JP2016035423A - Position detection apparatus and position detection method - Google Patents

Abstract

An object of the present invention is to reduce the influence of a measurement error of a relative position of a landmark while a moving body travels on a curved road on accuracy of a calculated position of the moving body.
In a case where a mobile object travels on a curved road and there is a landmark on the outside of the curved road, the position detection device 40 includes at least a land on the outside of the curved road. Select a combination of marks. The position detection device 40 calculates the position of the mobile body 1 on the map based on the relative position of the selected landmark and the position on the map.
[Selection] Figure 2

Description

  The present invention relates to a position detection device and a position detection method.

  As a technique for calculating the current position of the moving body, a position detection device that calculates the position of the moving body on the map from the relative position of the landmark to the moving body and the position of the landmark on the map is known. For example, the position detection device described in Patent Literature 1 images a landmark installed along a moving path of a moving body with an imaging unit. When at least three or more landmarks are recognized, the position detection device reads the position data of the three landmarks stored in advance in the storage unit. The position detection device calculates a relative angle θ between the landmarks as viewed from the moving body based on the image, and calculates the current position of the moving body based on the position data and the angle data.

JP 2001-142532 A

The relative position of the landmark with respect to the moving body changes as the moving body moves. While the moving body is moving linearly, there is no difference in the amount of change in relative position between different landmarks. However, the amount of change in the relative position of the landmark varies depending on the relative position of the landmark while the direction of the traveling direction of the moving body changes as the moving body travels on the curved road. The influence of the measurement error of the relative position of the landmark on the accuracy of the calculated position of the moving body becomes larger as the change amount of the relative position of the landmark accompanying the movement of the moving body is smaller.
The subject of this invention is reducing the influence which the measurement error of the relative position of the landmark during a mobile body drive | works on a curved road has on the precision of the calculation position of a mobile body.

  A position detection device according to one embodiment of the present invention detects a relative position of a landmark with respect to a moving object. The position detection device selects a combination of landmarks including at least a landmark outside the curved road when the moving body travels on the curved road and there is a landmark outside the curved road. . The position detection device calculates the position of the moving body on the map based on the relative position of the selected combination of landmarks and the position on the map.

  ADVANTAGE OF THE INVENTION According to this invention, the influence which the measurement error of the relative position of the landmark while a moving body drive | works a curve road has on the precision of the calculation position of a moving body can be reduced.

It is a figure which shows the structural example of a vehicle provided with the position detection apparatus which concerns on embodiment of this invention. It is a figure which shows the function structure of the position detection apparatus which concerns on 1st Embodiment of this invention. It is explanatory drawing of the measurement of the position of the scanning point by a position measurement part. It is explanatory drawing of an example of the calculation method of the provisional absolute position and provisional front-back direction of a vehicle. It is explanatory drawing of an example of the landmark extraction process by an extraction part. (A) is explanatory drawing of the positional relationship of the vehicle and landmark in the time t-1 and t, (b) is the variation | change_quantity of the relative position of the landmark with respect to the vehicle which occurred from the time t-1 to the time t. It is explanatory drawing of. (A) And (b) is explanatory drawing of an example of the landmark selected by the landmark selection part. It is explanatory drawing of an example of the calculation process of the absolute position of a vehicle. It is explanatory drawing of an example of operation | movement of the position detection apparatus which concerns on 1st Embodiment of this invention. It is explanatory drawing of an example of the landmark selection process shown in FIG. It is explanatory drawing of the change of the landmark selected by the landmark selection part. It is explanatory drawing of an example of the landmark selection process in the position detection apparatus which concerns on 2nd Embodiment of this invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
(First embodiment)
(Constitution)
Please refer to FIG. The position detection apparatus according to the first embodiment of the present invention is mounted on, for example, the vehicle 1 and is used to detect the current position on the map of the vehicle 1. The current position of the vehicle 1 detected by the vehicle detection device is used, for example, for calculating a route from the current position of the vehicle 1 to a desired destination in a navigation device mounted on the vehicle 1, or of the vehicle 1 Can be used for automatic operation. Further, the current position of the vehicle 1 detected by the vehicle detection device can be used for driving support for controlling the operation force so that the vehicle 1 continuously travels near the center of the travel lane, for example. The application of the position detection device according to the first embodiment of the present invention is not limited to this. The position detection apparatus according to the first embodiment of the present invention can be widely used for detection of the current position of a moving body.

The vehicle 1 includes a wheel 2, a vehicle speed sensor 3, a steering wheel 4, a steering angle sensor 5, a position measurement sensor 6, a yaw rate sensor 7, a controller 10, a map information storage unit 20, and a navigation device 30. Prepare.
The vehicle speed sensor 3 detects the wheel speed of the wheel 2 and calculates the speed of the vehicle 1 based on the wheel speed. The vehicle speed sensor 3 outputs speed information indicating the calculated speed to the controller 10. The vehicle speed sensor 3 may include a pulse generator such as a rotary encoder that measures wheel speed pulses. The steering angle sensor 5 detects the steering angle of the steering wheel 4 and outputs steering angle information indicating the detected steering angle to the controller 10. The steering angle sensor 5 is provided on a steering shaft or the like. The steering angle sensor 5 may detect the turning angle of the steered wheels as steering angle information.

  The position measurement sensor 6 scans the surface of an object existing around the vehicle 1 to measure the distance from the vehicle 1 to the scanning point on the object surface, and acquires the direction to the scanning point. The position measurement sensor 6 may include, for example, a laser range finder or a radar mounted on the vehicle 1. The position measurement sensor 6 outputs information on the measured distance and direction to the scanning point to the controller 10. The yaw rate sensor 7 detects the yaw rate of the vehicle 1. The yaw rate is the rate of change of the rotation angle in the direction in which the vehicle body turns. The yaw rate sensor 7 outputs yaw rate information indicating the detected yaw rate to the controller 10.

  The map information storage unit 20 stores map information. The map information may include information on geographical landmarks that can be used as a reference when calculating the position of the vehicle 1 on the map. In the following description, a geographical landmark that can be used as a reference when calculating the position of the vehicle 1 on the map may be referred to as a “landmark”. The map information may include landmark identification information and position information on the map. The landmark may be, for example, a utility pole, a signal pole, a sign pole, an illumination pole, etc. installed on the side of the road. For example, the landmark may be a three-dimensional object such as a guide board or a building.

The position on the map stored in the map information storage unit 20 is expressed by the absolute position on the absolute coordinate system with the fixed point on the ground as the origin. The absolute coordinate system may be, for example, a world coordinate system or a geodetic coordinate system. In the following description, the position on the map stored in the map information storage unit 20, that is, the absolute position on the absolute coordinate system may be referred to as “absolute position”.
The map information stored in the map information storage unit 20 further includes road information regarding roads. The road information may include, for example, information on the absolute position of a point where the road passes, information on the curvature of each curved road included in the road, and information on the road width of the road.

  The controller 10 is an electronic control unit including a CPU (Central Processing Unit) and CPU peripheral components such as a ROM (Read Only Memory) and a RAM (Random Access Memory). The controller 10 detects the relative position of the landmark based on the information on the distance and direction to the scanning point output from the position measurement sensor 6. The controller 10 reads the absolute position of the landmark from the map information storage unit 20. The controller 10 calculates the current absolute position of the vehicle 1 based on the measured relative position of the landmark and the read absolute position of the landmark.

  The navigation device 30 displays the traveling direction of the vehicle 1, road information around the vehicle 1, and the like based on the current position of the vehicle 1 and map information calculated by the controller 10. The navigation device 30 calculates a route to the set destination point and displays route guidance to the destination point. The navigation device 30 may use map information stored in the map information storage unit 20. The map information storage unit 20 may be built in the navigation device 30.

  Hereinafter, processing by the controller 10 will be described. Please refer to FIG. The position detection device 40 includes the position measurement sensor 6 and the map information storage unit 20 described above. The position detection device 40 includes a position measurement unit 41, a peripheral object detection unit 42, a movement amount calculation unit 43, a position calculation unit 44, and an extraction unit 45. The position detection device 40 includes a landmark selection unit 46, a first curvature radius estimation unit 47, a threshold specification unit 48, a distance calculation unit 49, a curve determination unit 50, a second curvature radius estimation unit 51, A position detection unit 52 is provided.

  Processing described later by the position measurement unit 41, the peripheral object detection unit 42, the movement amount calculation unit 43, the position calculation unit 44, and the extraction unit 45 is executed by the controller 10. Processing described later by the landmark selection unit 46, the first curvature radius estimation unit 47, the threshold specification unit 48, the distance calculation unit 49, the curve determination unit 50, the second curvature radius estimation unit 51, and the relative position detection unit 52 is performed by a controller. 10 is executed.

  The position measurement unit 41 measures the relative position of the scanning point with respect to the vehicle 1 based on the distance and direction information to the scanning point on the object around the vehicle 1 output from the position measurement sensor 6. With reference to FIG. 3, an example of the measurement of the position of the scanning point by the position measuring unit 41 will be described. In the accompanying drawings, the x-axis direction indicates the lateral direction of the vehicle 1, and the y-axis direction indicates the front-rear direction of the vehicle 1. The lateral direction of the vehicle 1 is a direction orthogonal to the moving direction of the vehicle 1. The front-rear direction of the vehicle 1 is a direction orthogonal to the lateral direction of the vehicle 1. Hereinafter, the lateral direction of the vehicle 1 is simply referred to as “lateral direction”. The front-rear direction of the vehicle 1 is simply referred to as “front-rear direction”.

  The position measurement unit 41 mounted on the vehicle 1 is determined by the distance r1 to the scanning point 61 and the direction θ1 as a relative position of the scanning point 61 on the surface of the object 60 in a relative coordinate system with the position of the vehicle 1 as the origin. The position may be measured. The direction θ1 may be a horizontal angle of the scanning point 61 with respect to the x-axis direction of the vehicle 1, for example. The direction θ1 may be, for example, the azimuth angle of the scanning point 61 with the vehicle 1 as the origin and the x-axis direction as a reference. In the following description, information on the relative position measured by the position measurement unit 41 may be referred to as “measurement position information”. Further, the relative position with respect to the vehicle 1 may be simply referred to as “relative position”. The position measurement unit 41 outputs the measurement position information to the peripheral object detection unit 42.

Please refer to FIG. The peripheral object detection unit 42 detects an object existing around the vehicle 1 based on the measurement position information received from the position measurement unit 41. For example, the peripheral object detection unit 42 performs a segmentation process on the measurement position information. At this time, the peripheral object detection unit 42 connects scanning points having a short separation distance. The peripheral object detection unit 42 divides the scanning points received from the position measurement unit 41 into segments by detecting a set of connected scanning points as a segment. The peripheral object detection unit 42 recognizes the detected segment as an object.
The peripheral object detection unit 42 detects the relative position of the object based on the measurement position information of the scanning points belonging to the segment. For example, the average coordinates of the measurement positions of the scanning points belonging to the segment may be calculated as the relative position of the object. The peripheral object detection unit 42 outputs the relative position of the detected object to the extraction unit 45.

The vehicle speed sensor 3 outputs the speed information of the vehicle 1 to the movement amount calculation unit 43. The yaw rate sensor 7 outputs the yaw rate information of the vehicle 1 to the movement amount calculation unit 43. The movement amount calculation unit 43 calculates the movement amount (dx, dy, dφ) of the vehicle 1 from time t-1 to time t based on the received speed information and yaw rate information. The time t is the latest measurement time of the position measurement sensor 6, and the time t-1 is the measurement time of the position measurement sensor 6 before the time t. For example, the measurement interval of the position measurement sensor 6 may be Tm, and the time t may be a time delayed by the measurement interval Tm from the time t-1.
dx and dy indicate translational movement amounts in the lateral direction and the front-rear direction, respectively. dφ represents the amount of rotation in the direction in which the vehicle body turns. The movement amount calculation unit 43 outputs the calculated movement amounts (dx, dy, dφ) to the position calculation unit 44.

  Based on the absolute position (Xt-1, Yt-1) and the front-rear direction (Φt-1) of the vehicle 1 at the time t-1 that has already been calculated, the position calculation unit 44 temporarily calculates the vehicle 1 at the time t. Absolute position (Xtt, Ytt) and provisional front-rear direction (Φtt) are calculated. When calculation of the absolute position of the vehicle 1 and the direction in the front-rear direction at time t-1 has not yet been completed, the measurement values obtained by other positioning means may be used. Other positioning means may be, for example, GPS (Global Positioning System). In the following description, the front-rear direction of the vehicle 1 may be simply referred to as “the direction of the vehicle 1”.

  Please refer to FIG. For example, the position calculation unit 44 receives the movement amount (dx, dy, dφ) from the movement amount calculation unit 43. The position calculation unit 44 corrects the absolute position (Xt−1, Yt−1) and the direction (Φt−1) at time t−1 with the movement amount (dx, dy, dφ), and the temporary absolute position at time t. (Xtt, Ytt) and a temporary direction (Φtt) are calculated. The position calculation unit 44 outputs the temporary absolute position (Xtt, Ytt) and temporary direction (Φtt) of the vehicle 1 to the extraction unit 45.

  Please refer to FIG. The extraction unit 45 receives the relative positions of objects around the vehicle 1 from the surrounding object detection unit 42. The extraction unit 45 receives the provisional absolute position (Xtt, Ytt) and provisional orientation (Φtt) of the vehicle 1 from the position calculation unit 44. The extraction unit 45 reads the landmark identification information and the absolute position from the map information storage unit 20. Based on the temporary absolute position and temporary orientation received from the position calculation unit 44 and the absolute position read from the map information storage unit 20, the extraction unit 45 selects a land from the objects received from the peripheral object detection unit 42. Extract marks.

  Please refer to FIG. Assume that the objects 62, 63, 64, and 65 are detected as objects around the vehicle 1. Assume that the relative positions of the objects 62, 63, 64 and 65 are pd1 (x1, y1), pd2 (x2, y2), pd3 (x3, y3) and pd4 (x4, y4), respectively. Further, it is assumed that absolute positions Pr1 (Xr1, Yr1) and Pr2 (Xr2, Yr2) of two landmarks are read from the map information storage unit 20.

  The extraction unit 45 assumes that the absolute position and orientation of the vehicle 1 are the provisional absolute position (Xtt, Ytt) and the provisional orientation (φtt), and the relative positions pd1 to pd4 of the objects 62 to 65 to the absolute position. Convert. The absolute coordinates of the objects 62, 63, 64 and 65 are denoted as Pd1 (Xd1, Yd1), Pd2 (Xd2, Yd2), Pd3 (Xd3, Yd3) and Pd4 (Xd4, Yd4), respectively.

  The extraction unit 45 compares the absolute coordinates Pd1 to Pd4 with the absolute positions Pr1 and Pr2 of the landmarks, and extracts objects 63 and 64 that are close to the absolute positions Pr1 and Pr2, respectively, as landmarks. For example, the extracting unit 45 may extract the object as a landmark when the distance between the object and the landmark is within a predetermined distance. The predetermined distance may be 0.5 m, for example. The extraction unit 45 outputs the extracted landmark identification information and its relative position to the landmark selection unit 46.

  The landmark selection unit 46 selects a combination of landmarks used for calculating the position of the vehicle 1 from the landmarks extracted by the extraction unit 45. The landmark selection unit 46 determines whether or not the vehicle 1 is traveling on a curved road. Whether or not the vehicle 1 travels on a curved road can be determined based on the temporary absolute position and orientation of the vehicle 1 and the road information stored in the map information storage unit 20. Further, whether or not the vehicle 1 travels on a curved road can be determined based on the steering angle detected by the steering angle sensor 5.

  The landmark selection unit 46 determines whether or not to select a combination of landmarks including at least a landmark on the outside of the curved road, depending on whether or not the vehicle 1 travels on the curved road. When the vehicle 1 travels on a curved road, the landmark selection unit 46 selects a combination of landmarks including at least a landmark outside the curved road. That is, the landmark selection unit 46 determines whether to select a landmark on the outside of the curved road with a higher priority than a landmark on the inside depending on whether or not the vehicle 1 travels on the curved road. To do. When the vehicle 1 travels on a curved road, the landmark selection unit 46 preferentially selects a landmark on the outside of the curved road over a landmark on the inside.

  In the following description, the meaning of the description “the vehicle 1 travels on a curved road” includes not only the case where the vehicle 1 is traveling on the curved road but also the traveling road ahead of the vehicle 1 is a curved road. Cases are also included. That is, the landmark selection unit 46 determines whether or not to select a combination of landmarks including at least landmarks outside the curved road, depending on whether or not the traveling road ahead of the vehicle 1 is a curved road. to decide. When the traveling road ahead of the vehicle 1 is a curved road, the landmark selection unit 46 selects a combination of landmarks including at least a landmark outside the curved road.

  When the vehicle 1 does not travel on a curved road, for example, the landmark selection unit 46 may select a combination of landmarks including either a landmark outside the curved road or a landmark inside the curved road. For example, when the traveling road ahead of the vehicle 1 is not a curved road, a combination of landmarks including either a landmark outside the curved road or a landmark inside the curved road may be selected.

In the following description, a landmark outside the curved road of the vehicle 1 may be referred to as an “outer landmark”. A landmark inside the curved road of the vehicle 1 may be referred to as an “inner landmark”.
The reason will be described with reference to FIGS. 6A and 6B. In FIG. 6A, reference numeral 66 indicates a curved road on which the vehicle 1 travels. Reference numeral 1a indicates the position of the vehicle 1 at time t-1, and reference numeral 1b indicates the position of the vehicle 1 at time t. Reference numeral 67 indicates an outer landmark. Reference numeral 68 indicates an inner landmark. When the vehicle 1 moves along the curved road while turning from the position 1a to the position 1b, the relative position of the outer landmark 67 and the relative position of the inner landmark 68 change.

  Reference is made to FIG. The positions of the outer landmark 67 and the inner landmark 68 are indicated by relative positions in a relative coordinate system with the position of the vehicle 1 as the origin. Reference numerals 67a and 67b indicate the relative positions of the outer landmarks 67 at time t-1 and time t, respectively. Reference numerals 68a and 68b indicate the relative positions of the inner landmarks 68 at time t-1 and time t, respectively.

  The amount of change L2 between the relative positions 68a and 68b of the inner landmark 68 is smaller than the amount of change L1 between the relative positions 67a and 67b of the outer landmark 67. For this reason, when the inner landmark 68 is used for calculating the position of the vehicle 1, the influence of the measurement error of the relative position of the landmark on the accuracy of the calculated position of the vehicle 1 is larger than when the outer landmark 67 is used. Become. For this reason, the landmark selecting unit 46 preferentially selects a combination of landmarks including the outer landmark 67, thereby affecting the influence of the measurement error of the relative position of the landmark on the accuracy of the calculated position of the vehicle 1. To reduce.

  For example, the landmark selection unit 46 may select a combination of landmarks including two or more outer landmarks as a combination of landmarks including at least the outer landmark. In addition, the landmark selection unit 46 includes, as a combination of landmarks including at least the outer landmark, a landmark including a number of outer landmarks corresponding to the number Co of outer landmarks and a number of inner landmarks corresponding to Co. A combination of these may be selected.

  For example, when the number Co of outer landmarks is less than a predetermined number Ct, the landmark selection unit 46 selects a combination of landmarks including Co outer landmarks and (Ct-Co) inner landmarks. It's okay. For example, when the number of outer landmarks Co is equal to or greater than a predetermined number Ct, the landmark selection unit 46 may select a combination of landmarks including only Ct outer landmarks.

  When the number Co of outer landmarks is equal to or greater than the predetermined number Ct, the landmark selecting unit 46 selects the outer landmark with priority over the inner landmark by selecting the predetermined number Ct of outer landmarks. When the number Co of outer landmarks is less than the predetermined number Ct, the landmark selecting unit 46 selects all the outer landmarks more preferentially than the inner landmark. For example, the predetermined number Ct may be the number of landmarks used to calculate the position of the vehicle 1 with a desired accuracy. For example, the predetermined number Ct may be “2”.

Reference is made to FIG. 7A and FIG. 7B. Reference numeral 69 indicates a curved road on which the vehicle 1 travels. Reference numerals 70, 71 and 72 denote outer landmarks. Reference numerals 73, 74, and 75 denote inner landmarks. Assume that the predetermined number Ct is “2”.
When the predetermined number Ct is “2”, if the number Co of outer landmarks is “2” or more, the landmark selection unit 46 selects a combination of landmarks including only two outer landmarks. For example, when there are three outer landmarks 70 to 72 as shown in FIG. 7A, the landmark selection unit 46 may select the two outer landmarks 70 and 72.

  On the other hand, if the number Co of outer landmarks is “1”, the landmark selecting unit 46 selects a combination of landmarks including one outer landmark and one inner landmark. For example, when there is one outer landmark 71 as shown in FIG. 7A, the landmark selection unit 46 may select a combination of landmarks including the outer landmark 71 and the inner landmark 73. .

When the number Co of outer landmarks is “0”, the landmark selecting unit 46 selects a combination of landmarks including only two inner landmarks. The predetermined number Ct is not limited to “2”. The predetermined number Ct may be an integer greater than or equal to “3”.
The landmark selection unit 46 may calculate the number Co of outer landmarks by detecting the outer landmark from the landmarks received from the extraction unit 45. For example, the landmark selection unit 46 receives the landmark received from the extraction unit 45 based on the landmark identification information received from the extraction unit 45 and the landmark absolute position and road information stored in the map information storage unit 20. May be detected as an outer landmark.

  Please refer to FIG. The landmark selection unit 46 determines whether or not to select a combination of landmarks including at least the outer landmark according to whether or not the curvature radius of the curved road on which the vehicle 1 travels is equal to or less than the first threshold radius Rt1. You can judge. The landmark selection unit 46 may select a combination of landmarks including at least the outer landmark when the curvature radius of the curved road on which the vehicle 1 travels is equal to or less than the first threshold radius Rt1. The landmark selection unit 46 may select a combination of landmarks including either the outer landmark or the inner landmark when the curvature radius of the curved road on which the vehicle 1 travels exceeds the first threshold radius Rt1. By selecting a combination of landmarks including any one of the outer landmark and the inner landmark, the vehicle 1 can be used when the radius of curvature is relatively large and the difference in the amount of change in the relative position between the outer landmark and the inner landmark is small. The number of landmark candidates that can be used for position calculation can be increased.

The first curvature radius estimation unit 47 estimates the curvature radius of the curved road on which the vehicle 1 travels. The curvature radius estimated by the first curvature radius estimation unit 47 is expressed as “first curvature radius Re1”. For example, the first curvature radius estimation unit 47 may estimate the first curvature radius Re1 of the curve road ahead of the vehicle 1.
For example, the first curvature radius estimation unit 47 may estimate the first curvature radius Re1 based on the temporary absolute position and orientation of the vehicle 1 and the road information stored in the map information storage unit 20. For example, the first curvature radius estimation unit 47 calculates the first curvature radius Re1 of the curve road ahead of the vehicle 1 based on the route information calculated by the navigation device 30 and the road information stored in the map information storage unit 20. May be estimated. The first curvature radius estimation unit 47 outputs the estimated first curvature radius Re1 to the landmark selection unit 46.

  The threshold designation unit 48 designates the first threshold radius Rt1. The first threshold radius Rt1 may be a fixed value stored in advance, or may be dynamically calculated according to at least one of the road width of the curved road and the speed of the vehicle 1. For example, the first threshold radius Rt1 is an allowable value when the first curvature radius Re1 is equal to or smaller than the first threshold radius Rt1 and the difference in the change amount of the relative position between the outer landmark and the inner landmark that occurs during the measurement interval Tm. You may set so that it may become above. In the following description, the difference in the change amount of the relative position between the outer landmark and the inner landmark that occurs during the measurement interval Tm may be simply referred to as “change amount difference”.

For example, if the speed of the vehicle 1 is V, the road width of the curved road is W, and the allowable value is ΔL, the first threshold radius Rt1 may be specified so as to satisfy the following expression (1).

When the above equation (1) is modified, the following equation (2) is obtained.

For example, if the speed V is 30 km / h, the road width W is 3 m, the measurement interval Tm is 0.1 second, and the allowable value ΔL is 5 cm, the first threshold radius Rt1 is 100 m.
As apparent from the above equation (2), when the road width W increases, the first threshold radius Rt1 also increases. This is because the change amount difference increases as the road width W increases, and the change amount difference becomes greater than or equal to the allowable value on a gentler curve road. As the speed V increases, the first threshold radius Rt1 also increases. This is because the change amount difference increases as the speed V increases, and thus the change amount difference becomes greater than or equal to the allowable value on a gentler curve road.

  For this reason, for example, the threshold value specifying unit 48 may read the road width of the curved road from the map information storage unit 20 and dynamically calculate the first threshold radius Rt1 according to the above equation (2). That is, the threshold value specifying unit 48 may dynamically calculate the first threshold radius Rt1 so that the first threshold radius Rt1 increases as the road width increases. For example, the threshold value specifying unit 48 may read the speed information output by the vehicle speed sensor 3 and dynamically calculate the first threshold radius Rt1 according to the above equation (2). That is, the first threshold radius Rt1 may be dynamically calculated so that the first threshold radius Rt1 increases as the speed of the vehicle 1 increases. The threshold designation unit 48 may dynamically calculate the first threshold radius Rt1 according to the above equation (2) according to both the read road width and the speed of the vehicle 1.

By calculating the first threshold radius Rt1 so as to increase in accordance with the increase in the road width, it is possible to avoid using only the inner landmarks with the change amount difference exceeding the allowable value due to the increase in the road width. . Further, even if the change amount difference is less than the allowable value due to the reduction of the road width, it is possible to avoid limiting landmark candidates that can be used for calculating the position of the vehicle 1.
By calculating the first threshold radius Rt1 so as to increase in accordance with the increase in the speed of the vehicle 1, it is possible to avoid using only the inner landmark while the change amount difference is not less than the allowable value due to the increase in speed. it can. Moreover, even if the change amount difference is less than the allowable value due to the decrease in speed, it is possible to avoid limiting landmark candidates that can be used for calculating the position of the vehicle 1.

The landmark selection unit 46 determines whether or not the radius of curvature of the travel path where the vehicle 1 is currently traveling exceeds the first threshold radius Rt1. When the radius of curvature of the travel path at the point where the vehicle 1 is currently traveling exceeds the first threshold radius Rt1, the landmark selection unit 46 is either the outer landmark or the inner landmark. Select a combination of landmarks including.
The second curvature radius estimator 51 estimates the curvature radius of the travel path at the point where the vehicle 1 is currently traveling. The curvature radius estimated by the second curvature radius estimation unit 51 is expressed as “second curvature radius Re2”. For example, the second curvature radius estimation unit 51 may estimate the second curvature radius Re <b> 2 based on the temporary absolute position and orientation of the vehicle 1 and the road information stored in the map information storage unit 20. The second curvature radius estimation unit 51 outputs the second curvature radius Re2 to the landmark selection unit 46.

The landmark selection unit 46 includes at least a landmark on the outside of the curved road depending on whether or not the vehicle 1 has reached a position separated from the entrance of the curved road ahead of the vehicle 1 by the threshold distance Dt. It may be determined whether or not to select a combination of marks. That is, the landmark selection unit 46 may select a combination of landmarks including at least a landmark outside the curved road from a point separated from the entrance of the curved road ahead of the vehicle 1 by the threshold distance Dt. The threshold distance Dt may be a distance greater than zero. By switching landmarks in advance before entering a curved road, the accuracy of the calculated position of the vehicle 1 can be improved compared to switching landmarks after entering a curved road. The threshold distance Dt may be determined according to the following equation (3) according to, for example, the speed V of the vehicle 1 and the allowance time T until the vehicle 1 enters the curve road.
Dt = V × T (3)

The distance calculation unit 49 calculates a distance Dc between the vehicle 1 and the entrance of the curve road ahead. For example, the distance calculation unit 49 may calculate the distance Dc based on the temporary absolute position and orientation of the vehicle 1 and the road information stored in the map information storage unit 20. The distance calculation unit 49 outputs the distance Dc to the landmark selection unit 46.
The landmark selection unit 46 receives the distance Dc from the distance calculation unit 49. The landmark selection unit 46 detects whether or not the distance Dc has changed from a value greater than the threshold distance Dt to a value less than or equal to the threshold distance Dt. When the distance Dc changes from a value greater than the threshold distance Dt to a value less than or equal to the threshold distance Dt, the landmark selection unit 46 switches the landmark to be selected to a combination of landmarks including at least the outer landmark.

  At this time, the landmark selection unit 46 switches the value of the first flag FLG1 from “0” to “1”. The first flag FLG1 indicates whether or not the landmark selection unit 46 selects a landmark including at least the outer landmark. The value “1” indicates that the landmark selection unit 46 selects a combination of landmarks including at least the outer landmark. The value “0” indicates that the landmark selection unit 46 selects a combination of landmarks including either the outer landmark or the inner landmark.

  The curve determination unit 50 determines whether or not the vehicle 1 has entered a curved road. For example, the curve determination unit 50 may determine whether or not the vehicle 1 has entered a curved road based on the temporary absolute position and orientation of the vehicle 1 and the road information stored in the map information storage unit 20. The curve determination unit 50 outputs the determination result to the landmark selection unit 46. The landmark selection unit 46 switches the value of the second flag FLG2 from “0” to “1” when the vehicle 1 enters a curved road. The second flag FLG2 indicates whether or not the vehicle 1 has entered a curved road. The value “0” indicates a state before the vehicle 1 enters the curved road. The value “1” indicates a state after the vehicle 1 enters the curved road.

The landmark selecting unit 46 determines whether or not the second curvature radius Re2 exceeds the first threshold radius Rt1 again after the vehicle 1 enters the curved road. That is, the landmark selection unit 46 determines whether or not the second curvature radius Re2 again exceeds the first threshold radius Rt1 while the value of the second flag FLG2 is “1”. When the second curvature radius Re2 does not exceed the first threshold radius Rt1, the landmark selection unit 46 selects a combination of landmarks including at least the outer landmark.
When the second curvature radius Re2 exceeds the first threshold radius Rt1, the landmark selection unit 46 selects a combination of landmarks including either the outer landmark or the inner landmark. The landmark selection unit 46 switches the value of the first flag FLG1 and the value of the second flag FLG2 from “1” to “0”.

  Thus, the first flag FLG1 is set to “1” when the vehicle 1 arrives at a point separated from the entrance of the curved road by the threshold distance Dt, and thereafter, until the vehicle 1 enters the curved road “ 1 "is maintained. The landmark selection unit 46 selects a combination of landmarks including at least the outer landmark while the first flag FLG1 is “1”. Thereby, even if the second curvature radius Re2 exceeds the first threshold radius Rt1 before the vehicle 1 enters the curved road, the landmark selection unit 46 continues to select a combination of landmarks including at least the outer landmark. Can do.

  Please refer to FIG. The landmark selection unit 46 outputs the identification information of the combination of the selected landmarks to the position calculation unit 44. The landmark selection unit 46 outputs the relative position of the selected landmark to the relative position detection unit 52. The relative position detection unit 52 measures the relative position of the object near the received relative position by the position measurement unit 41, and detects the relative position obtained by the measurement as the landmark position. The relative position detection unit 52 outputs the detected relative position to the position calculation unit 44.

  Please refer to FIG. The position calculation unit 44 receives the relative position of the landmark from the relative position detection unit 52. The position calculation unit 44 receives the identification information of the selected landmark from the landmark selection unit 46. The position calculation unit 44 reads the absolute position of the selected landmark from the map information storage unit 20 based on the identification information received from the landmark selection unit 46. Based on the absolute position of the landmark read from the map information storage unit 20 and the relative position of the landmark received from the relative position detection unit 52, the position calculation unit 44 determines the absolute position of the vehicle 1 and the vehicle 1 at time t. Calculate the orientation.

  Please refer to FIG. Let Pm1 and Pm2 be the absolute positions of the landmarks. Further, the coordinates of Pm1 are (Xm1, Ym1), and the coordinates of Pm2 are (Xm2, Ym2). The position calculation unit 44 assumes that the absolute position and orientation of the vehicle 1 are the temporary absolute position (Xtt, Ytt) and the temporary orientation (φtt), and the relative position of the landmark received from the relative position detection unit 52. Is converted to an absolute position. The absolute positions of the converted landmarks are P1 and P2, respectively. The coordinates of P1 are (X1, Y1), and the coordinates of P2 are (X2, Y2). Further, an error between the absolute positions Pm1 and P1 is D1, and an error between the absolute positions Pm2 and P2 is D2.

上式（４）において、Ｎは、ランドマーク選択部４６に選択されたランドマークの総数を示す。図８の例ではＮ＝２である。 The position calculation unit 44 calculates the error E between the absolute positions Pm1 and Pm2 read from the map information storage unit 20 and the absolute positions P1 and P2 obtained by conversion from the relative positions received from the relative position detection unit 52. It calculates based on Formula (4).

In the above equation (4), N indicates the total number of landmarks selected by the landmark selection unit 46. In the example of FIG. 8, N = 2.

  The position calculation unit 44 calculates the provisional absolute position (Xtt, Ytt) and provisional orientation (φtt) of the vehicle 1 that minimizes the error E as the absolute position and orientation of the vehicle 1 at time t. For example, the position calculation unit 44 may calculate the absolute position and orientation of the vehicle 1 at time t by optimization calculation that minimizes the error E. Instead of minimizing the error E, the position calculation unit 44 calculates a temporary absolute position (Xtt, Ytt) and a temporary orientation (φtt) of the vehicle 1 such that the error E is less than a predetermined value. Also good.

(Operation)
Next, the operation of the position detection device 40 according to the first embodiment will be described. Please refer to FIG. Steps S10 to S17 described in FIG. 9 are repeatedly executed until an ignition switch (not shown) of the vehicle 1 is switched off.
In step S <b> 10, the position measuring unit 41 measures the relative position of the scanning point on the object around the vehicle 1. The peripheral object detection unit 42 detects an object existing around the vehicle 1 based on the measurement position information received from the position measurement unit 41.

  In step S <b> 11, the position calculation unit 44 determines the vehicle 1 at time t based on the absolute position (Xt−1, Yt−1) and the front-rear direction (Φt−1) of the vehicle 1 calculated at time t−1. The temporary absolute position (Xtt, Ytt) and the temporary front-rear direction (Φtt) are calculated. The extraction unit 45 reads the provisional absolute position (Xtt, Ytt) and provisional orientation (Φtt) of the vehicle 1. In step S <b> 12, the extraction unit 45 reads the absolute position of the landmark from the map information storage unit 20.

In step S <b> 13, the extraction unit 45 extracts landmarks from the objects received from the peripheral object detection unit 42. In step S14, the landmark selection unit 46, the first curvature radius estimation unit 47, the threshold value specification unit 48, the distance calculation unit 49, the curve determination unit 50, and the second curvature radius estimation unit 51 execute a landmark selection process.
The landmark selection process S14 shown in FIG. 9 will be described with reference to FIG. In step S20, the landmark selection unit 46 determines whether or not the value of the first flag FLG1 is “1”. When the value of the first flag FLG1 is “1” (step S20: Y), the operation proceeds to step S28. When the value of the first flag FLG1 is “0” (step S20: N), the operation proceeds to step S21.

  In step S21, the first curvature radius estimation unit 47 estimates the first curvature radius Re1. In step S22, the landmark selection unit 46 determines whether or not the first curvature radius Re1 is equal to or less than the first threshold radius Rt1. When the first curvature radius Re1 is equal to or smaller than the first threshold radius Rt1 (step S22: Y), the operation proceeds to step S23. When the first curvature radius Re1 is not equal to or less than the first threshold radius Rt1 (step S22: N), the operation proceeds to step S34.

  In step S23, the distance calculation unit 49 calculates a distance Dc between the vehicle 1 and the entrance of the curve road ahead. In step S24, the landmark selection unit 46 determines whether or not the distance Dc is equal to or less than the threshold distance Dt. When the distance Dc is less than or equal to the threshold distance Dt (step S24: Y), the operation proceeds to step S25. If the distance Dc is not less than or equal to the threshold distance Dt (step S24: N), the operation proceeds to step S34. In step S25, the landmark selection unit 46 switches the value of the first flag FLG1 from “0” to “1”.

  In step S26, the landmark selection unit 46 calculates the number Co of outer landmarks. In step S27, the landmark selection unit 46 selects a combination of landmarks according to the number Co of outer landmarks. When the number Co of outer landmarks is equal to or greater than the predetermined number Ct, the landmark selection unit 46 selects a combination of landmarks including only Ct outer landmarks. When the number Co of outer landmarks is less than the predetermined number Ct, the landmark selection unit 46 selects a combination of landmarks including Co outer landmarks and (Ct-Co) inner landmarks. Thereafter, the operation proceeds to step S15 shown in FIG.

In step S28, the landmark selection unit 46 determines whether or not the value of the second flag FLG2 is “1”. When the value of the second flag FLG2 is “1” (step S28: Y), the operation proceeds to step S31. When the value of the second flag FLG2 is “0” (step S28: N), the operation proceeds to step S29.
In step S29, the curve determination unit 50 determines whether or not the vehicle 1 has entered a curved road. When the vehicle 1 enters the curved road (step S29: Y), the operation proceeds to step S30. If the vehicle 1 has not yet entered the curved road (step S29: N), the operation proceeds to step S27. In step S30, the landmark selection unit 46 switches the value of the second flag FLG2 from “0” to “1”. Thereafter, the operation proceeds to step S27.

In step S31, the second curvature radius estimation unit 51 estimates the second curvature radius Re2. In step S32, the landmark selection unit 46 determines whether or not the second curvature radius Re2 exceeds the first threshold radius Rt1. When the second curvature radius Re2 does not exceed the first threshold radius Rt1 (step S32: N), the operation proceeds to step S27. When the second curvature radius Re2 exceeds the first threshold radius Rt1 (step S32: Y), the operation proceeds to step S33.
In step S33, the landmark selection unit 46 switches the value of the first flag FLG1 and the value of the second flag FLG2 from “1” to “0”. In step S34, the landmark selection unit 46 selects a combination of landmarks including either the outer landmark or the inner landmark. Thereafter, the operation proceeds to step S15 shown in FIG.

Please refer to FIG. In step S15, the relative position detector 52 detects the relative position of the landmark selected in step S14. In step S <b> 16, the position calculation unit 44 determines the absolute position of the vehicle 1 and the direction of the vehicle 1 based on the absolute position of the landmark read from the map information storage unit 20 and the relative position of the landmark received from the relative position detection unit 52. Is calculated.
In step S17, the position detection device 40 determines whether or not the ignition switch (IGN) of the vehicle 1 is off. If the ignition switch is not off (step S17: N), the operation returns to step S10. When the ignition switch is off (step S17: Y), the operation of the position detection device 40 ends.

  The vehicle speed sensor 3 corresponds to a speed detection unit. The map information storage unit 20 corresponds to a position storage unit. The first curvature radius estimation unit 47 corresponds to a curvature radius estimation unit. The first curvature radius Re1 corresponds to the curvature radius estimated by the curvature radius estimation unit. The first threshold radius Rt1 corresponds to the threshold radius. The second curvature radius estimation unit 51 corresponds to a second curvature radius estimation unit. The second curvature radius Re2 corresponds to the second curvature radius. The threshold designation unit 48 corresponds to a road width detection unit and a threshold determination unit.

(Effect of embodiment)
(1) The position detection device 40 according to the first embodiment is a case where the vehicle 1 travels on a curved road, and there is a landmark outside the curved road, the landmark includes at least an outer landmark. Select a combination. That is, the position detection device 40 preferentially selects the outer landmark over the inner landmark when the vehicle 1 travels on a curved road. The position detection device 40 calculates the absolute position of the vehicle 1 using the selected combination of landmarks. For this reason, the influence which the measurement error of the relative position of the landmark while the vehicle 1 travels on the curved road has on the accuracy of the calculated position of the vehicle 1 can be reduced.

(2) The position detection device 40 according to the first embodiment selects a combination of landmarks including at least an outer landmark when the traveling road ahead of the vehicle 1 is a curved road. For this reason, when the vehicle 1 travels and enters a forward curve road, the influence of the measurement error of the relative position of the landmark on the accuracy of the calculated position of the vehicle 1 can be reduced.
(3) The position detection device 40 according to the first embodiment selects a combination of landmarks including at least the outer landmark from a point separated from the curve road ahead of the vehicle 1 by the threshold distance Dt. For this reason, the accuracy of the calculated position of the vehicle 1 can be improved rather than switching the landmark after entering the curved road.

(4) The position detection device 40 according to the first embodiment estimates the first curvature radius Re1 of the curved road, and includes a land including at least an outer landmark when the first curvature radius Re1 is equal to or less than the first threshold radius Rt1. Select a combination of marks. Therefore, landmark candidates that can be used for calculating the position of the vehicle 1 when the radius of curvature is relatively large and the change amount difference is small can be increased.
(5) The position detection device 40 according to the first embodiment selects a combination of landmarks including at least two outer landmarks as a combination of landmarks including at least the outer landmark. For this reason, the accuracy of the calculated position of the vehicle 1 can be increased.

  (6) The position detection device 40 according to the first embodiment selects a combination of landmarks including at least the outer landmark when the number Co of outer landmarks is equal to or greater than the predetermined number Ct. When the number of outer landmarks Co is less than a predetermined number Ct, a combination of landmarks including an outer landmark and an inner landmark is selected. For this reason, even if the number Co of outer landmarks is less than the predetermined number Ct, the position of the vehicle 1 can be calculated using a combination of Ct landmarks by using the inner landmarks together.

(7) The position detection device 40 according to the first embodiment selects a combination of landmarks including at least two outer landmarks when there are two or more outer landmarks. The position detection device 40 selects a combination of landmarks including an outer landmark and an inner landmark when there is one outer landmark. Therefore, even if the number of outer landmarks is less than two, the position of the vehicle 1 can be calculated using a combination of two landmarks by using the inner landmarks together.
(8) The position detection device 40 according to the first embodiment selects a combination of landmarks including only the outer landmark as a combination of landmarks including at least the outer landmark. For this reason, the fall of the precision of the calculation position of the vehicle 1 by using an inner landmark can be avoided.

(Modification)
The position calculation unit 44 may calculate the average coordinates of the relative positions of the landmarks received from the relative position detection unit 52 when calculating the absolute position of the vehicle 1. Further, the position calculation unit 44 may calculate the average coordinates of the absolute position of the landmark received from the landmark selection unit 46. The position calculation unit 44 may calculate the absolute position of the vehicle 1 by subtracting the X coordinate value and the Y coordinate value of the average coordinate of the relative position from the X coordinate value and the Y coordinate value of the average coordinate of the absolute position. According to such a calculation method, the absolute position of the vehicle 1 can be calculated more easily.

  The landmark selection unit 46 may output the relative position of the selected landmark to the position calculation unit 44. The position calculation unit 44 may calculate the absolute position of the vehicle 1 based on the relative position of the landmark received from the landmark selection unit 46 and the absolute position of the landmark. According to such a configuration, since the detection of the relative position of the landmark by the relative position detection unit 52 can be omitted, the absolute position of the vehicle 1 can be calculated more easily.

The movement amount calculation unit 43 may calculate the yaw rate based on the speed of the vehicle 1 and the steering angle of the steering wheel 4 detected by the steering angle sensor 5. The movement amount calculation unit 43 may calculate the rotation amount dφ of the vehicle 1 based on the calculated yaw rate. The yaw rate sensor 7 can be omitted by calculating the yaw rate based on the speed and the steering angle.
The extraction unit 45 may determine an object whose size is equal to or less than a predetermined value S as a landmark. For example, the extraction unit 45 may calculate the distance between points that are the farthest among the scanning points belonging to the segment as the size of the object. For example, the predetermined value S may be 0.5 m. By determining whether or not the landmark is based only on the size information of the object, the landmark can be extracted by simpler calculation.

The first curvature radius estimation unit 47 may estimate the first curvature radius Re1 based on the steering angle detected by the steering angle sensor 5. The second curvature radius estimation unit 51 may estimate the second curvature radius Re2 based on the steering angle detected by the steering angle sensor 5.
The first curvature radius estimation unit 47 may estimate the first curvature radius Re1 based on an image of a white line on the road ahead of the vehicle 1 taken by an in-vehicle camera (not shown). The distance calculation unit 49 may calculate the distance Dc based on the white line image of the road ahead of the vehicle 1 taken by the in-vehicle camera. The curve determination unit 50 may determine whether or not the vehicle 1 has entered the curved road based on the white line image of the road ahead of the vehicle 1 taken by the in-vehicle camera. The threshold designation unit 48 may detect the road width W of the curved road based on the white line image of the road ahead of the vehicle 1 taken by the in-vehicle camera.

  The first threshold radius Rt1 may be a fixed value calculated based on a fixed estimated value of the road width W and a fixed estimated value of the speed V of the vehicle 1. The threshold designation unit 48 may dynamically calculate the first threshold radius Rt1 based on the road width W read from the map information storage unit 20 and the assumed value of the fixed speed V of the vehicle 1. The threshold designation unit 48 may dynamically calculate the first threshold radius Rt1 based on the speed information output from the vehicle speed sensor 3 and the assumed value of the fixed road width W.

(Second Embodiment)
(Constitution)
Next, the position detection device 40 according to the second embodiment of the present invention will be described. The functional configuration of the position detection device 40 according to the second embodiment of the present invention is the same as the functional configuration shown in FIG. The threshold designation unit 48 designates a second threshold radius Rt2 that is larger than the first threshold radius Rt1. Even if the second curvature radius Re2 exceeds the first threshold radius Rt1 while the vehicle 1 is traveling on a curved road, the landmark selection unit 46 changes the outer landmark until the second curvature radius Re2 exceeds the second threshold radius Rt2. Select at least the combination of landmarks to include.

For this reason, it is possible to prevent the landmark selection unit 46 from selecting a combination of landmarks including either the outer landmark or the inner landmark in the middle of the curved road. For example, the second threshold radius Rt2 may be set so that a road having the second threshold radius Rt2 can be regarded as a straight road. For example, the second threshold radius Rt2 may be 1000 m.
The landmark selection unit 46 determines whether or not the second curvature radius Re2 exceeds the second threshold radius Rt2 while the value of the second flag FLG2 is “1”. When the second curvature radius Re2 does not exceed the second threshold radius Rt2, the landmark selection unit 46 selects a combination of landmarks including at least the outer landmark. When the second curvature radius Re2 exceeds the second threshold radius Rt2, the landmark selection unit 46 selects a combination of landmarks including either the outer landmark or the inner landmark. The landmark selection unit 46 switches the value of the first flag FLG1 and the value of the second flag FLG2 from “1” to “0”.

  With reference to FIG. 11, changes in landmarks selected by the landmark selection unit 46 when the vehicle 1 travels on a curved road will be described. Reference numeral 1a indicates the position of the vehicle 1 at time t1, and reference numeral 1b indicates the position of the vehicle 1 at time t2 after time t1. Reference numeral 1c indicates the position of the vehicle 1 at time t3 after time t2, and reference numeral 1d indicates the position of the vehicle 1 at time t4 after time t3.

  Reference numeral 80 indicates a straight section connected to the entrance of the curved road. Reference numeral 81 indicates a first curve section having a curvature radius of 90 m. Reference numeral 82 indicates a second curve section having a curvature radius of 120 m. Reference numeral 83 indicates a straight section connected to the exit of the curved road. Reference numerals 90, 92, 94, and 96 denote outer landmarks. Reference numerals 91, 93, 95, and 97 denote inner landmarks. Assume that the first threshold radius Rt1 is 100 m and the second threshold radius Rt2 is 1000 m.

  When the distance Dc between the vehicle 1 traveling in the straight section 80 at the time t1 and the entrance of the curved road is equal to or smaller than the threshold distance Dt, the landmark selection unit 46 selects a combination of landmarks including at least the outer landmark. . For example, the landmark selection unit 46 selects a combination of landmarks including only the two outer landmarks 90 and 92. While the vehicle 1 is traveling in the first curve section 81 at time t2, the second curvature radius Re2 is 90 m and does not exceed the first threshold radius Rt1, so the landmark selection unit 46 sets the outer landmark at least. Select a combination of landmarks to include. For example, the landmark selection unit 46 selects a combination of landmarks including only the two outer landmarks 92 and 94.

  While the vehicle 1 is traveling on the second curve section 82 at time t3, the second curvature radius Re2 is 120 m, which exceeds the first threshold radius Rt1. However, since the second curvature radius Re2 does not exceed the second threshold radius Rt2, the landmark selection unit 46 selects a combination of landmarks including at least the outer landmark. For example, the landmark selection unit 46 selects a combination of landmarks including only the two outer landmarks 94 and 96.

When the vehicle 1 travels in the straight section 83 at time t4, the second curvature radius Re2 exceeds the second threshold radius Rt2, and therefore the landmark selection unit 46 selects a landmark including either the outer landmark or the inner landmark. Select a combination. For example, the landmark selection unit 46 selects a combination of landmarks including the outer landmark 96 and the inner landmark 97. The landmark selection unit 46 may select a combination of landmarks including only the inner landmarks 95 and 97.
The second threshold radius Rt2 corresponds to the second threshold radius.

(Operation)
Next, the operation of the position detection device 40 according to the second embodiment will be described. Please refer to FIG. The landmark selection process shown in FIG. 12 may be executed as the landmark selection process S14 shown in FIG. The operations in steps S40 to S51 are the same as the operations in steps S20 to S31 shown in FIG.
In step S52, the landmark selection unit 46 determines whether or not the second curvature radius Re2 exceeds the first threshold radius Rt1. When the second curvature radius Re2 does not exceed the first threshold radius Rt1 (step S52: N), the operation proceeds to step S47. When the second curvature radius Re2 exceeds the first threshold radius Rt1 (step S52: Y), the operation proceeds to step S53.

  In step S53, the landmark selection unit 46 determines whether or not the second curvature radius Re2 exceeds the second threshold radius Rt2. When the second curvature radius Re2 does not exceed the second threshold radius Rt2 (step S53: N), the operation proceeds to step S47. When the second curvature radius Re2 exceeds the second threshold radius Rt2 (step S53: Y), the operation proceeds to step S54. The operations in steps S54 and S55 are the same as the operations in steps S33 and S34 shown in FIG.

(Effect of embodiment)
The position detection device 40 according to the second embodiment selects a combination of landmarks including at least the outer landmark until the vehicle 1 enters the curved road and the second curvature radius Re2 exceeds the second threshold radius Rt2. For this reason, it is possible to prevent the position of the vehicle 1 from being calculated using only the inner landmark in the middle of the curved road.
Although the present invention has been described with reference to a limited number of embodiments, the scope of rights is not limited thereto, and modifications of the embodiments based on the above disclosure are obvious to those skilled in the art.

3 Vehicle speed sensor 5 Steering angle sensor 6 Position measurement sensor 7 Yaw rate sensor 10 Controller 20 Map information storage unit 30 Navigation device 40 Position detection device 41 Position measurement unit 42 Peripheral object detection unit 43 Movement amount calculation unit 44 Position calculation unit 45 Extraction unit 46 Landmark selection unit 47 First curvature radius estimation unit 48 Threshold specification unit 49 Distance calculation unit 50 Curve determination unit 51 Second curvature radius estimation unit 52 Relative position detection unit

Claims (10)

A position measuring unit for measuring a relative position of an object around the moving body with respect to the moving body;
A relative position detection unit for detecting a relative position of a landmark among the relative positions measured by the position measurement unit;
A location storage unit that stores locations of landmarks on the map;
When the moving body travels on a curved road, and there is a landmark outside the curved road, a landmark selection that selects a combination of landmarks including at least a landmark outside the curved road And
A position calculating unit that calculates a position of the moving body on the map based on the relative position of the landmark selected by the landmark selecting unit and the position on the map;
A position detection device comprising:   The position detection device according to claim 1, wherein the landmark selection unit selects the combination when a traveling road ahead of the moving body is the curved road.   The position detection according to claim 2, wherein the landmark selection unit selects the combination when the moving body reaches a position separated from the curved road ahead of the moving body by a threshold distance. apparatus. A curvature radius estimation unit for estimating a curvature radius of the curved road,
The said landmark selection part selects the said combination, when the said curvature radius estimated by the said curvature radius estimation part is below a threshold radius, The Claim 1 characterized by the above-mentioned. Position detection device. A second radius-of-curvature estimator that estimates a radius of curvature of a travel path at a point where the moving body is traveling as a second radius of curvature;
The landmark selection unit selects the combination until the moving body enters the curved road and the second curvature radius estimated by the second curvature radius estimation unit exceeds a second threshold radius. ,
The position detection device according to claim 4, wherein the second threshold radius is larger than the threshold radius.   The position detection apparatus according to claim 1, wherein the combination includes at least two landmarks outside the curved road.   When the number of landmarks outside the curved road is a predetermined number or more, the combination includes at least a predetermined number of landmarks outside the curved road, and the landmark outside the curved road is the 6. The position detection device according to claim 1, wherein when the number is less than a predetermined number, the combination includes a landmark outside the curved road and a landmark inside. .   When there are two or more landmarks outside the curved road, the combination includes at least two landmarks outside the curved road, and one landmark is outside the curved road. 8. The position detection apparatus according to claim 7, wherein the combination includes a landmark outside the curved road and a landmark inside.   The position detection device according to any one of claims 1 to 6, wherein the combination includes only a landmark outside the curved road. Measuring the relative position of an object around the moving body relative to the moving body;
Detect the relative position of the landmark among the measured relative positions,
When the moving body travels on a curved road, and there is a landmark outside the curved road, a combination of landmarks including at least a landmark outside the curved road is selected,
Based on the relative position of the selected combination of landmarks and the position on the map of the selected combination of landmarks among the positions on the map of landmarks previously stored in the storage device, A position detection method characterized by calculating a position on a map.

Images