JP2006038755A - Device for detecting object around vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for detecting an object around a vehicle for removing the influence due to error when finding the relative position of the object, even when the error is included at any one relative position of grouped detection points. <P>SOLUTION: Concerning the object lying forward of own vehicle, the device for detecting the object around the vehicle detects a position coordinate in each detection point (step S1), and prepares a group of the detection points at each object, by collecting the detection points satisfying a prescribed condition to perform grouping (steps S6 and S10). Concerning each of the grouped detection points, the device calculates a reliability determining value R1(t) or R2(t) (steps S7 and S11), and selects representative detection points from among the detection points grouped in the same group, on the basis of the value (steps S8 and S12). The device determines the relative position of the object, on the basis of the relative position of the selected representative detection points (steps S9 and S13). <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an apparatus for detecting an object mounted on a vehicle and present around the vehicle.

  A method of detecting a relative positional relationship with another vehicle traveling in front of the host vehicle using a laser radar or the like is widely known. For example, in the apparatus disclosed in Patent Document 1, the irradiation range of the laser radar is divided for each predetermined angle, and any one of a plurality of distance data detected in each divided area is selected and used. To compress the distance data. Based on the distance data thus compressed, the relative position and the relative speed with respect to the host vehicle are obtained for the detection points in each region. Furthermore, when the relative positions and relative speeds of a plurality of detection points are within a certain range, the detection points are grouped into one group and processed as representing the same vehicle. At this time, the relative position of the vehicle is obtained using the average value of the relative positions of the detection points included in the same group. In this way, the relative positional relationship with other vehicles is obtained.

JP 2001-343460 A

  In the device disclosed in Patent Document 1, when an error is included in the value of any of the relative positions among the grouped detection points, the other vehicle obtained using the average value of each detection point There is a problem that the error also affects the positional relationship.

  A vehicle surrounding object detection device according to the present invention includes a position detection unit that detects a relative position of an object around the host vehicle with respect to the host vehicle at one or a plurality of detection points, and a detection point in which the relative position is detected by the position detection unit. Among the detection points satisfying a predetermined grouping condition and creating a group of detection points for each object, and the reliability of each relative position for each of the detection points grouped by the grouping means. A reliability calculation means for calculating, a representative point selection means for selecting a representative detection point from detection points grouped in the same group based on the reliability value calculated by the reliability calculation means, and a representative detection point Position determining means for determining the relative position of the object based on the relative position.

  According to the present invention, with respect to an object around the own vehicle, the relative position with respect to the own vehicle is detected at one or a plurality of detection points, and detection points satisfying a predetermined grouping condition are grouped to create a group of detection points for each object. By doing so, grouping is performed. For each detection point grouped by this grouping, the reliability of each relative position is calculated, and based on the calculated reliability value, representative detection points are selected from the detection points grouped in the same group. Based on the relative position of the representative detection point, the relative position of each object is determined. Since the positional relationship with each object representing another vehicle or the like is thus determined, even if the relative position value of any of the grouped detection points includes an error, the influence of the error is affected. The positional relationship with other vehicles can be obtained by removing.

  FIG. 1 shows the configuration of a vehicle surrounding object detection device (hereinafter referred to as the present device) according to an embodiment of the present invention. This apparatus is mounted on a vehicle and includes a laser radar 1, a vehicle behavior detection unit 2, a calculation unit 3, and a vehicle control unit 4. The apparatus detects an object such as another vehicle (hereinafter simply referred to as an object) existing in front of the host vehicle by using these components, and controls the traveling of the host vehicle based on the detection result. Each component will be described below.

  The laser radar 1 is installed at the front end of the host vehicle, irradiates laser light toward the front of the host vehicle, and detects reflected light generated by the laser beam. The laser radar 1 is a kind of what is called a scanning laser radar, and can scan (scan) the laser light within a predetermined scanning range by changing the irradiation direction of the laser light.

  When an object exists within the scan range of the laser radar 1, the irradiated laser light is reflected by the object. At this time, since the irradiation direction is changed by the scanning of the laser beam, a plurality of reflection points are generated on the object. By detecting the reflected light by the laser radar 1 using the plurality of reflection points as detection points and measuring the time from when the laser light is irradiated to each detection point until the reflected light is detected, the own vehicle To the respective detection points are calculated. Further, the direction of each detection point with respect to the host vehicle is obtained from the irradiation direction of the laser light when the reflected light is detected. Based on the distance and direction of each detection point thus obtained, the relative position of the object ahead of the host vehicle with respect to the host vehicle is detected at each detection point. The detection result of the relative position at each detection point is output from the laser radar 1 to the calculation unit 3 as position information and reflection intensity information representing the measurement value at each detection point together with the intensity of the reflected light detected at each detection point. Is done.

  FIG. 2 is a diagram showing an example of mounting the laser radar 1 on the own vehicle. The laser radar 1 mounted near the center of the front end of the host vehicle 11 irradiates the laser beam in front of the host vehicle 11, that is, in the direction on the left side in the figure. The optical axis of the laser beam at this time is represented as an optical axis L. Also, the left and right coordinate axes with respect to the host vehicle 11 are defined as the X axis, the vertical coordinate axis is defined as the Y axis, and the longitudinal coordinate axis is defined as the Z axis. When the optical axis L and the Z axis coincide with each other, the laser beam is irradiated toward the front surface of the host vehicle 11. The origin position 11a of each coordinate axis is assumed to be fixed with respect to the host vehicle 11.

  The laser radar 1 can rotate the optical axis L within a predetermined angle in the left-right direction (around the Y axis) around the origin of each coordinate axis. Thereby, the irradiation direction of a laser beam changes horizontally within a predetermined angle. At this time, the surface along which the optical axis L rotates is called a scanning surface, and this scanning surface is equal to the horizontal plane (XZ plane). In this way, the laser radar 1 scans the laser beam within a predetermined scan range along the scanning surface.

  The vehicle behavior detection unit 2 acquires information (host vehicle travel information) for representing the travel state of the host vehicle by detecting the behavior of the host vehicle. The vehicle behavior detection unit 2 includes a wheel speed sensor that detects the wheel speeds of the left and right rear wheels, and a steering angle sensor that detects a steering angle. Then, the yaw rate and slip angle of the host vehicle calculated based on the sensor signals from these sensors are acquired as host vehicle travel information. The own vehicle travel information acquired by the vehicle behavior detection unit 2 is output to the calculation unit 3.

  The calculation unit 3 is configured by a microcomputer including a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), an input / output I / F, and the like mounted in the host vehicle. The calculation unit 3 is based on the position information and reflection intensity information of each detection point output from the laser radar 1 and the own vehicle travel information output from the vehicle behavior detection unit 2, that is, the yaw rate and slip angle of the own vehicle. Thus, a process of detecting an object existing on the estimated course ahead of the host vehicle (hereinafter referred to as an object detection process) is executed. The result of the object detection process is output from the calculation unit 3 to the vehicle control unit 4. For example, the relative distance from the host vehicle to the object and the relative speed of the object with respect to the host vehicle are output from the calculation unit 3 to the vehicle control unit 4 as a result of the object detection process. The specific contents of the object detection process will be described later.

  The vehicle control unit 4 controls the host vehicle based on the result of the object detection process output from the calculation unit 3. For example, as described above, the relative distance from the host vehicle and the relative speed with respect to the host vehicle are output from the calculation unit 3 as the object detection processing result. When the relative distance is equal to or less than a predetermined value and the relative speed is equal to or greater than a predetermined value, the traveling speed of the host vehicle is reduced by controlling the engine speed, the brake operation, and the like of the host vehicle. . By doing in this way, based on the result of an object detection process, the own vehicle can be controlled so that the distance between the own vehicle and the other vehicle can be maintained at a certain level or more.

  Note that the content output as the object detection processing result from the calculation unit 3 to the vehicle control unit 4 and the content for controlling the host vehicle in the vehicle control unit 4 may be other than the above. For example, when the relative positional relationship between the host vehicle and the object is output from the calculation unit 3 to the vehicle control unit 4 and the relative positional relationship satisfies a predetermined condition, for example, when the traveling direction of the host vehicle is within a predetermined distance The vehicle control unit 4 can issue a warning to the driver.

  Next, the contents of the object detection process executed in the calculation unit 3 will be described by taking the case shown in FIG. 3 as an example. FIG. 3 shows how the laser radar 1 installed in the host vehicle 11 of FIG. 2 detects the objects A and B ahead of the host vehicle 11 existing in the scan range at time t−1 and time t, respectively. Is shown. Note that the X axis and the Z axis in FIG. 3 are defined in the same direction as in FIG.

  The laser radar 1 scans the laser beam within a predetermined scan range along the above-described scanning plane, and outputs position information and reflection intensity information obtained for each detection point to the calculation unit 3. Here, reflected light is detected at each of the detection points a, b, c, d, e, and f shown in FIG. 3 at time t−1, and the position information and reflection intensity information at each of these detection points is the laser. It is assumed that the signal is output from the radar 1 to the calculation unit 3. Hereinafter, the position coordinates of each detection point at time t−1 are represented as (Xω (t−1), Zω (t−1)), and the reflection intensity at each detection point is represented as Pω (t−1). However, any of a, b, c, d, e, and f enters ω. In FIG. 3, each detection point is shown in a circular range having a different size. This is for indicating the magnitude of the reflection intensity Pω (t−1) at each detection point. Actually, each detection point is located at the center of each circular range. For example, since the detection points b and d are larger in the circular range than the other detection points, the values of the reflection intensities Pb (t−1) and Pd (t−1) are the values of the other detection points. It can be seen that it has a larger value than the reflection intensity.

  The calculation unit 3 groups the detection points detected at time t−1 and groups the detection points representing the same object into one group. Here, the detection points c and f represent the object A, and the detection points a, b, d, and e represent the object B, and are grouped in the same group. Note that one detection point may be grouped.

When the detection points of the same object are grouped as described above, based on the position information and reflection intensity information of each detection point output from the laser radar 1, the objects A and B are used as parameters of each object at time t-1. The lateral width (width in the X-axis direction), position, relative velocity, and reflection intensity are respectively calculated. In the following, the horizontal width, position coordinates, relative speed, and reflection intensity of the object A at time t−1 are respectively expressed as follows.
Width of object A: WA (t−1)
Position coordinates of the object A: (XA (t-1), ZA (t-1))
Relative velocity of object A: VA (t−1)
Reflection intensity of object A: PA (t-1)

Similarly, the horizontal width, position coordinates, relative velocity, and reflection intensity of the object B at time t−1 are respectively expressed as follows.
Width of object B: WB (t-1)
Position coordinates of the object B: (XB (t-1), ZB (t-1)),
Relative velocity of object B: VB (t-1)
Reflection intensity of object B: PB (t−1)

  A specific grouping method for grouping the detection points as described above and a specific calculation method for calculating the parameters of each object will be described in detail below by taking the case at time t as an example. To do.

  Next, at time t, reflected laser light is detected for each of the detection points a to f shown in FIG. 3 in the same manner as at time t−1, and position information and reflection intensity information at each detection point are obtained. It is assumed that the laser radar 1 outputs to the calculation unit 3. Hereinafter, the position coordinates of each detection point at time t are represented as (Xω (t), Zω (t)) (where ω is any one of a to f), and the reflection intensity at each detection point is represented by Pω (t). It expresses.

When the position information and the reflection intensity information about the detection points a to f at the time t are input from the laser radar 1 to the calculation unit 3, the calculation unit 3 firstly calculates the position of the object A at the time t-1. Based on (XA (t−1), ZA (t−1)) and relative velocity VA (t−1), estimated coordinates (XA ′ (t), ZA ′ ( t)) is calculated by the following equations (1) and (2).
ZA ′ (t) = ZA (t−1) + VA (t−1) · cos (Slip) (1)
XA ′ (t) = XA (t−1) + VA (t−1) · sin (Slip)
+ ZA '(t) tan (Yaw) (2)

  In the above formulas (1) and (2), Yaw and Slip represent the yaw rate and slip angle of the host vehicle output from the vehicle behavior detector 2, respectively. That is, the position of the object A at the current time t is estimated by adding the deviation due to the slip angle of the host vehicle and the deviation due to the yaw rate to the position of the object A obtained at the previous time t-1.

Further, the estimated coordinates (XA ′ (t), ZA ′ (t)) of the position of the object B at time t are also calculated by the following equations (3) and (4) in the same manner as described above.
ZB ′ (t) = ZB (t−1) + VB (t−1) · cos (Slip) (3)
XB ′ (t) = XB (t−1) + VB (t−1) · sin (Slip)
+ ZB '(t) tan (Yaw) (4)

  When the positions of the objects A and B at the time t are estimated as described above, the detection points are grouped for each object based on the estimated coordinate values. Here, it is known that the characteristic of the reflected light detected by the laser radar 1 changes depending on the relative distance from the own vehicle to the detection point. In general, reflected light having a relatively weak reflection intensity and a large variation is detected from a detection point of an object far away from the host vehicle. The higher the reflection intensity, the smaller the error of the detected position information. On the other hand, reflected light with relatively high reflection intensity and small variation is generally detected from detection points of objects close to the host vehicle, and its position information error is small. In such a case, a detection result with a large error may be obtained even at a relatively short distance. For this reason, in the present apparatus, the detection points are grouped by different methods depending on whether the distance from the host vehicle is long or close.

  In the following description, it is assumed that the object A exists farther from the predetermined position when viewed from the host vehicle, and the object B exists closer to the predetermined position. In the following, the relative distance from the vehicle to the predetermined position is represented as L. The value of L is predetermined and is 60 m, for example. At this time, with respect to the object A, with respect to each detection point existing around the estimated coordinates calculated by the above formulas (1) and (2), the reflection intensity of each detection point and the previous time for the object A A difference value from the reflection intensity calculated at t−1 is calculated for each detection point. Then, the detection points whose calculated difference values are equal to or less than a predetermined value are grouped together as ones representing the same object A, thereby grouping the objects A.

Specifically, for the detection points whose position coordinates satisfy the following expressions (5) and (6), the detection points whose reflection intensity satisfies the expression (7) are assigned to the same group representing the object A. Group as included. In Equation (6), the value of ZA ′ (t) is assumed to be equal to or greater than the predetermined value L. Such a grouping method is hereinafter referred to as long-distance grouping.
XA ′ (t) − (1/2) · WA (t−1) ≦ Xω (t)
≦ XA ′ (t) + (1/2) · WA (t−1) (5)
ZA ′ (t) − (1/2) · Z0 ≦ Zω (t)
≦ ZA ′ (t) + (1/2) · Z0 (6)
| Pω (t) −PA (t−1) | ≦ Pt1 (7)

  When the long-distance grouping is performed on the object A as described above, a grouping range 40 as indicated by a broken line in FIG. The grouping range 40 is set in a range of a lateral width WA (t−1) and a front-rear width Z0 around the estimated coordinates (XA ′ (t), ZA ′ (t)) of the object A. Of the detection points located within the grouping range 40, detection points satisfying Expression (7) are grouped into the same group. A predetermined value is set in advance for Z0. That is, according to the equation (6), Zω (t) is at least a predetermined value (L−Z0 / 2) or more. Pt1 is a threshold value, and a predetermined value is set in advance.

  As described above, the detection point where the value of Zω (t) indicating the relative distance from the host vehicle is equal to or larger than the predetermined value (L−Z0 / 2) is obtained by performing long-distance grouping, and thereby within the grouping range 40. Further, the difference | Pω (t) between the reflection intensity Pω (t) at the detection point and the reflection intensity of the previous object A, that is, the reflection intensity PA (t−1) at the previous representative detection point. The detection points whose −PA (t−1) | is equal to or less than a predetermined threshold value Pt1 are grouped together as one representing the object A. At this time, as described above, one detection point may be grouped. The representative detection points will be described later.

On the other hand, for the object B, the detection points whose reflection intensity is a predetermined value or more are the same for each detection point existing around the estimated coordinates represented by the above formulas (3) and (4). The object B is grouped by grouping it into one group as representing the object B. Specifically, for the detection points whose position coordinates satisfy the following expressions (8) and (9), the detection points whose reflection intensity satisfies the expression (10) are assigned to the same group representing the object B. Group as included. In Equation (9), the value of ZB ′ (t) is assumed to be smaller than the predetermined value L described above. Such a grouping method is hereinafter referred to as short distance grouping.
XB ′ (t) − (1/2) · WB (t−1) ≦ Xω (t)
≦ XB ′ (t) + (1/2) · WB (t−1) (8)
ZB ′ (t) − (1/2) · Z0 ≦ Zω (t)
≦ ZB ′ (t) + (1/2) · Z0 (9)
Pω (t) ≧ Pt2 (10)

  When short distance grouping is performed on the object B as described above, a grouping range 50 as indicated by a broken line in FIG. The grouping range 50 is set in a range of a lateral width WB (t−1) and a front-rear width Z0 with the estimated coordinates (XB ′ (t), ZB ′ (t)) of the object B as the center. Of the detection points located within the grouping range 50, detection points satisfying the expression (10) are grouped into the same group. Z0 is the same as that in the above-described formula (6), and a predetermined value is set in advance. That is, according to the equation (9), Zω (t) is at least less than a predetermined value (L + Z0 / 2). Pt2 is a threshold value, and a predetermined value is set in advance.

  As described above, detection points where the value of Zω (t) indicating the relative distance from the host vehicle is less than the predetermined value (L + Z0 / 2) are positioned within the grouping range 50 by performing short-distance grouping. Furthermore, the detection points whose reflection intensity Pω (t) is equal to or greater than a predetermined threshold value Pt2 are grouped together as one representing the object B. Also at this time, as described above, one detection point may be grouped.

  In this way, at the time t, the detection points are grouped based on the estimated coordinate values of the objects A and B. According to the above formulas (5) to (7), the detection points c and f are grouped into the same group as representing the object A. Further, the detection points a, b, d, and e are grouped into the same group as those representing the object B by the equations (8) to (10).

  After the detection points are grouped as described above, the lateral width, position coordinates, relative velocity, and reflection intensity are calculated as parameters at time t for each of the object A and the object B. At this time, as described above, the characteristic of the reflected light that is detected changes depending on the distance of the object from the host vehicle. Therefore, the position coordinate and the reflection intensity in the Z-axis direction are close to the object A that is far away from the object A. A different calculation method is used for each.

For the object A, the lateral width WA (t), the position coordinates (XA (t), ZA (t)), the relative velocity VA (t), and the reflection intensity PA (t) at time t are expressed by the following equations, respectively. calculate.
WA (t) = MAX {Xω (t)} − MIN {Xω (t)} (11)
XA (t) = 1/2 · [MAX {Xω (t)} + MIN {Xω (t)}] (12)
ZA (t) = R1MAX {Zω (t)} (13)
VA (t) = ZA (t) −ZA (t−1) (14)
PA (t) = R1MAX {Pω (t)} (15)

  In the above formulas (11) and (12), MAX {Xω (t)} and MIN {Xω (t)} are the maximum and minimum X coordinate values of the detection points grouped in the same group. Respectively. Here, in FIG. 4, the detection point c and the detection point f of the same group representing the object A are located further upward in the drawing. In FIG. 4, the X axis and the Z axis are defined in the same direction as in FIG. 3, and the X coordinate value is larger as the detection point is located above the paper surface. Therefore, it can be seen that the X coordinate value Xc (t) of the detection point c is larger than the X coordinate value Xf (t) of the detection point f. Therefore, MAX {Xω (t)} = Xc (t) and MIN {Xω (t)} = Xf (t). By substituting this into the equations (11) and (12), the lateral width of the object A at time t is WA (t) = Xc (t) −Xf (t), and the X coordinate value is XA (t) = 1 / 2 · {Xc (t) + Xf (t)}, respectively.

In the above equations (13) and (15), R1MAX {Zω (t)} and R1MAX {Pω (t)} are expressed by the following equation (16) among the detection points grouped in the same group. The Z coordinate and the reflection intensity of the largest reliability judgment value R1 (t) represented are respectively represented.
R1 (t) = Pω (t) (16)

  From the above equation (16), the reliability judgment value R1 (t) is obtained for each detection point representing the object A. Thereby, the reliability of each relative position is calculated about each of the detection points put together in one group. Here, the reliability judgment value R1 (t) according to the equation (16) is the reflection intensity Pω (t) at each detection point. That is, among the detection points in the same group representing the object A, the Z coordinate and the value of the reflection intensity of the highest reflection intensity are selected in the equations (13) and (15), and the object A is represented on behalf of each detection point. Z coordinate and reflection intensity. Hereinafter, such a detection point is referred to as a representative detection point of the object A.

  The manner in which the representative detection point of the object A is selected will be described with reference to FIG. In FIG. 4, both the detection point c and the detection point f are within the grouping range 40 and are grouped into the same group representing the object A. Here, the size of the circular range of each detection point in FIG. 4 represents the value of the reflection intensity Pω (t) at that detection point, as in FIG. In FIG. 4, since the detection point c has a larger circular range than the detection point f, the reflection intensity Pc (t) at the detection point c is higher than the reflection intensity Pf (t) at the detection point f. It is clear that That is, of the detection points c and f in the same group representing the object A, the detection point c has the largest reliability judgment value R1 (t). At this time, the detection point c is set as a representative detection point of the object A. As a result, according to the equations (13) and (15), the Z coordinate value of the object A at time t is calculated as ZA (t) = Zc (t), and the reflection intensity is calculated as PA (t) = Pc (t). .

  Note that, at the detection points grouped as representing the object A by the long distance grouping, the value of Zω (t) is equal to or greater than the predetermined value (L−Z0 / 2) as described above. That is, as described above, for a detection point where the value of Zω (t) indicating the relative distance from the host vehicle is equal to or greater than a predetermined value (L−Z0 / 2), the reflection intensity Pω (t ) To determine the reliability judgment value R1 (t), the reliability of each detection point is calculated.

On the other hand, for the object B, the horizontal width WB (t), the position coordinates (XB (t), ZB (t)), the relative velocity VB (t) and the reflection intensity PB (t) at time t are respectively expressed by the following equations. Calculated by
WB (t) = MAX {Xω (t)} − MIN {Xω (t)} (17)
XB (t) = 1/2 · [MAX {Xω (t)} + MIN {Xω (t)}] (18)
ZB (t) = R2MIN {Zω (t)} (19)
VB (t) = ZB (t) −ZB (t−1) (20)
PB (t) = R2MIN {Pω (t)} (21)

  In the above equations (17) and (18), MAX {Xω (t)} and MIN {Xω (t)} are the detections grouped in the same group as in the above equations (11) and (12). The maximum and minimum values of the X coordinate values of the points are respectively shown. Here, in FIG. 5, among the detection points a, b, d and e of the same group representing the object B, the detection point a is located at the uppermost position in the figure, and the detection point e is located at the lowermost position in the figure. ing. In FIG. 5, the X axis and the Z axis are defined in the same direction as in FIG. 3 and FIG. 4, and the X coordinate value is larger as the detection point is located above the paper surface. Therefore, it can be seen that the X coordinate value Xa (t) of the detection point a is larger than the other detection points, and the X coordinate value Xe (t) of the detection point e is smaller than the other detection points. Therefore, MAX {Xω (t)} = Xa (t) and MIN {Xω (t)} = Xe (t). By substituting this into the equations (17) and (18), the lateral width of the object B at time t is WB (t) = Xa (t) −Xe (t), and the X coordinate value is XB (t) = 1 / 2 · {Xa (t) + Xe (t)}, respectively.

In the above equations (19) and (21), R2MIN {Zω (t)} and R2MIN {Pω (t)} are expressed by the following equation (22) among the detection points grouped in the same group. Represents the Z coordinate and the reflection intensity of the smallest reliability judgment value R2 (t).
R2 (t) = | Zd (t) | + | Vd (t) | + 2 · | Pd (t) |
However, Zd (t) = Zω (t) −ZB ′ (t) (23)
Vd (t) = Vω (t) −VB (t−1) (24)
Where Vω (t) = Zω (t) −Zω (t−1)
Pd (t) = Pω (t) −PB (t−1) (25)

  The reliability determination value R2 (t) is obtained for each detection point representing the object B by the above equations (22) to (25). Thereby, the reliability of each relative position is calculated about each of the detection points put together in one group. Here, the reliability judgment value R2 (t) calculated for each detection point by the equations (22) to (25) is the Z coordinate value of the detection point (that is, the distance from the host vehicle to the detection point) Zω ( It can be seen that the smaller the difference between t) and the estimated Z coordinate value obtained for the object B (that is, the estimated distance from the host vehicle to the object B) ZB ′ (t), the smaller. It can also be seen that the smaller the difference between Vω (t) representing the relative velocity of the detection point and the relative velocity VB (t−1) of the object B at the previous time t−1, the smaller the velocity becomes. It can also be seen that the smaller the difference between the reflection intensity Pω (t) at the detection point and the reflection intensity PB (t−1) of the object B at the previous time t−1, the smaller the difference. One detection point having the smallest reliability determination value R2 (t) is selected from each detection point of the object B, and the Z coordinate and reflection intensity value of the selected detection point are expressed by the equation By (19) and (21), the Z coordinate and the reflection intensity of the object B are represented on behalf of each detection point. Hereinafter, such a detection point is referred to as a representative detection point of the object B.

  The manner in which the representative detection point of the object B is selected will be described with reference to FIG. In FIG. 5, the detection points a, b, d, and e are all within the grouping range 50 and are grouped into the same group representing the object B. Here, the size of the circular range at each detection point in FIG. 5 represents the value of the reflection intensity Pω (t) at that detection point, as in FIGS. 3 and 4. For example, it is assumed that the reliability determination value R2 (t) is calculated with the smallest value for the detection point b. At this time, the detection point b is set as a representative detection point of the object B. As a result, according to the equations (19) and (21), the Z coordinate value of the object B at time t is calculated as ZB (t) = Zb (t), and the reflection intensity is calculated as PB (t) = Pb (t). .

  At the detection points grouped as representing the object B by the short distance grouping, the value of Zω (t) is less than the predetermined value (L + Z0 / 2) as described above. That is, as described above, for the detection point where the value of Zω (t) indicating the relative distance from the host vehicle is less than the predetermined value (L + Z0 / 2), the reliability determination is made based on the following three differences. The reliability of each detection point is calculated by obtaining the value R1 (t). The first difference is a difference between the position Zω (t) at the detection point and the estimated position ZB ′ (t) of the object B. The second difference is the difference between the relative speed Vω (t) at the detection point and the previous relative speed of the object B, that is, the relative speed VB (t−1) at the previous representative detection point. The third difference is a difference between the reflection intensity Pω (t) at the detection point and the reflection intensity of the previous object B, that is, the reflection intensity PB (t−1) at the previous representative detection point.

  As described above, each parameter (width) of object A and object B at time t is calculated using the calculation result at the previous time t-1 and the measurement value of each detection point grouped at the current time t. , Position coordinates, relative velocity and reflection intensity) are calculated. Note that each parameter is calculated in the same manner at other times. For example, when each parameter is calculated at time t-1, the calculation result at the previous time t-2 and the detection value for each detection point at the current time t-1 are used.

  Since there is no previous calculation result for a newly detected object, each parameter is calculated using only the detection value of the detection point at that time. Specifically, among the detected detection points, detection points that do not satisfy any of the expressions (5) and (6) or the expressions (8) and (9), that is, the object A or B detected last time. Detection points that are not located in any of the grouping ranges are processed as detection points that represent new objects, and those detection points that are located within a certain range are combined into one new object. Is grouped to represent By using the equations (11) to (13) and (15) for each detection point thus grouped, the lateral width, position coordinates, and reflection intensity of the new object are calculated. In addition, since the relative speed of the new object cannot be calculated by the equation (14), for example, an average value of the relative speeds of the grouped detection points may be used.

  By calculating each parameter as described above, the object A and the object B are detected in this apparatus. Among the detected objects, an object moving with a predetermined width (for example, 1.0 m) or more is extracted as a vehicle candidate. Further, even when the horizontal width is equal to or smaller than a predetermined value, if there are a plurality of objects moving in the same direction adjacent to each other within a predetermined distance, they are grouped and extracted as one vehicle candidate. The previous object information is updated with the vehicle candidate information extracted in this manner.

  Further, the estimated course of the host vehicle is calculated from the yaw rate and slip angle calculated by the vehicle behavior detection unit 2. Then, based on the estimated course and the object information updated as described above, an object such as another vehicle existing on the estimated course of the host vehicle is specified, and the relative distance and the relative speed are used as the object detection processing result. Output to the vehicle control unit 4. As described above, the object detection process is executed in the calculation unit 3.

  FIG. 6 shows a flowchart of a processing procedure when the calculation unit 3 performs the object detection process. This flowchart is started when the host vehicle is traveling. In step S1, the laser radar 1 scans laser light within the scan range with respect to the front of the host vehicle, and obtains reflected laser light from the objects A and B existing in front. Thereby, the distance and direction from the own vehicle to the detection points a to f are detected, and the position information (Xω (t), Zω (t)) and reflection intensity of each detection point from the laser radar 1 to the calculation unit 3 are detected. Information Pω (t) is output.

  In step S2, the vehicle behavior detector 2 detects the wheel speed and steering angle of the left and right rear wheels of the host vehicle, and calculates the yaw rate and slip angle as host vehicle travel information. As a result, the yaw rate Yaw and slip angle Slip of the host vehicle are output from the vehicle behavior detection unit 2 to the calculation unit 3.

  In step S3, the calculation unit 3 calculates the estimated positions of the objects A and B. Here, the previous coordinate values (XA (t−1), ZA (t−1)) and relative speed VA (t−1) of the object A, and the yaw rate output from the vehicle behavior detection unit 2 in step S2 Based on Yaw and the slip angle Slip, the estimated coordinates (XA ′ (t), ZA ′ (t)) of the object A are calculated by using the expressions (1) and (2) described above. Further, the previous coordinate value (XB (t−1), ZB (t−1)) and relative speed VB (t−1) of the object B, and the yaw rate output from the vehicle behavior detecting unit 2 in step S2 are also used. Based on Yaw and the slip angle Slip, the estimated coordinates (XB ′ (t), ZB ′ (t)) of the object B are calculated by using the equations (3) and (4) described above.

  In step S4, one of the objects A and B whose estimated position is calculated in step S3 is selected. In the next step S5, the distance from the host vehicle to the object A or B selected in step S4, that is, the value of ZA ′ (t) (in the case of the object A) or ZB ′ (t) (in the case of the object B) is predetermined. It is determined whether or not it is a value (for example, 60 m) or more. If it is equal to or greater than the predetermined value, the process proceeds to step S6, and after performing the processes of steps S6 to S9, the process proceeds to step S14. This corresponds to the case where the object A is selected in step S4. On the other hand, when the distance from the own vehicle is less than the predetermined value, the process proceeds to step S10, and after performing the processes of steps S10 to S13, the process proceeds to step S14. This corresponds to the case where the object B is selected in step S4.

  The case where it progresses to step S6 previously is demonstrated. In this case, the object A is selected. In step S6, long-distance grouping is performed, and detection points that satisfy the conditions of equations (5) to (7) are grouped as being included in the same group representing the object A that is relatively far away. In the next step S7, for each detection point grouped in step S6, each reliability judgment value R1 (t) is calculated using equation (16).

  In step S8, the detection point having the largest reliability judgment value R1 (t) calculated in step S7 is selected as the representative detection point. Here, the detection point c is selected as the representative detection point of the object A as described above. In the next step S9, the lateral width WA (t), the position coordinates (XA (t), ZA (t)), the relative velocity VA ( t) and the reflection intensity PA (t) are calculated. After executing step S9, the process proceeds to step S14.

  Next, the case where it progresses to step S10 is demonstrated. In this case, the object B is selected. In step S10, short distance grouping is performed, and detection points that satisfy the conditions of equations (8) to (10) are grouped as being included in the same group representing the object B at a relatively short distance. In the next step S11, the reliability determination value R2 (t) is calculated for each detection point grouped in step S10 using equations (22) to (25).

  In step S12, the detection point with the smallest reliability judgment value R2 (t) calculated in step S11 is selected as the representative detection point. Here, the detection point b is selected as the representative detection point of the object B as described above. In the next step S13, the horizontal width WB (t), the position coordinates (XB (t), ZB (t)), the relative velocity VB ( t) and the reflection intensity PB (t) are calculated. After executing step S13, the process proceeds to step S14.

  In step S14, it is determined whether or not all objects have been selected in step S4. When all the objects, that is, both the objects A and B have been selected, the process proceeds to step S15. If there is an object that has not yet been selected, the process returns to step S4. In step S15, the preceding vehicle information is output to the vehicle control unit 4 as described above based on the detection results of the objects A and B. After executing step S15, the process returns to step S1 to repeat the above processing.

  In step S6 and step S10 described above, detection points that are not located within the grouping range of any object are processed as described above as detection points representing a new object. Thereby, each parameter is calculated also about a new object like other objects.

According to the embodiment described above, the following operational effects can be obtained.
(1) For the objects A and B ahead of the host vehicle, position coordinates (Xω (t), Zω (t)) indicating relative positions with respect to the host vehicle are detected at each detection point (step S1), and a predetermined condition is satisfied. By grouping the detection points and creating a group of detection points for each object, grouping is performed (steps S6 and S10). The reliability of each relative position is calculated by calculating the reliability judgment value R1 (t) or R2 (t) for each of the detection points collected by this grouping (steps S7 and S11). Based on the calculated reliability judgment value R1 (t) or R2 (t), representative detection points are selected from the detection points grouped in the same group (steps S8 and S12), and the representative detection points are selected. Z coordinate value ZA (t) indicating the relative position of the object A in the Z axis direction and the relative position of the object B in the Z axis direction based on the Z coordinate value Zω (t) indicating the relative position in the Z axis direction The Z coordinate value ZB (t) shown is determined (steps S9 and S13). Since it did in this way, even if it is a case where the error is contained in the value of the relative position of one of the grouped detection points, the influence of the error can be removed in the calculation result of the relative position of each object. .

(2) The conditions for grouping the detection points are changed according to the relative positions of the detection points with respect to the host vehicle. Specifically, for a detection point where the value of Zω (t) indicating the relative distance from the host vehicle is equal to or greater than a predetermined value (L−Z0 / 2), by performing long-distance grouping, Further, the difference | Pω (t) −PA () between the reflection intensity Pω (t) at the detection point and the reflection intensity of the previous object A, that is, the reflection intensity PA (t−1) at the previous representative detection point. t-1) The detection points having a value of | equal to or less than a predetermined threshold value Pt1 are grouped together to represent the object A in the same group. Further, detection points where the value of Zω (t) indicating the relative distance from the host vehicle is less than a predetermined value (L + Z0 / 2) are located within the grouping range 50 by performing short-distance grouping, and further reflected. The detection points whose intensity Pω (t) is equal to or greater than a predetermined threshold value Pt2 are grouped together in the same group as representing the object B. Since it did in this way, according to the characteristic of the reflected light which changes with the relative distance from the own vehicle, the detection points showing the same object can be grouped correctly.

(3) The calculation method for calculating the reliability of each detection point is changed according to the relative position of each detection point with respect to the host vehicle. Specifically, for a detection point where the value of Zω (t) indicating the relative distance from the host vehicle grouped for the object A is equal to or greater than a predetermined value (L−Z0 / 2), the reflection intensity Pω at the detection point. By calculating the reliability judgment value R1 (t) based on (t), the reliability of each detection point is calculated. For a detection point where the value of Zω (t) indicating the relative distance from the host vehicle grouped for the object B is less than a predetermined value (L + Z0 / 2), the reflection intensity Pω (t) at the detection point, The reliability of each detection point is calculated by obtaining the reliability judgment value R2 (t) based on the difference from the previous reflection intensity of the object B, that is, the reflection intensity PB (t-1) at the previous representative detection point. To do. Since it did in this way, according to the characteristic of the reflected light which changes with the relative distance from the own vehicle, the detection point which has exact position information can be selected as a representative detection point.

  In the above embodiment, the distances of the boundary positions of the long distance grouping and the short distance grouping may be equalized. That is, far-distance grouping may be performed for detection points where the relative distance Zω (t) from the host vehicle is equal to or greater than the predetermined value L, and short-distance grouping may be performed for detection points where the relative distance is less than the predetermined value L.

  In the above-described embodiment, the example in which the relative position and the reflection intensity of each detection point are detected using the scanning laser radar has been described. For example, a radio wave radar may be used. Any object can be used as long as it can irradiate an object with an electromagnetic wave such as a laser beam or a radio wave, and detect the relative position to the vehicle and the reflection intensity of the electromagnetic wave at each detection point. .

  Furthermore, in the above-described embodiment, an example in which the present invention is applied to an apparatus that detects an object existing in front of the host vehicle has been described. However, the present invention can also be applied to an apparatus that detects an object that exists in a direction other than the front, such as rearward or laterally, around the host vehicle.

  In the above embodiment, the example in which the relative position in the Z-axis direction of the object is determined based on the relative position (relative distance) in the Z-axis direction of one representative detection point has been described. However, not only the Z-axis direction but also the relative position in the X-axis direction may be determined based on the relative position of the representative detection point. Further, two or more representative detection points may be selected, and the relative position of the object may be determined based on the relative positions of the two or more representative detection points.

  In the above embodiment, the position detection means, the electromagnetic wave irradiation means, and the reflection intensity detection means are realized by the laser radar 1, and the grouping means, the reliability calculation means, the representative point selection means, and the position determination means are realized by the calculation unit 3. Yes. In the calculation unit 3, the grouping means is processed by the processes of steps S6 and S10 in FIG. 6, the reliability calculation means is processed by the processes of steps S7 and S11, the representative point selecting means is processed by the processes of steps S8 and S12, and the processes of steps S9 and S13. Each position determination means is realized by processing. However, the configuration in the above embodiment including these is merely an example, and each component is not limited to the above embodiment unless the characteristics of the present invention are impaired. Other embodiments conceivable within the scope of the technical idea of the present invention are also included in the scope of the present invention.

It is a figure which shows the structure of the vehicle surrounding object detection apparatus by one Embodiment of this invention. It is the figure which showed the example of mounting to the own vehicle of a laser radar. It is a figure which shows a mode that the objects A and B ahead of the own vehicle which exist in the scanning range by a laser radar are detected. It is a figure for demonstrating the long distance grouping performed with respect to the object A, and the selection method of a representative detection point. It is a figure for demonstrating the short distance grouping performed with respect to the object B, and the selection method of a representative detection point. It is a flowchart which shows the process sequence of the object detection process performed in the vehicle surrounding object detection apparatus by one Embodiment of this invention.

Explanation of symbols

1: Laser radar 2: Vehicle behavior detection unit 3: Calculation unit 4: Vehicle control unit 11: Own vehicle 40, 50: Grouping range

Claims (7)

Position detecting means for detecting a relative position with respect to the host vehicle at one or a plurality of detection points for an object around the host vehicle;
Grouping detection points that satisfy a predetermined grouping condition among detection points whose relative positions are detected by the position detection unit, and grouping detection points for each object; and
Reliability calculation means for calculating the reliability of each relative position for each of the detection points grouped into the group by the grouping means;
Based on the reliability value calculated by the reliability calculation means, representative point selection means for selecting representative detection points from detection points grouped in the same group;
A vehicle surrounding object detection device comprising: position determination means for determining a relative position of the object based on a relative position of the representative detection point. In the vehicle surrounding object detection device according to claim 1,
The vehicle surrounding object detection device, wherein the grouping condition is set according to a relative position of each detection point. In the vehicle surrounding object detection device according to claim 2,
Electromagnetic wave irradiation means for irradiating the object with electromagnetic waves;
A reflection intensity detecting means for detecting the reflection intensity of the electromagnetic wave irradiated by the electromagnetic wave irradiation means for each detection point;
The position detecting means detects a relative distance from the own vehicle to each detection point as a relative position of each detection point,
The grouping means is located within a predetermined first range for a detection point whose relative distance is greater than or equal to a predetermined value, and a difference between the reflection intensity at the detection point and the reflection intensity at the previous representative detection point is predetermined. The detection points that are less than or equal to the first threshold value are grouped into the same group as satisfying the predetermined condition,
Detection points whose relative distance is less than a predetermined value are located within a predetermined second range, and detection points whose reflection intensity at the detection point is equal to or higher than a predetermined second threshold are A vehicle surrounding object detection device characterized by grouping into the same group as satisfying a predetermined condition. In the vehicle surrounding object detection device according to claim 1 or 2,
The vehicle surrounding object detection device, wherein the reliability calculation means calculates the reliability of the relative position of each detection point using a plurality of reliability calculation methods according to the relative position of each detection point. In the vehicle surrounding object detection device according to claim 4,
Electromagnetic wave irradiation means for irradiating the object with electromagnetic waves;
A reflection intensity detecting means for detecting the reflection intensity of the electromagnetic wave irradiated by the electromagnetic wave irradiation means for each detection point;
The position detecting means detects a relative distance from the own vehicle to each detection point as a relative position of each detection point,
The reliability calculation means calculates the reliability based on the reflection intensity at the detection point for the detection point where the relative distance is a predetermined value or more,
For detection points whose relative distance from the host vehicle is less than a predetermined value, the reliability is calculated based on the difference between the reflection intensity at the detection point and the reflection intensity at the representative detection point selected previously. A vehicle surrounding object detection device. In the vehicle surrounding object detection device according to claim 3,
The vehicle surrounding object detection device, wherein the reliability calculation means calculates the reliability of the relative position of each detection point using a plurality of reliability calculation methods according to the relative position of each detection point. In the vehicle surrounding object detection device according to claim 6,
The reliability calculation means calculates the reliability based on the reflection intensity at the detection point for the detection point where the relative distance is a predetermined value or more,
For detection points whose relative distance from the host vehicle is less than a predetermined value, the reliability is calculated based on the difference between the reflection intensity at the detection point and the reflection intensity at the representative detection point selected previously. A vehicle surrounding object detection device.

Images