” in the curvature stability data d input to the check unit 6, and the curvature stability data d is input to the curvature fusion unit 7. When the curvature Q is not large in comparison with the previous value P in the comparison unit 8 (when the difference between the two is smaller than the threshold value), the curvature stability data d is input to the curvature fusion unit 7 as it is. You.

The reliability data a, the reliability data b, the GPS reliability data c, and the curvature stability data d
Is input to the curvature fusion unit 7 and the reliability judgment unit 14
Is input to In the curvature fusion section 7, curvature fusion is performed, and curvature fusion data P is obtained. Here, the curvature fusion data P is given by the following equation. P = (aX + bY + cZ + dQ) / (a + b + c + d) Thus, the reliability data a, reliability data b, G
The PS reliability data c and the curvature stability data d are weighted by their reliability and are reflected in the curvature fusion data P according to the contribution. Therefore, accurate calculation according to the reliability of each data can be performed. Since the reliability data a, the reliability data b, the GPS reliability data c, and the curvature stability data d are rewritten to “0” as described above when it is determined that there is no reliability, the curvature fusion is performed. The data P does not affect the data P, and the reliability is improved.

The reliability determining section 14 comprehensively determines reliability based on the reliability data a, the reliability data b, the GPS reliability data c, and the curvature stability data d. If a + b + c + d) falls below the threshold, the curvature fusion data P is forcibly set to “0”.
Is set. On the other hand, when (a + b + c + d) is equal to or larger than the threshold, the curvature fusion data P is output.

Therefore, the reliability is determined individually, and the reliability is comprehensively determined by the reliability determining unit 14.
Only when the reliability exceeds the threshold value, the curvature fusion data P is output, so that not only individual data but also reliability as fused data can be improved. Thereby, for example, the reliability data a, reliability data b, G
When “0” is set in most of the PS reliability data c and the curvature stability data d, it is possible to prevent a situation in which the curvature fusion data P is determined depending only on a specific small amount of data. , The reliability of the curvature fusion data P can be improved. Further, at the time of sensor failure or sensor abnormal data input, the data is not reflected in the curvature fusion data P, so that abnormal data can be suppressed. As a result, a high-precision road curvature can be obtained, and can be used for automatic driving, steering angle control, lane departure suppression, and the like.

The curvature fusion data P obtained in this manner is stored in the curvature storage device 13 for each process, for example,
The value is stored together with data indicating how many times the data is, such as the previous value P and the value P before the last time. When the curvature fusion data P is “0”, the previous value P, which is the previous curvature fusion data P, is used.

Next, control of curvature calculation will be described with reference to the flowchart shown in FIG. In step S1, image data is captured and white lines are extracted, and the number of detected white lines is counted (for the left and right white lines).
Then, in the next step S3, it is converted into reliability data by the number of white line detection points. That is, the reliability data is obtained by multiplying the number of detection results / the number of previous searches by a coefficient. Here, the reliability data is the reliability data a of the white line on the right and the reliability data b of the white line on the left. Next, the process proceeds to step S5, in which the extracted data is converted into plane coordinates, and in the next step S7, an approximate curve is calculated by the least square method.

Then, the process proceeds to step S9, where the curvature of an arbitrary position ahead of the approximate curve, that is, the curvature data of the white line is calculated. The white line curvature data at this time is right white line curvature data X and left white line curvature data Y. Steps S1 to S9 are the same as the image data input step S
100.

Next, GPS data is received in step S11. At this time, data of the H-DOP value and the P-DOP value are input, and the number of received satellites is counted. Then, the process proceeds to step S13. In step S13, the GPS data is converted into reliability data. That is, the reliability data is ((1 / H-DOP) ×
Coefficient + (1 / P-DOP) x Coefficient + (1 / Number of received satellites)
× coefficient) × coefficient. This reliability data is GPS
This is indicated by reliability data c.

Then, in step S15, the GPS
The current position is calculated from the data, and the next step S1
In step 7, coordinates in the vehicle traveling direction are fetched from the current position of the electronic map, and the flow advances to step S19. In step S19, an approximate curve is calculated from the coordinates of the electronic map by the least squares method, and map curvature data Z is obtained. Steps S11 to S19 constitute a GPS data input step S200.

Next, the process proceeds to step S21 in FIG. 5, where the YAW sensor data and the distance sensor data are captured. Then, the process proceeds to step S23. In step S23, the YAW data and the distance data are converted into YAW curvature data Q. Then, the process proceeds to the next step S25. In step S25, the previous value of the curvature fusion data P is fetched and set as P, and the value two times before is taken as P. Then, step S27
Proceed to.

In step S27, a curvature changing speed is calculated. That is, (P−P) / (P
(Calculation time−P calculation time). Then, the process proceeds to step S29. In step S29, the curvature changing speed is converted into curvature stability data d. That is, the curvature stability data d is represented by (1 / curvature change rate) × coefficient. Steps S21 to S29 constitute a vehicle data input step S300.

Next, in step S31 of FIG.
It is determined whether (right white line curvature data X-previous value P) is within a threshold value. If the result of the determination in step S31 is "N
If "O", that is, it is determined that the right white line curvature data-previous value P is not within the threshold value, the process proceeds to step S33. In step S33, the reliability data a of the right white line is rewritten to "0" and the process proceeds to step S35. When the result of the determination in step S31 is “YES”, that is, when it is determined that (right white line curvature data X−previous value P) is within the threshold value,
Proceed to step S35.

In step S35, it is determined whether (left white line curvature data Y-previous value P) is within a threshold value. When the result of the determination in step S35 is "NO", that is, when it is determined that (left white line curvature data Y-previous value P) is not within the threshold value, the process proceeds to step S37, and step S37 is performed.
, The reliability data b of the left white line is rewritten to “0”, and the process proceeds to step S39. If the result of the determination in step S35 is "YES", that is, if it is determined that (left white line curvature data Y-previous value P) is within the threshold, step S35
Proceed to 39.

In step S39, it is determined whether (map curvature data Z-previous value P) is within a threshold value. If the result of the determination in step S39 is “NO”, that is, if it is determined that (map curvature data Z−previous value P) is not within the threshold value, the process proceeds to step S41, and the GPS reliability data c is set to “0” in step S41. Then, the process proceeds to step S43. The result of the determination in step S39 is "YES", that is, (map curvature data Z-previous value P).
Is determined to be within the threshold, the process proceeds to step S43.

Next, in step S43, (YAW
It is determined whether or not the curvature data Q-previous value P) is within the threshold value. If the result of the determination in step S43 is “NO”, that is, if it is determined that (YAW curvature data Q−previous value P) is not within the threshold, the process proceeds to step S45, and the curvature stability data d is set to “0” in step S45. The process proceeds to step S47. If the result of the determination in step S43 is "YES", that is, if it is determined that (YAW curvature data Q-previous value P) is within the threshold value, the flow proceeds to step S47.

In step S47, curvature fusion data P is obtained. This curvature fusion data P is, as described above, P = (aX + bY + cZ + dQ) / (a + b + c)
+ D). Then, the curvature fusion data P is stored in the curvature storage device 13 in step S49. Then, the process proceeds to step S51. In step S51, reliability is determined. This reliability judgment is (a + b + c + d)
Is determined by whether or not exceeds a threshold. This step S
51 is “NO”, that is, (a + b +
If it is determined that (c + d) does not exceed the threshold, the process proceeds to step S53, where the curvature fusion data P is rewritten to “0” in step S53, and the process proceeds to step S55.

If the result of the determination in step S51 is "Y
ES ", that is, when it is determined that (a + b + c + d) exceeds the threshold value, the process is terminated, and the curvature fusion data P is output for use in various controls. For example, when the vehicle is traveling, a warning is issued when the vehicle deviates from the travel path determined by the curvature fusion data P, or the driver is alerted to be able to travel along the travel path. be able to. Step S31 to step S5
The steps up to 3 constitute the data check step S400. Therefore, the flow chart is roughly divided into an image data input step S100, a GPS data input step S200, and a vehicle data input step S30.
0, a data check step S400, and the data reliability is ensured by the data check step S400.

According to the above embodiment, the right white line curvature data X and the left white line curvature data Y are used to calculate the curvature fusion data P in consideration of the reliability data a and the reliability data b, respectively. Further, the map curvature data Z is used in consideration of the GPS reliability data c, and the YAW curvature data Q is used.
Can be used in a state in which the curvature stability data d is added, the reliability of each curvature data can be improved, and the curvature fusion data P is obtained by weighting each data. The curvature fusion data P is calculated in accordance with the degree of, so that accurate and reliable curvature fusion data P can be obtained. So, for example,
If the data obtained from the image device 1 due to the pitching is not reliable (a and b are set to “0”), the data from the image device 1 does not participate in the calculation of the curvature fusion data P. Therefore, the reliability of the curvature fusion data P is not reduced by this data. Also, when the GPS satellites are arranged close to each other and the positioning accuracy cannot be ensured (when the DOP value is large and c is set to “0”), the data from the GPS satellites still has a curvature. Since it is not involved in the calculation of the fusion data P, the reliability of the curvature fusion data P is not reduced by this data.

Since the YAW curvature data Q obtained by the YAW sensor 11 and the distance sensor 12 is used in consideration of the curvature stability data d using the past curvature fusion data P, the past curvature fusion data Q is used. It is possible to further increase the reliability by taking into account the calculation results of the data P (previous value P, last-before-time value P, etc.), and accurately calculate the curvature fusion data P. Therefore, when the data from the YAW sensor 11 and the data from the distance sensor 12 are not stable, this output is not reflected on the curvature fusion data P.
Only when the YAW sensor 11 and the distance sensor 12 are stable, the data becomes valid and the accuracy of the data can be improved.

The reliability judgment unit 14 compares the sum of the reliability data a, the reliability data b, the GPS reliability data c, and the curvature stability data d with a threshold value and uses the sum to determine the overall reliability. As a result, we make comprehensive judgments not only on the reliability of each data but also on the overall reliability,
Since this is used when outputting the curvature fusion data P,
More accurate curvature calculation can be performed. As a result, accurate and accurate curvature fusion data P can be obtained, which can be effectively used for automatic driving, steering angle control, lane departure warning, and lane departure suppression.

[0042]

As described above, according to the first aspect of the present invention, the curvature data obtained by the first, second, and third curvature calculating means is used for dividing the line recognition state determining means. The reliability of the calculation data can be increased, and the curvature can be calculated accurately, because the reliability of the calculation data can be increased.

According to the second aspect of the present invention, in addition to the output results of the lane marking recognition state determining means and the receiving state determining means, the reliability is determined by the curvature changing speed calculated by the curvature changing speed calculating means. Can be used in a state in which is taken into account, so that there is an effect that the reliability can be further improved by the amount in consideration of the past calculation result, and the curvature can be calculated accurately.

According to the third aspect of the present invention, it is possible to represent the appropriateness of photographing by the image pickup means by a numerical value, so that information relating to reliability is included in the first curvature calculation means to further improve reliability. There is an effect that the curvature can be improved and the curvature can be accurately calculated.

According to the fourth aspect of the present invention, the GPS
Since it is possible to represent the appropriateness of the reception state of the radio wave from the satellite by a numerical value, it is necessary to include reliability-related information in the second curvature calculation means to further enhance the reliability and accurately calculate the curvature. There is an effect that can be.

According to the fifth aspect of the present invention, it is possible to increase the calculation accuracy of the third curvature calculating means based on the curvature stability data, so that the reliability is further improved and the curvature is accurately calculated. There is an effect that can be.

According to the invention described in claim 6, it is possible to use the output result of the fourth curvature calculation data in a state in which the reliability data, the GPS reliability data, and the curvature stability data are added. Therefore, there is an effect that the reliability of the fourth curvature calculation data is improved and the curvature can be calculated accurately.

[Brief description of the drawings]

FIG. 1 is a block diagram corresponding to claims 1 and 2 of the present invention.

FIG. 2 is a block diagram of an embodiment of the present invention.

FIG. 3 is a display diagram of the image device according to the embodiment of the present invention.

FIG. 4 is a flowchart of the embodiment of the present invention.

FIG. 5 is a flowchart of the embodiment of the present invention.

FIG. 6 is a flowchart of the embodiment of the present invention.

FIG. 7 is a flowchart of the embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image device (imaging means A) 2 White line (traveling line division line) 3 Image area 10 GPS receiver (receiving means E) 11 YAW sensor (detecting means D) 12 Distance sensor (detecting means D) a, b Reliability data c GPS reliability data d Curvature stability data P Curvature fusion data (curvature) B Road map data C Road data storage means F First curvature calculation means G Separation line recognition state determination means H Second curvature calculation means I Receiving state Judgment means J Third curvature calculation means K Fourth curvature calculation means L Curvature change rate calculation means Q YAW curvature data (curvature) X Right white line curvature data (curvature) Y Left white line curvature data (curvature) Z Map curvature data ( curvature)

Claims (6)

[Claims] 1. An image pickup means for recognizing a lane marking on a road surface, a road data storage means for storing road map data in advance, a detection means for detecting a value indicating a turning state of a vehicle, and a positioning satellite. Receiving means for receiving the signal of the following, first curvature calculating means for obtaining a curvature of the traveling road based on recognition of the lane marking by the imaging means, and classification for judging a recognition state of the lane marking by the imaging means. Line recognition state determination means, second curvature calculation means for calculating the curvature of the traveling road based on the road data stored in the road data storage means, and reception determining whether the reception state is good or bad based on the reception result by the reception means State determination means, third curvature calculation means for estimating the curvature of the traveling path based on the detection result of the detection means,
And a fourth curvature calculating means for calculating the curvature of the traveling road by integrating the output results of the first, second, and third curvature calculating means, the lane marking recognition state determining means, and the receiving state determining means. Characteristic device for calculating the curvature of a traveling path. 2. A method according to claim 1, further comprising a curvature changing speed calculating unit for obtaining a changing speed of the curvature based on a past calculation result of the fourth curvature calculating unit. The curvature calculation device for a traveling path according to claim 1, wherein the curvature of the traveling road is calculated. 3. The lane marking recognition state determining means based on reliability data obtained by multiplying a value obtained by dividing the number of detected lane markings in the image area obtained by the imaging means by all candidate points by a coefficient. The travel road curvature calculating device according to claim 1, wherein the state of the travel road division line is determined. 4. The receiving device receives a radio wave from a GPS satellite, and the receiving state determining means receives the radio wave from the GPS satellite based on the GPS reliability data obtained from the number of the GPS receiving satellites and the GPS DOP value. The travel path curvature calculation device according to claim 1, wherein the radio wave reception state is determined. 5. The method according to claim 2, wherein said curvature change speed calculating means obtains curvature stability data, and based on said curvature stability data, third curvature calculation means estimates the curvature of the traveling road. An apparatus for calculating a curvature of a traveling path according to claim 1. 6. The reliability data according to claim 3, wherein
An apparatus for calculating a curvature of a traveling road, comprising: determining a total reliability based on the GPS reliability data described in claim 4 and the curvature stability data described in claim 5.