JP3201128B2 - Inter-vehicle distance measurement device and vehicle equipped with the same - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inter-vehicle distance measuring device for measuring an inter-vehicle distance from a preceding vehicle and a vehicle equipped with the same.

[0002]

2. Description of the Related Art Conventionally, an inter-vehicle distance detecting device for detecting an inter-vehicle distance with a preceding vehicle is disclosed in Japanese Patent Application Laid-Open No. 61-23985. This inter-vehicle distance detecting device emits an electromagnetic wave, receives a reflected wave reflected by a front reflecting object, detects a reflecting object, and, from the reflecting object, a roadside reflector provided on both the left and right of the roadside, and a preceding reflector. Identify the vehicle. Based on the angles (positions) of the left and right roadside reflectors on the roadside, it is determined which lane the host vehicle and the preceding vehicle are traveling. It outputs the inter-vehicle distance to a preceding vehicle traveling in the same lane as the own vehicle.

However, in such a conventional inter-vehicle distance detecting apparatus, both the right and left roadside reflectors on the roadside must be detected. If the roadside reflector is provided only on one of the right and left roadsides, Alternatively, if one roadside reflector cannot be detected behind another vehicle, the lane of the preceding vehicle cannot be accurately identified.

[0004] When the vehicle is traveling on a curve, the determination of the lane in which the preceding vehicle is traveling must be corrected based on the steering angle of the own vehicle.

[0005]

SUMMARY OF THE INVENTION An object of the present invention is to provide an inter-vehicle distance measuring device for measuring an inter-vehicle distance between a host vehicle and a preceding vehicle traveling on the same lane.

An inter-vehicle distance measuring device according to the present invention is mounted on a vehicle. A vehicle equipped with this inter-vehicle distance measuring device is referred to as a host vehicle.

An inter-vehicle distance measuring apparatus according to the present invention emits a laser beam at a predetermined time interval at a predetermined angle interval within a predetermined measurement angle range, and emits a projection timing signal indicating a projection time point of the laser beam; A light projecting means for outputting a sweep angle signal representing a light angle, receiving a reflected light of the laser light projected from the light projecting means from an object to be detected, and receiving a light receiving timing signal representing a point in time when the reflected light is received A light receiving means for outputting, a distance calculating means for calculating a distance to an object to be detected based on a time interval between a light emitting timing signal from the light emitting means and a light receiving timing signal from the light receiving means;
And calculating the relative position of the detected object based on the distance to the detected object calculated by the distance calculating means and the sweep angle signal obtained from the light projecting means, and calculating the relative position of the same detected object. Calculates the motion vector of the detected object based on the two relative positions at different times, and identifies whether the detected object is a roadside reflector or a vehicle based on the motion vector of the detected object Then, the roadside shape is recognized based on the relative position of the detected object identified as one of the right and left roadside reflectors among the detected objects identified as the roadside reflector, and based on the recognized roadside shape. The lane in which the vehicle is traveling is determined for the detected object identified as the vehicle by
Signal processing means for outputting a vehicle having a minimum distance to a vehicle traveling in the same lane as an inter-vehicle distance is provided.

According to the present invention, the relative position of the detected object is detected by projecting the laser beam in the predetermined measurement angle range, and the movement vector of the detected object is calculated. The roadside reflector and the vehicle are identified based on the movement vector of the detected object. The roadside shape is recognized using one of the roadside reflectors identified, and the lane in which the vehicle travels is determined based on the recognized roadside shape. If it is determined that the vehicles are on the same lane, the inter-vehicle distance of the vehicle with the shortest distance is output.

Therefore, when one of the right and left roadside reflectors among the roadside reflectors provided on the roadside can be detected, the traveling lane of the preceding vehicle is determined by recognizing the roadside shape based on the position of the roadside reflector. it can. This makes it possible to measure the inter-vehicle distance of a vehicle traveling on the same lane as the own vehicle.

Further, since the lane on which the vehicle travels is determined by recognizing the roadside shape, it is not necessary to correct the curve.

More specifically, an inter-vehicle distance measuring apparatus according to the present invention emits a laser beam at a constant time interval at a constant angular interval over a predetermined measuring angle range, and indicates a time point at which the laser beam is projected. A light projecting means for outputting a light projecting timing signal and a sweep angle signal representing the light projecting angle, receiving the reflected light of the laser light emitted from the light projecting means from the object to be detected, and receiving the reflected light A light-receiving means for outputting a light-receiving timing signal indicating a point in time; calculating a distance to an object to be detected based on a time interval between the light-emitting timing signal supplied from the light-emitting means and the light-receiving timing signal supplied from the light receiving means; Calculating means, a vehicle speed detecting means for detecting a vehicle speed of the own vehicle, a distance to the detected object calculated by the distance calculating means, and a sweep given by the light projecting means. Relative position calculating means for calculating the relative position of the detected object based on the degree signal and whether or not the detected object is a preceding vehicle on the own lane based on at least the relative position calculated by the relative position calculating means. A lane determining means for determining whether or not the detected object which is determined to be a preceding vehicle on the own lane by the lane determining means outputs a distance to a vehicle having the smallest relative position as an inter-vehicle distance; Output means is provided.

According to the present invention, an object to be detected is detected by projecting a laser beam in a predetermined measurement angle range. The object to be detected is, for example, a vehicle reflector attached to a general four or more vehicle such as a passenger car, a bus, a truck, a two-wheeled vehicle, and a roadside reflector provided on a roadside. Based on the detected distance to the detected object and the sweep angle signal, the relative position is calculated, and if necessary, the movement vector is calculated.

Further, the roadside reflector is detected based on the movement vector of the detected object and the vehicle speed of the host vehicle as required. The detected roadside reflectors are divided into a group provided on the left side of the road and a group provided on the right side of the road. The roadside shape is recognized based on the relative positions of the roadside reflectors of the group to which the roadside reflector belongs in a large number. The roadside shape is recognized as a straight line or a circle.

[0014] In the first embodiment of the present invention, the lane judging means determines, for each detected object, a distance from the vehicle to the roadside represented by the roadside shape and a distance from the detected object to the roadside shape. It is determined whether or not the detected object is the preceding vehicle on the own lane based on the distance to the roadside represented.

In a second embodiment of the present invention, the lane determining means determines, for each detected object, that the detected object is a preceding vehicle on the own lane based on the relative position and the movement vector of the detected object. It is to determine whether there is.

In a third embodiment of the present invention, the lane determining means determines, for each detected object, whether or not the detected object is a preceding vehicle on the own lane based on a relative position of the detected object. Is determined.

In a fourth embodiment of the present invention, for each detected object, the distance from the vehicle to the roadside represented by the roadside shape and the distance from the detected object to the roadside represented by the roadside shape are determined. Based on the distance, a first own vehicle lane presence degree indicating the degree of the detected object existing on the own lane and a first own degree indicating the existing degree on the other lane.
The first evaluation means for calculating the other lane existence degree is further provided. The lane determining means is realized as determining whether the detected object is a preceding vehicle on the own lane based on at least the first own lane abundance calculated by the first evaluating means. .

In a fifth embodiment according to the present invention,
For each detected object, based on the relative position and the movement vector of the detected object, a second self-lane abundance indicating the degree of the above-described object existing on the own lane and a degree of being present on the other lane are shown. Second evaluation means for calculating a second other lane existence degree is further provided. The lane determining means is realized as determining whether the detected object is a preceding vehicle on the own lane based on at least the second own lane abundance calculated by the second evaluating means. .

In a sixth embodiment of the present invention, for each of the detected objects, a third self-lane presence indicating the degree of existence of the detected object on the own lane based on the relative position of the detected object. Third evaluation means for calculating a third degree of presence of another lane indicating the degree and the degree of existence on another lane is further provided. The lane determining means is realized as a means for determining whether or not the detected object is a preceding vehicle on the own lane based on at least the third own lane abundance calculated by the third evaluating means. .

In a seventh embodiment of the present invention, for each detected object, the distance from the vehicle to the roadside represented by the roadside shape and the distance from the detected object to the roadside represented by the roadside shape are determined. A first self-lane abundance indicating the degree of existence of the detected object on the own lane and a first other lane abundance indicating the degree of existence on another lane based on the distance Evaluation means, for each detected object, based on the relative position and the movement vector of the detected object, a second own lane existence degree indicating the degree of existence of the detected object on the own lane and an existence degree on another lane A second evaluation means for calculating a second other lane existence degree representing the degree of the presence of the detected object, and for each detected object, the degree of the existence of the detected object on the own lane based on the relative position of the detected object. Third to represent Third evaluation means for calculating a third other lanes abundance indicating a degree that exists in the own lane abundance and other lanes on are further provided. The lane determining means calculates the sum of the first, second, and third own lane abundances and the first, second, and third other lane abundances for each of the detected objects. This is realized as determining the detected object whose sum of degrees is larger than the sum of the degrees of other lanes as the preceding vehicle on the own lane.

According to the first embodiment and the fourth embodiment, of the roadside reflectors provided on the roadside of the road,
If the roadside reflector provided on one of the right and left roadsides can be detected, the roadside shape can be recognized based on the relative position of the roadside reflector. By recognizing the roadside shape, the distance from the detected object (preceding vehicle) to the roadside represented by the roadside shape can be calculated. Further, the traveling lane of the detected object (preceding vehicle) can be evaluated based on the distance to the road side.

Further, since the traveling lane of the detected object (preceding vehicle) is determined based on the distance from the detected object (preceding vehicle) to the road side, the road curves as in the prior art. There is no need to make corrections based on the steering angle when the vehicle is running.

In place of or in addition to the evaluation of the traveling lane based on the distance from the detected object to the roadside, the evaluation of the traveling lane is performed based on the relative position and the movement vector of the detected object (second and second). 5, Seventh Embodiment), and a more reliable determination of the traveling lane can be performed. Further, when the evaluation based on only the relative position of the detected object is added (third, sixth, and seventh embodiments), even if the roadside reflector cannot be detected, the traveling lane of the preceding vehicle is determined. can do.

In an eighth embodiment according to the present invention, the relative position calculating means checks whether or not the detected objects whose relative positions have been calculated depend on light reflected from the same detected object. Gender identification is performed, and the detected objects determined to be the same by the identity identification are collectively determined as one detected object, and the relative position of the collected single detected object is calculated again.

According to the eighth embodiment, even if a plurality of reflected lights are detected from the same detected object, they are combined into one detected object by the identity discrimination.
The same detected object is no longer treated as a plurality of detected objects.

According to a ninth embodiment of the present invention, for each detected object, it is determined whether the relative positions and the movement vectors of one detected object and another detected object satisfy a pair condition indicating a relationship of being a vehicle. It is checked whether there are two detected objects satisfying the pair condition, and a vehicle identification means for combining them into one detected object is further provided.

According to the ninth embodiment, since two reflectors attached to a vehicle are combined into one detected object, one vehicle cannot be treated as two preceding vehicles.

In the first, fourth and seventh embodiments, preferably, the road-side shape recognizing means detects, for each detected object, a movement vector calculated by the movement vector calculating means and a vehicle speed detected by the vehicle speed detecting means. Based on the vehicle speed, the roadside reflector provided on the roadside of the road is detected, and the detected roadside reflector is divided into a group to which the one provided on the right side of the road belongs and a group to which the one provided on the left side of the road belongs. The roadside shape is recognized based on the relative positions of the roadside reflectors of the group to which the number belongs.

[0029] In the first, fourth and seventh embodiments, more preferably, the roadside shape recognizing means includes a group to which a roadside reflector provided on the right side of the road belongs by one of the following two methods. And the group to which the one provided on the left side of the road belongs.

The first method is to classify the roadside reflector into a group belonging to the group provided on the right side of the road and a group belonging to the group provided on the left side of the road according to the relative position of the roadside reflector and the moving direction based on the moving vector. What separates.

In the second method, a group to which the roadside reflector is provided on the right side of the road and a group to which the roadside reflector is provided on the left side of the road belong to the relative movement based on the relative position and the movement vector of the roadside reflector. It is divided into

Therefore, the roadside reflectors can be accurately grouped by the first or second method.

Preferably, in the seventh embodiment, the road-side shape recognizing means further evaluates the reliability of whether or not the recognized road-side shape is appropriate. If there is no reliability of the roadside shape recognized by the roadside shape recognition means, the own lane abundance and the other lane abundance are not calculated for all detected objects.

By evaluating the reliability of the roadside shape recognized by the roadside shape recognition means, it is possible to determine whether the roadside shape is correct. If the recognized roadside shape is unreliable, the first evaluation means does not perform the evaluation based on the distance from the detected object to the roadside. , The traveling lane of the detected object (preceding vehicle) can be accurately determined using the evaluation results of the other second or third evaluation means.

Preferably, in the seventh embodiment, when the movement vector of the detected object is equal to or less than a predetermined threshold value, the second evaluating means determines whether the detected object has the second own lane abundance and the second vehicle lane. No other lane existence degree is calculated.

When the inter-vehicle distance between the host vehicle and the preceding vehicle is substantially constant, the magnitude of the movement vector of the preceding vehicle becomes small. When the magnitude of the movement vector is small, the second evaluation means may not be able to accurately evaluate the detected object. For this reason, the second evaluation means does not evaluate the detected object (preceding vehicle) having a small movement vector. The lane judging means does not consider the inaccurate evaluation result by the second evaluating means and uses the evaluation results of the other first or third evaluating means to travel the detected object (preceding vehicle). The lane can be accurately determined.

In the sixth and seventh embodiments, preferably, the third evaluating means increases the third vehicle lane abundance of a detected object located in a short distance in front of the vehicle, and increases the third vehicle lane abundance. The third self-lane abundance decreases as the distance increases and as the distance increases.

Therefore, the first evaluation means cannot evaluate the traveling lane of the detected object (preceding vehicle) based on the distance from the detected object (preceding vehicle) to the road side because the recognized roadside shape is not reliable. Sometimes. In addition, the second evaluation means may not be able to accurately evaluate the traveling lane for the detected object because the magnitude of the movement vector of the detected object (preceding vehicle) is small. Even in such a case, if the preceding vehicle can be detected as an object to be detected, the preceding vehicle existing at a short distance on the own lane can always be detected by using the third evaluation means. The distance between vehicles can be measured.

[0039]

[Explanation of the embodiment]

Table of contents 1. Overview 2. Hardware configuration 3. Detailed description of inter-vehicle distance measurement (signal processing device) 3.1 Collection of paired data and vehicle speed data 3.2 Calculation of relative position of detected object 3.2.1 Coordinate transformation of relative position of detected object 3.2.2 Identification of the identity of the detected object 3.3 Calculation of the movement vector of the detected object 3.4 Identification of the vehicle and roadside reflector 3.4.1 Identification of the vehicle 3.4.2 Identification of the roadside reflector 3.5 Clustering of the roadside reflector 3.5.1 Relative position and movement of the roadside reflector Clustering of roadside reflectors based on vectors (Part 1) 3.5.2 Clustering of roadside reflectors based on relative position and movement vector of roadside reflectors (Part 2) 3.6 Recognition of roadside shape and its reliability evaluation 3.6.1 Recognition of roadside shape 3.6 .2 Reliability evaluation of roadside shape 3.7 Evaluation of driving lane of preceding vehicle based on distance from preceding vehicle to roadside 3.8 Evaluation of preceding vehicle To location and movement evaluation of the traffic lane of the preceding vehicle by vector 3.9 prior evaluation of the traffic lane of the preceding vehicle by only the relative position of the vehicle 3.10 Output of overall judgment 3.11 headway distance traffic lane of the preceding vehicle

1 Outline

FIG. 1 shows a state where three vehicles are traveling on a road. Only the road on one side of the center line C is shown, and this road has two lanes on each side. These lanes are denoted by L1 and L2.

The vehicles 1 and 2 are traveling on the inner lane L1, and the vehicle 3 is traveling on the outer lane L2. A number of reflectors R are provided at appropriate intervals along the roadside. The reflector R provided on the road side is hereinafter simply referred to as “roadside reflector”. These roadside reflectors are used for detecting the roadside as will be described in detail later.

In general, vehicles having four or more wheels (hereinafter, simply referred to as "vehicles") such as passenger cars, buses, trucks, etc., are provided at both ends (near the location where the taillights are attached) at a rear end for a total of two vehicles. Reflectors are mounted. One reflector is mounted on the motorcycle. The reflector attached to the vehicle and the motorcycle is hereinafter simply referred to as “vehicle reflector”. These vehicle reflectors are used to detect the presence of vehicles and two-wheeled vehicles, as described in more detail below. There is also a vehicle including a vehicle and a motorcycle. In particular, when referring to motorcycles, the term motorcycle rather than vehicle is used.

The roadside reflector and the vehicle reflector are called retroreflectors, and have a property that the reflection direction is almost the same as the incident direction.

The inter-vehicle distance measuring device 10 is mounted on a vehicle.
The vehicle on which the inter-vehicle distance measuring device 10 is mounted is referred to as “own vehicle”.
That. The traveling lane of the own vehicle is called "own lane", and lanes other than the own lane are called "other lanes". Vehicles existing in front of the own vehicle are referred to as "preceding vehicles" regardless of their own lane or another lane and include motorcycles and the like.

In this embodiment, it is assumed that the inter-vehicle distance measuring device 10 is mounted on the vehicle 1. A case in which the inter-vehicle distance is measured between the vehicle 1 which is the own vehicle and the vehicle 2 which is the preceding vehicle existing on the own lane will be described below.

The inter-vehicle distance measuring device 10 projects a pulsed laser beam toward the front of the host vehicle. This laser light is swept over a predetermined measurement angle range. The laser light is projected at a constant angular interval within the above measurement angle range.

For example, the measurement angle range is 200 [mrad], and has a spread of 100 [mrad] on the left and right sides in the traveling direction of the host vehicle. This measurement angle range is 80 at equal angular intervals.
The laser beam is divided and a laser beam is projected at each angular interval.
Laser light is projected at an angular interval of 2.5 [mrad].

The laser beam projected from the inter-vehicle distance measuring device 10 is reflected by the roadside reflector R or a vehicle reflector attached to the preceding vehicles (vehicles 2 and 3) and returns to the inter-vehicle distance measuring device 10. The inter-vehicle distance measuring device 10 communicates with the roadside reflector R and the preceding vehicle (vehicle 2) based on the reflected light.
And the vehicle reflector of the vehicle 3) are detected as detected objects.

As will be described in detail later, the relative position of the detected object (position based on the own vehicle (vehicle 1)) is calculated. A movement vector of the detected object is calculated based on the relative position of the detected object obtained at least two times at different times. Vehicles (vehicles 2 and 3) are identified based on the relative position of the detected object and its movement vector. The roadside reflector R is identified based on the movement vector of the detected object.

The roadside shape is recognized based on the relative position of the detected object determined to be the roadside reflector. For all detected objects (including roadside reflectors), the distance from each detected object to the roadside based on the recognized roadside shape (hereinafter, "distance to roadside"
(A distance in a direction perpendicular to the traveling direction of the vehicle 1) is calculated. For all of the detected detected objects, the preceding vehicle (vehicle 2 and vehicle 3) is determined based on the distance from the detected object to the roadside, the relative position and movement vector of the detected object, and the relative position of the detected object. The evaluation for determining the traveling lane of (2) is performed. Based on these evaluation results, it is comprehensively determined that the preceding vehicle running on the own lane of the vehicle 1 (own vehicle) is the vehicle 2.

In this manner, the distance between the preceding vehicle 2 traveling on the determined own lane and the vehicle 1 (own vehicle) is the inter-vehicle distance. When there are a plurality of preceding vehicles traveling on the own lane, the smallest one of the preceding vehicles is output as the inter-vehicle distance.

2 Hardware Configuration

FIG. 2 is a block diagram showing an electrical configuration of the inter-vehicle distance measuring device 10. As shown in FIG. The inter-vehicle distance measuring device 10 includes a light emitting device 11, a laser light sweeping device 12, a sweep angle detecting device 13, a light receiving device 14, a distance calculating device 15, a vehicle speed sensor 16, a warning device 17, an actuator 18, and a signal processing device 19.

The light projector 11 repeatedly emits a pulsed laser beam at a predetermined time interval, and outputs a light emission timing signal indicating a point in time when the laser beam is emitted. For the light projector 11, for example, a semiconductor laser light emitting device is appropriate in terms of miniaturization of the device and high-speed driving. The light projection timing signal is provided from the light projector 11 to the distance calculator 15.

For example, assuming that the measurable distance is 150 [m], the time interval for projecting the laser light is the time required for the laser light to reciprocate, that is, (150 [m] ×
2) / (3 × 10 ⁸ [m / s]) = 1 [μs] or more.

The laser light sweeping device 12 sweeps the laser light emitted by the light projector 11 within a measurement angle range. Laser light is
The light is projected radially forward at a constant angular interval (equal angular interval described above) at each light emitting time interval. The laser light sweeping device 12 is appropriate in that a laser light reflecting mirror that vibrates and changes the angle at a predetermined cycle does not easily spread the laser light. The sweep angle position signal indicating the angle at which the laser beam was projected is sent from the laser beam sweep unit 12 to the sweep angle detection unit.
Given to 13.

The sweep angle detector 13 is a laser beam sweeper.
The sweep angle θ at which the laser beam is projected (based on the direction perpendicular to the traveling direction of the vehicle 1) is calculated based on the sweep angle position signal given from step S12. The data representing the calculated sweep angle θ is sent from the sweep angle detector 13 to the signal processor 19.
Given to.

The laser light projected from the laser sweeping device 12 is reflected by an object to be detected (the above-described roadside reflector or vehicle reflector), and the reflected light is received by the light receiver 14.

When receiving the light reflected by the object to be detected, the light receiver 14 outputs a light reception timing signal indicating the time when the light is received. The reflected light is received between the time when the laser light is projected from the laser light sweeping device 12 and the time when the next laser light is projected. The light reception timing signal is provided from the light receiver 14 to the distance calculator 15. The light receiver 14 converts reflected light incident on a light receiving element (for example, a photodiode, a phototransistor, or the like) into an electric signal, and outputs a light receiving timing signal when the signal is equal to or larger than a predetermined threshold. Noise is removed by performing threshold processing.

The distance calculating device 15 measures the time interval from the input time of the light emitting timing signal given from the light emitting device 11 to the input time of the light receiving timing signal given from the light receiving device 12, and calculates the distance d to the object to be detected. Is calculated. Data representing the calculated distance d to the detected object is provided from the distance calculation device 15 to the signal processing device 19.

The vehicle speed sensor 16 detects the vehicle speed v (the vehicle speed of the vehicle 1 on which the inter-vehicle distance measuring device 10 is mounted). Data representing the vehicle speed v is provided from the vehicle speed sensor 16 to the signal processing device 19.

The signal processor 19 will be described later based on the sweep angle θ given from the sweep angle detector 13, the distance d to the detected object given from the distance calculator 15, and the vehicle speed v given from the vehicle speed sensor 16. The calculation of the inter-vehicle distance to the preceding vehicle and other processing are performed.

The signal processor 19 also calculates the risk based on the following distance of the own vehicle lane to the preceding vehicle, its movement vector, the vehicle speed, etc., outputs a warning signal based on this risk, and outputs a warning signal. The amount of operation to be given to an actuator (brake, throttle valve, etc.) for avoiding the above is determined and output.

For example, when the inter-vehicle distance to the preceding vehicle is short, there is a high risk of collision with the preceding vehicle. In such a case, an alarm signal is output to the alarm device 17, and an operation amount for deceleration or stop is given to the actuator 18.

The alarm device 17 sounds an alarm in response to an alarm signal given from the signal processing device 19. As a result, the driver of the host vehicle can be notified of the danger of colliding with the preceding vehicle as described above.

The actuator 18 performs operations such as braking the vehicle and cutting off the supply of fuel based on the operation amount given from the signal processing device 19.

3 Detailed description of distance measurement (signal processing device)

FIG. 3 and FIG. 4 are flow charts showing the processing procedure of one processing cycle of the following distance measurement in the signal processing device 19. The signal processing device 19 can be realized by a computer system and software for operating the computer system.

3.1 Collection of paired data and vehicle speed data (FIG. 3; step 20)

The one cycle process is synchronized with one sweep of the laser beam over the measurement angle range. From one end of the measurement angle range to the other end (for example, from the left end to the right end)
The laser light is swept in one direction, and the object to be detected is detected by one sweep of the laser light. After this, step 20 ~
A series of processes in step 30 is performed. After this series of processing, the laser beam is swept again in the other direction over the measurement angle range. The sweep of the laser beam and a series of subsequent processes are repeatedly performed at a fixed sweep time interval. The time during which one sweep of the laser light and the series of processes shown in FIGS. 3 and 4 are performed is referred to as a processing cycle time interval T [s]. For example, the processing cycle time interval T is T = 0.1 [s].

The processing of the signal processing device 19 performed on the detected object detected by the k-th laser beam sweep will be described below as the current processing cycle k.

Each time a laser beam is projected, the sweep angle detecting device 13 gives the detected sweep angle θ to the signal processing device 19. In addition, the distance calculation device 15 detects the object to be detected (if any).
The distance d is calculated and given to the signal processor 19. The signal processing device 19 stores the sweep angle θ and the distance d as pair data in a pair data storage area of the memory.

FIG. 5 shows an example of pair data stored in the pair data storage area of the memory. Assuming that the laser beam is projected 80 times in one sweep cycle, a maximum of 80 pairs of data will be stored in the pair data storage area of the memory. The pair data is stored in the pair data storage area of the memory corresponding to the detection order i. In FIG. 5, eleven pairs of data are stored.

The vehicle speed sensor 16 gives the detected vehicle speed v of the own vehicle to the signal processor 19. The signal processing device 19
This vehicle speed v is only one for one sweep of laser light,
At a predetermined timing (for example, during or before or after the sweep of the laser beam), data representing the vehicle speed v is fetched and stored in the vehicle speed data storage area. Data representing the vehicle speed stored in the vehicle speed data storage area of the memory is defined as a vehicle speed vk. For example, the vehicle speed vk is vk = 21 [m / s] (= 75 [km / s
h]).

3.2 Calculation of the relative position of the detected object (FIG. 3;
Step 21)

3.2.1 Coordinate transformation of relative position of detected object

The distance d and the sweep angle θ of the detected object detected by the sweep of the laser beam are expressed in a polar coordinate system. The relative position of the detected object expressed in this polar coordinate system is
Coordinate conversion into a relative position in a rectangular coordinate system.

The ith detected object OBk, i
Is pair data dk, i and θk, i. Based on the pair data dk, i and θk, i, the detected object O
The relative position Qk, i (xk, i, yk, i) of Bk, i is calculated. The distance dk, i to the object to be detected OBk, i and the sweep angle θk, i are expressed in a polar coordinate system. The polar coordinate system is transformed into a rectangular coordinate system, and the relative position of the detected object is represented by the rectangular coordinate system. The relative position Qk, i (xk, i, yk, i) of the detected object OBk, i is coordinate-transformed by the following equation.

[0080]

Xk, i = dk, i · cos θk, i (1) yk, i = dk, i · sin θk, i (2)

FIG. 6A shows a polar coordinate system representing the detection position of the detected object. FIG. 6B is a diagram in which the position display of the detected object OBk, i shown in FIG. 6A is coordinate-transformed from the polar coordinate system to the rectangular coordinate system according to the equations (1) and (2). In these figures, a vehicle 1 on which an inter-vehicle distance measuring device 10 is mounted is shown.
Is the Y axis, and the direction orthogonal to the Y axis in the horizontal plane is the X axis and the direction of θ = 0. The origin O is located on the vehicle 1 in any coordinate system.

FIG. 7 is a flowchart showing the coordinate conversion process. In the flow chart of FIG. 7, Nk is the detected object OB detected in the k-th laser beam sweep.
The total number of k and i, that is, the total number of pair data stored in the pair data storage area of the memory. In this relative position calculation process, coordinate conversion is performed for all pair data stored in the pair data storage area of the memory.

I is initialized to 1 (step 31), and the first
The pair data θk, 1 and dk, of the object to be detected OBk, 1
1 is read from the paired data storage area of the memory, and the coordinate conversion is started from the first detected object OBk, 1.

In the example shown in FIG. 5, the first paired data θ
k, 1 = 1668.3 [mrad] and dk, 1 = 48 [m] are read.

Coordinate transformation is performed on the pair data θk, i and dk, i of the detected object OBk, i read from the pair data storage area of the memory according to the above equations (1) and (2) (step 33). ).

For example, when i = 1, the pair data is θk,
Since 1 = 1668.3 [mrad] and dk, 1 = 48 [m], the equation
When coordinates are transformed according to (1) and (2), xk, 1 =
48 × cos 1.6683 = -4.7, yk, 1 = 48 × sin 1.6683 = 47.8
Becomes The data representing the relative position Qk, i (xk, i, yk, i) of the coordinate-transformed detected object OBk, i is stored in the relative position data temporary storage area of the memory as shown in FIG.

I is incremented (step 34),
Returning to step 33, steps 33 and 34 for a new i
Is repeated.

The above processing (steps 33 and 34) is repeated while sequentially incrementing i. If i> Nk, the coordinate conversion has been completed for all of the detected objects OBk, i. . (Step 32).

When the coordinate conversion process is performed on the pair data shown in FIG. 5, relative position data in the orthogonal coordinate system as shown in FIG. 8 is obtained.

3.2.2 Identification of identity of detected object

A plurality of data (objects to be detected OBk, i) may be obtained from one reflector. For example, this is when a roadside reflector, a preceding vehicle (vehicle reflector), or the like exists at a position at a short distance from the host vehicle. Therefore, when a plurality of data are detected within a certain range (identity identification window to be described later), the identity identification for combining the plurality of data into one data as a result of one detected object. Processing can be performed.

As described above, the measurement angle range is 200 [mrad], and this range is divided into 80 equal angles, so that the laser beam is projected at an angular interval of 2.5 [mrad]. At a position 10 [m] away from the vehicle, the distance between the optical paths of two adjacent laser beams is 10 [m] x sin (2.5 [mr
ad]) = 0.025 [m] = 2.5 [cm]. Therefore, if there is a preceding vehicle at a distance of 10 [m] from the own vehicle, two or more data can be obtained from one vehicle reflector even if the vehicle reflector attached to the preceding vehicle is about several cm. Sometimes.

In order to always represent one detected object with one piece of relative position data, identity detection processing of the detected object is performed. In this identity discrimination processing, the relative position Qk, i (xk, i, y) of the detected object OBk, i is determined according to the detection order i.
k, i) is set as the reference. The next object to be detected OBk, i + 1 is included in the identity identification window.
Exists, it is determined that these two detected objects OBk, i and OBk, i + 1 are actually the same reflector.

As shown in FIG. 9, the identity discrimination window has two diagonal points, ie, a starting point G1 (x1G, y1G).
And an end point G2 (x2G, y2G). The coordinates of the start point G1 and the end point G2 are respectively relative positions Qk, i (xk, i, yk, i) of the detected object OBk, i.
Is represented by the following equation based on

[0095]

X1G = xk, i (3) y1G = yk, i + 1 (4) x2G = xk, i + 0.5 (5) y2G = yk, i-1 (6)

It is determined whether or not the next detected object OBk, i + 1 exists in the set identity discrimination window. If the detected object OBk, i + 1 exists in the identity identification window, the detected object OBk, i and the detected object OBk, i +
It is determined that 1 is the same (the same reflector).

The detected object OBk, i and the detected object OBk, i +
1 is determined to be the same, the detected object OBk, i +
The identity identification window is set (updated) again based on the relative position Qk, i (xk, i + 1, yk, i + 1) of 1 and the next detected object OBk, i + 2 is newly set. It is determined whether or not it exists in the same identity identification window. In this way, while sequentially updating the identity identification window,
The detected object OBk, i + sequentially adjacent to the detected object OBk, i
, OBk, i + 2,... Are subjected to identity discrimination processing to determine whether they are the same detected object.

A plurality of detected objects determined to be the same are collected into one detected object. At this time, the center relative position of the detected object combined into one is calculated as follows.

The x-coordinate of the center relative position is represented by an intermediate value of the x-coordinates of the relative positions of the detected objects located at both ends among a plurality of detected objects determined to be the same. That is, the relative positions of the detected objects located at both ends are
Assuming that (xk, L, yk, L) and (xk, R, yk, R), the x coordinate xk, m of the center relative position is represented by the following equation.

[0100]

Xk, m = (xk, L + xk, R) / 2 (7)

The y coordinate of the center relative position is represented by the average value of the y coordinates of all the relative positions of the detected object placement determined to be the same. That is, assuming that the relative position of the detected object is (xk, i, yk, i) (i = L to R) and the total number of the detected objects OBk, i determined to be the same is ND, the center relative position Is represented by the following equation.

[0102]

Yk, m = (Σyk, i) / ND (8) where Σ represents the addition for all Nk yk, i of i = L to R.

FIGS. 10 and 11 are flow charts showing the process of identifying the identity of the detected object.

I is initialized to 1 (FIG. 10; step 3)
6), relative position Qk, 1 of the first detected object OBk, 1
Data representing (xk, 1, yk, 1) is read from the relative position data temporary storage area of the memory. The identity discrimination process is started from the first detected object OBk, 1. In the example shown in FIG. 8, the relative position Qk, 1 (-4.7, 47.8) is read from the relative position data temporary storage area.

It is determined whether or not the identity identification window has been set (FIG. 10, step 38). If the identity identification window has not been set (N in step 38)
O), relative position Qk, i of read object OBk, i
An identity discrimination window is set based on (xk, i, yk, i) (FIG. 10; step 43).

Thereafter, i is incremented (FIG. 10;
After step 46) and step 37, the process returns to step 38 again. If the identity window is not set,
The object to be detected OBk, i read first, that is, only when i = 1. If i ≧ 2, the identity identification window has already been set.

In step 38, if the identity identification window has already been set (YES in step 38), it is determined whether the read object to be detected OBk, i exists in the identity identification window (step 38). Figure 10; step 39).

If the object to be detected OBk, i exists in the identity discrimination window (YES in step 39), the object to be detected OBk, i-1 which is the reference of the identity discrimination window with the object to be detected OBk, i Are regarded as the same detected objects, and these are stored (registered) in the temporary registration area of the memory (FIG. 10; step 44).

In the example shown in FIG. 8, when the identity identification window is set based on the relative position Qk, 1 (-4.7, 47.8) of the detected object OBk, 1, the next detected object OBk, 2 Since the relative position Qk, 2 (−4.6, 47.8) of the object exists in the identity discrimination window, the detected objects OBk, 1 and OBk, 2 are registered in the temporary registration area of the memory.

When the detected object OBk, i is registered, the relative position Qk, i (xk, i, yk, i) of the detected object OBk, i is registered.
) Is updated as a reference (a new identity identification window is set) (FIG. 10;
Step 45).

In step 39, the detected object OBk, i
Is determined not to exist in the identity identification window (NO in step 39), it is determined whether the detected object OBk, i is registered in the temporary registration area of the memory (FIG. 10; step 40). . If there are a plurality of detected objects already registered in the temporary registration area of the memory (YE
S), they are put together, and the center relative positions of the put-up plurality of detected objects are calculated according to the above equations (7) and (8) (FIG. 10; step 41). Thereafter, the temporary registration area of the memory is cleared.

For example, if the detected objects OBk, 1 and OBk, 2 are registered in the temporary registration area of the memory, the detected objects OBk, 1 and OBk, 2 are registered based on their relative positions.
2 and the center relative position is xk, 1 = ｛(-4.
7) + (-4.6)｝ / 2 = -4.7, yk, 1 = {47.8 + 47.8} / 2
= 47.8). As will be described later, after this identity identification processing, a serial number is newly assigned from 1 as a number i indicating the detection order, so that the detected objects OBk, 1 and OBk,
The center relative position of 2 is represented by (xk, 1, yk, 2).
The center relative position calculated in this way is stored in the current registration data storage area (k) provided in the memory in the order of processing. An example of the current registration data storage area (k) created based on the data shown in FIG. 8 is shown in FIG.

If the detected object is not registered in the temporary registration area of the memory (NO in step 40), there is nothing to collect, so that the processing in step 41 is not performed, and the relative position of the detected object OBk, i Qk, i (xk, i, yk, i) is stored in the current registration data storage area (k) (FIG. 1).
0; Step 42).

After the processing in step 41 or 42, the identity discrimination window is set based on the relative position Qk, i (xk, i, yk, i) of the detected object OBk, i previously read from the temporary registration area. It is set (FIG. 10; step 43), i is further incremented (FIG. 10; step 46), and the process returns to step 38 to repeat the processing of steps 38 to 46 for a new i.

The above processing (steps 38 to 46) is repeated while sequentially incrementing i, and if i> Nk (FIG. 10; step 37), the detected object registered in the temporary registration area of the memory is It is determined whether there is (Figure 11;
Step 47).

If there are detected objects registered in the temporary registration area of the memory (YES in step 47), they are collected, and the central relative position of the collected detected objects is calculated by the above equation.
It is calculated according to (7) and equation (8) (FIG. 11; step 50). The calculated center relative position is stored in the current registration data storage area (k) of the memory in the order of processing.

The relative position data or center relative position data stored in the current registration data storage area (k) includes
They are assigned a detection order i again in the order of storage.
In the following description, the relative position data and the center relative position data obtained by the identity identification processing are both referred to as the relative position data of the detected object OBk, i, and these are newly identified in the detection order i. .

In the current registration data storage area (k) shown in FIG. 11, the detected objects OBk, 1 and OB shown in FIG.
k, 2, OBk, 8 and OBk, 9, and OBk, 10 and OBk11
Are collected, and the detected objects OBk, 1, OBk, 7
And OBk, 8 are stored in the current registration data storage area (k) as relative position data. The other detected objects OBk, 3, OBk, 4, OBk, 5, OBk,
The relative position data of 6 and OBk, 7 remain unchanged (detection number i is changed), and the detected objects OBk, 2, OBk, 3
, OBk, 4, OBk, 5 and OBk, 6 are stored in the current registration data storage area (k) as relative position data.

3.3 Calculation of Movement Vector of Detected Object (FIG. 3; Step 22)

The motion vector Vk, i of the detected object OBk, i
(Vx, k, i, Vy, k, i) is the last processing cycle (k−
Relative position Qk-1, j of detected object OBk-1, j in 1)
(Xk-1, j, yk-1, j) and the detected object OBk-1, j
Relative position Qk, i (xk, i, yk, y) of the detected object OBk, i in the current processing cycle k considered to be the same as
i).

The movement vector Vk, i (Vx, k, i, Vy, k, i
), First, the detected object OBk, i in the current processing cycle k, which is considered to be the same as the detected object OBk-1, j detected in the previous processing cycle (k-1), is found. Have to do it. The detected object OBk-1, j in the previous processing cycle (k-1)
And the detected object OBk, i in the current processing cycle k
Is determined as follows.

The relative position Qk-1, i (xk-1, j, y) of the detected object OBk-1, j in the previous processing cycle (k-1)
The window is set based on k-1, j). The windows include a first window W1 and a second window W2.

The first window W1 is a detected object OBk-1, j for which the motion vector Vk-1, j has not been calculated in the previous processing cycle (k-1) (in the previous processing cycle (k-1)). The newly detected object OBk-
1, j; hereinafter simply referred to as “newly detected object”).

As shown in FIG. 13, the first window W1 is a rectangle represented by two diagonal points, that is, a start point S1 (x1S, y1S) and an end point S2 (x2S, y2S). The coordinates of the start point S1 and the end point S2 are respectively the detected object OBk-1, j in the previous processing cycle (k-1).
Relative position Qk-1, j (xk-1, j, yk-1, j), vehicle speed vk
And the processing cycle time interval T, it is expressed by the following equation.

[0125]

X1S = xk-1, j-0.5 (9) y1S = {yk-1, j + (27.8-vk) .times.T} +0.5 (10) x2S = xk-1, j + 0. 5… (11) y2S = {yk-1, j−vk × T} −0.5… (12)

The first window W1 is based on the assumption that the difference between the speed of the host vehicle and the speed of the preceding vehicle is 100 [km / h] (= 27.8 [m / s]) or less. In other words, in the extreme case, even if one of the host vehicle and the preceding vehicle is stopped and the other is traveling at 100 [km / h], the identity of the detected object can be recognized. Is set to Therefore, in equation (10), (27.8−vk) × T term is
In (12), the term (−vk × T) is added. A system error (in this embodiment, 0.5 [m]) is added to or subtracted from each coordinate of the start point S1 and the end point S2 due to a measurement error or the like. On a highway, the maximum speed allowed on the road is usually less than 100 km / h, and if the preceding vehicle is traveling at a speed exceeding 100 km / h, the speed will be excessive. This is because it is a violation.

The detected object OBk-1, j (preceding vehicle) is 27.
When traveling at 8 [m / s] (= 100 [km / h]), the speed difference from the host vehicle is (27.8-vk) [m / s], and the processing cycle time interval T [s] Is the distance difference at (27.8−vk) × T
[m]. The relative position Qk-1, j (xk-1, j, y) of the detected object OBk-1, j in the previous processing cycle (k-1)
k-1, j) and the detected object OBk, i detected in the same processing cycle k this time fall within W1 of the first window.

When the detected object OBk-1, j (preceding vehicle) is stationary, the speed difference from the own vehicle is (-vk), and
The distance difference in the processing cycle time interval is (−vk) × T
It is. The relative position Qk-1, j (xk-1, j, yk-1, of the detected object OBk-1, j in the previous processing cycle (k-1)
j) and the detected object OBk, i detected in the same current processing cycle k fall within W1 of the first window.

The second window W2 has the motion vector Vk-1, j (Vx, k-1, V) in the previous processing cycle (k-1).
j, Vy, k-1, j) for which the detected object OBk-1, j has been calculated.
(The detected object O which has been continuously detected twice or more in the past
Bk-1, j), its relative position Qk-1, j (xk-1, j)
, Yk-1, j) and the movement vector Vk-1, j (Vx, k-1,
j, Vy, k-1, j).

The relative position Qk-1, j (xk-1, j, y) of the detected object OBk-1, j in the previous processing cycle (k-1)
k-1, j) and the movement vector Vk-1, j (Vx, k-1, j, V
y, k-1, j), the detected object OBk, i detected in the current processing cycle k relative position Qk, i (xk, i)
i, yk, i) can be predicted. The second window W2 is set so as to track the same detected object in time series.

As shown in FIG. 14, the second window W2 has two diagonal points, ie, a starting point T1 (x1T, y1T).
And an end point T2 (x2T, y2T). The starting point T1 and the ending point T2 are respectively the relative position Qk-1, j (xk-1, j, yk-1, j) and the movement vector of the detected object OBk-1, j in the previous processing cycle (k-1). Based on Vk-1, j (Vx, k-1, j, Vy, k-1, j) and the processing cycle time interval T, it is expressed by the following equation.

[0132]

X1T = (xk-1, j + Vx, k-1, j × T) -0.5 (13) y1T = (yk-1, j + Vy, k-1, j × T) +0.5 ... (14) x2T = (xk-1, j + Vx, k-1, j × T) +0.5 (15) y2T = (yk-1, j + Vy, k-1, j × T) -0.5 ... ( 16)

The coordinates of the second window W2 are based on the relative position Qk-1, j (xk-1, j, yk-1, j) of the detected object OBk-1, j.
) And the movement vector Vk-1, j (Vx, k-1, j, Vy, k-1, j
), The expected position (xk-1, j + Vx, k-1, j × T, yk) of the detected object OBk, i in the current processing cycle k.
This is obtained by adding and subtracting a system error (0.5 [m]) around (−1, j + Vy, k−1, j × T).

The window (first window W1 or second window W2) set for the detected object OBk-1, j detected in the previous processing cycle (k-1)
) Is checked for all the detected objects OBk, i in the current processing cycle k. The window is set for all of the previous detected objects OBk-1, j.

As a result of the inspection, the previous processing cycle (k-
If one detected object OBk, i exists in the window set for the detected object OBk-1, j in 1), the relative position Qk, i (x
k, i, yk, i) and the object to be detected OBk-1, j serving as the reference of the window are considered to be the same, and their relative positions Qk-1, j (xk-1, j, yk) -1, j) and relative position Q
Based on k, i (xk, i, yk, i) and the processing cycle time interval T, the movement vector Vk, i of the detected object OBk, i
(Vx, k, i, Vy, k, i) will be calculated. The movement vector Vk, i (Vx, k, i, Vy, k, i) is given by the following equation.

[0136]

Vx, k, i = (xk-1, j−xk, i) / T (17) Vy, k, i = (yk−1, j−yk, i) / T (18)

FIG. 15 shows the relative position Qk-1, j (xk-j) of the detected object OBk-1, j in the previous processing cycle (k-1).
1, j, yk-1, j) and the movement vector Vk, i (in the current processing cycle k) calculated based on the relative position Qk, i (xk, i, yk, i) in the current processing cycle k. V
x, k, i, Vy, k, i).

Since the calculated movement vector may include a noise component such as a measurement error, the noise is removed by using a digital filter or the like.
More accurate movement vectors can be calculated.

The current processing cycle k in one window
If there are a plurality of detected objects in, the detected object OB located closest to the center of the window
k and i are selected. The selected object to be detected OBk, i,
The detected object OBk-1, j which is the reference of the window is regarded as the same, and their relative positions Qk-1, j (xk-1, j,
yk-1, j), the relative position Qk, i (xk, i, yk, i) and the processing cycle time interval T,
According to (18), the movement vector Vk, i (Vx, k, i, Vy, k, i) of the detected object OBk, i is calculated.

The current processing cycle k in one window
, The detected object OBk, i located closest to the center of the window
Is one of the following:

(A) Object that is actually the same as the detected object in the previous processing cycle (b) Object that is actually different from the detected object in the previous processing cycle (c) Object that appears due to noise

Since the detected object OBk, i in (a) is the same, it is preferable that this is selected.

Regarding the detected object OBk, i in (b), different objects (referred to as different detected objects) are actually regarded as the same, so that the calculated motion vector becomes erroneous, and Means that there is another true motion vector obtained for the same detected object. If the same object as the foreign object detection object exists in the second window W2 set in the next processing cycle (k + 1), the foreign object detection object will be tracked. The detected object will be tracked as a new detected object from the next processing cycle (k + 1). The first window W1 is set for other detected objects not selected.

Since the detected object in (c) is due to noise, even if the movement vector is calculated in the current processing cycle k, it is not always detected in the next processing cycle (k + 1). Also, even if it is detected, it does not necessarily exist in the set second window W2.
Therefore, in the next processing cycle (k + 1), most of the detected objects due to noise will disappear. Even if it appears in the window two or more times and is recognized as the same object to be detected, it will eventually disappear with the passage of time.

FIG. 16 shows the relative position Q of the detected object OBk-1, j in the previous processing cycle (k-1) stored in the previous registration data storage area (k-1) in the memory.
k-1, j (xk-1, j, yk-1, j) and the movement vector Vk-
1, j (Vx, k-1, j, Vy, k-1, j) are shown. FIG.
, "*" Indicates that no data is stored. In FIG. 16, data of eight detected objects OBk-1, j are stored, and no movement vector is stored only for the detected objects OBk, 4.

FIG. 17 is a flow chart showing the calculation processing of the movement vector. In the flow chart of FIG. 17, j represents the j-th detected object OBk-1, j in the previous processing cycle (k-1), and Nk-1 is the previously registered data storage area (k- 1) represents the total number of detected objects OBk-1, j stored in 1).

J is initialized to 1 (step 51), and the relative position Qk-1,1 (xk-1,1) of the first detected object OBk-1,1 in the previous processing cycle (k-1) 1, yk-1,1
) And the movement vector Vk-1,1 (Vx, k-1,1, Vy, k-
1,1) is read from the previous registration data storage area (k-1). The process of calculating the motion vector is started from the first detected object OBk-1,1.

In the example shown in FIG. 16, the relative position Qk-1,1 (-
4.5, 49.9) and the movement vector Vk-1,1 (-2, -22)
Is read.

It is determined whether the detected object OBk-1, j is a newly detected object (step 53). The detected object OBk-1,
Whether j is a newly detected object is determined by the detected object O
The motion vector Vk-1, j (Vx, k-1, j, Vy, k-
1, j) is stored in the previous registration data storage area (k-1). The movement vector Vk-1,
If j is not stored, the detected object OBk-1, j
Is determined as a newly detected object.

If the detected object OBk-1, j is a new detected object (YES in step 53), the detected object OBk-1, j
The first window W1 is set on the basis of the relative position Qk-1, j (xk-1, j, yk-1, j) (step 54).

In the example shown in FIG. 16, the object to be detected OBk-1,4
Since no movement vector is stored for only the detected object OBk-1,4, the first window W
1 will be set.

In step 53, the detected object OBk-1,
If j is determined not to be a newly detected object (at step 53
NO), the relative position Qk-1, j of the detected object OBk-1, j
(Xk-1, j, yk-1, j) and the movement vector Vk-1, j
The second window W2 is set based on (Vx, k-1, j, Vy, k-1, j) (step 55).

In the example shown in FIG. 16, since the movement vectors are stored for the detected objects OBk-1, j other than the detected object OBk, 4, the second window is used except for the detected object OBk, 4. W2 will be set.

When either the first window W1 or the second window W2 is set, the detected object OBk, i in the current processing cycle k stored in the current registration data storage area (k) is set. It is checked whether any one or more of them are present in the window (step 56).

If the detected object OBk, i in the current processing cycle k exists in the window (YE at step 56)
S), the relative position Qk-1, j (xk-
1, j, yk-1, j) and the relative position Qk, i of the detected object OBk, i
(17) and (1) based on (xk, i, yk, i)
According to 8), the movement vector V of the detected object OBk, i
k, i (Vx, k, i, Vy, k, i) is calculated (step 5).
7).

For example, the detected object OBk-1, shown in FIG.
Since the detected object OBk, 1 shown in FIG. 12 exists in the second window W2 based on the above equation (2), the above equations (17) and (1)
The movement vector Vk, 1 (Vx, k, 1, Vy, k, 1) is calculated for the detected object OBk, 1 according to 8). That is, Vx, k, 1 = ((-4.7)-(-4.5)) / 0.1 = (-2), Vy,
k, 1 = (49.9-47.8) /0.1 = (-22).

The calculated movement vector Vk, i (Vx, k, i
, Vy, k, i), as shown in FIG.
It is stored in the current registration data storage area (k) of the memory in association with the relative position Qk, i (xk, i, yk, i) of k, i.

J is incremented (step 58).
Returning to step 53, steps 53-58 for the new j
Is repeated.

The above processing (steps 53 to 58) is repeated while j is sequentially incremented. If j> Nk-1 is satisfied, Nk-1 detected objects in the previous processing cycle (k-1) are processed. , The motion vector calculation processing is completed. (Step 52).

In this way, as shown in FIG. 18, the movement vector of the detected object in the current processing cycle k is calculated. In FIG. 18, the detected object OBk,
Since 4 is the first to appear in the current processing cycle k (that is, the detected object that is the same as OBk, 4 has no previous data, no motion vector has been calculated.

3.4 Identification of Vehicle and Roadside Reflector (FIG. 3; Step 23)

As described above, the object detected as the object to be detected includes the roadside reflector and the vehicle reflector attached to the vehicle or the motorcycle. It is determined whether the detected object is a roadside reflector or two vehicle reflectors attached to one vehicle. The detected object determined to be a roadside reflector is used in the processing of “3.6 Recognition of Roadside Shape and Its Reliability Evaluation” through the processing of “3.5 Classification of Roadside Reflector” described later. The vehicle reflectors mounted on one vehicle are combined into one and handled as one object (vehicle) as described below.

3.4.1 Vehicle Identification

As described above, two vehicle reflectors are mounted at the rear of the vehicle. The relative positions and the movement vectors of two vehicle reflectors (detected objects) mounted on one vehicle have a fixed relationship determined by the physical condition that they are mounted on one vehicle.
This fixed relationship (hereinafter referred to as “pair condition”) is defined as 2
The relative positions Qk, i (xk, i, yk, i) and Qk, j (xk, j, y) of the detected objects OBk, i and OBk, j
k, j) at a distance corresponding to the width of the vehicle in the width direction,
And their movement vectors Vk, i (Vx, k, i, Vy, k,
i) and Vk, j (Vx, k, j, Vy, k, j) are substantially equal. That is, the pair condition is represented by the following equation.

[0165]

| Xk, i −xk, j | <Thc1 (19) | yk, i −yk, j | <Thc2 (20) | Vx, k, i −Vx, k, j | <Thc3 ... (21) | Vy, k, i−Vy, k, j | <Thc4 (22) Thc1 is a threshold value representing a distance corresponding to the width of one vehicle (almost in the X direction), and Thc2 is a traveling direction (almost Y). The threshold values Thc3 and Thc4 indicating the allowable range in the direction (i.e., direction) are threshold values indicating that the motion vectors are substantially equal.

The vehicle is identified by the pair conditions shown in equations (19) to (22). Two detected objects OB satisfying the pair condition
k, i and OBk, j are combined into one. In the following, a combination of two detected objects is referred to as a “vehicle candidate object”.

The center relative position of the vehicle candidate object is determined by the relative position Qk, i (x
k, i, yk, i) and the relative position Qk, of the detected object OBk, j.
j (xk, j, yk, j). That is,
[(Xk, i + xk, j) / 2, (yk, i + yk, j) / 2]
It is.

The movement vector of the vehicle candidate object is the movement vector Vk, i (Vx, k, i, V) of the detected object OBk, i.
y, k, i) and the movement vector Vk, j of the detected object OBk, j
(Vx, k, j, Vy, k, j). That is,
[(Vx, k, i + Vx, k, j) / 2, (Vy, k, i + Vy, k, j
) / 2].

When a candidate vehicle object is found in the following vehicle identification processing, a flag Fc (for example, Fc is "010") is used to indicate that.

In this vehicle identification processing, two different detected objects OBk, i and OB in the current processing cycle k
It is checked whether the above pair condition is satisfied for all combinations of k and j. In this vehicle identification process,
One detected object and another first detected object may satisfy a pair condition, and the one detected object may satisfy a pair condition with another second detected object.

For example, when the vehicle is stopped near the roadside, two vehicle reflectors attached to the left and right sides of the vehicle and a roadside reflector on the left side in the traveling direction of the vehicle are detected as detected objects. Is done. The pair of two vehicle reflectors clearly satisfies the pair condition and is determined to be a vehicle candidate object. A pair of a vehicle reflector and a roadside reflector on the left side of the vehicle may also satisfy the pair condition, and these are also determined to be vehicle candidate objects. The pair of the vehicle reflector and the roadside reflector on the left side of the vehicle is regarded as “vehicle”. The center relative position of the vehicle candidate object is an intermediate position between the relative position of the vehicle reflector on the left side of the vehicle and the relative position of the roadside reflector. There is no problem because the vehicle candidate object exists at a position closer to the road side than the actual vehicle position.

When the motorcycle is stopped near the road, the vehicle reflector attached to the motorcycle and the road reflector on the left side of the motorcycle are detected as objects to be detected. A pair of a vehicle reflector and a roadside reflector may satisfy a pair condition, and these are determined to be vehicle candidate objects. The vehicle reflector and the roadside reflector of the motorcycle are regarded as “vehicles”. in this case,
As in the case where the vehicle reflector on the left side of the vehicle and the roadside reflector are regarded as “vehicle candidate objects”, the center relative position of the vehicle candidate object is intermediate between the relative position of the two-wheeled vehicle reflector and the relative position of the roadside reflector. Position,
There is no problem because the vehicle candidate object exists at a position closer to the road side than the position of the motorcycle.

FIG. 19 is a flowchart showing a process of identifying a vehicle candidate object. In this flow chart, i and j (j> i) represent the ith detected object OBk, i and the jth detected object OBk, j in the current processing cycle k, respectively, and Nk is the memory Represents the total number of detected objects OBk, i stored in the current registration data storage area (k).

I is initialized to 1 (step 61), and the first
The relative position Qk, 1 (xk, 1, y) of the object to be detected OBk, 1
k, 1) and the movement vector Vk, 1 (Vx, k, 1, Vy, k, 1
) Is read from the current registration data storage area (k). The identification processing is started from the first detected object OBk, 1.

In the example shown in FIG. 18, the relative position Qk, 1 (-4.
7, 47.8) and the movement vector Vk, 1 (-1, -21) are read.

J is set to i + 1 (step 63),
The relative position Qk, j (xk, j) of the j-th detected object OBk, j
j, yk, j) and the movement vector Vk, j (Vx, k, j, V
y, k, j) are read from the current registration data storage area (k).

The relative position Qk, i (xk, i, yk, i) and the movement vector Vk, i of the i-th detected object OBk, i
(Vx, k, i, Vy, k, i) and the j-th object to be detected OB
The relative position Qk, j (xk, j, yk, j) of k, j and the movement vector Vk, j (Vx, k, j, Vy, k, j) are expressed by the above equation (19).
It is determined whether the pair condition of Expression (22) is simultaneously satisfied (step 65).

If the detected objects OBk, i and OBk, j satisfy the pair condition (YES in step 65), these detected objects OBk, i and OBk, j become vehicle candidate objects. The center relative value and the movement vector of the vehicle candidate object are calculated, and the calculated relative position and the movement vector are calculated together with the flag Fc representing the vehicle candidate object in the processing order according to the evaluation order provided in the memory shown in FIG. It is registered (stored) in the data storage area (step 66).

In the example shown in FIG. 18, the detected objects OBk, 5 and OBk, 6 are determined to be vehicle candidate objects. The detected objects OBk, 7 and OBk, 8 are also determined to be vehicle candidate objects.

J is incremented (step 67), and
The processing of steps 65 and 66 is repeated until j = Nk (step 64).

When j> Nk (NO in step 64), it is determined whether or not the detected object OBk, i is determined to constitute a vehicle candidate object (whether at least once in step 65 is YES). (Step 68).

When it is determined that the detected object OBk, i forms a vehicle candidate object (YES in step 68), no processing is performed, but the detected object OBk, i forms a vehicle candidate object. If it is not determined that the object OBk, i is detected (NO in step 68), the relative position Qk, i
(Xk, i, yk, i) and the movement vector Vk, i (Vx, k,
i, Vy, k, i) are directly registered (stored) in the evaluation data storage area (step 69).

In the example shown in FIG. 18, the objects to be detected OBk, 1,
Since OBk, 2, OBk, 3 and OBk, 4 have no other detected objects that satisfy the pair condition, they are not determined to constitute vehicle candidate objects. As shown in FIG. 20, the relative position and the movement vector of these detected objects are registered (stored) in the evaluation data storage area as they are, and the flag is set to “*” (for example, the flag “*” is set to “000”). is there)
Becomes

Thereafter, i is incremented (step 70), and the process returns to step 43, where the processing of steps 64 to 69 is repeated for a new i.

The above processing (steps 63 to 69) is repeated while sequentially incrementing i. If i> Nk−1, any two selected from the Nk detected objects are detected. This means that the inspection as to whether the pair condition is satisfied has been completed for all combinations of the detected objects.
(Step 62).

When the vehicle identification process is performed on the data shown in FIG. 18 in this manner, the data shown in FIG. 20 is obtained. The data (xk, i, yk, i, Vx, k, i, Vy, k, i and the flags) stored in the evaluation data storage area are renumbered according to the order of the storage area. .

3.4.2 Roadside reflector identification

The roadside reflector identification processing is performed for all the detected objects OBk, i except for the vehicle candidate objects (the detected objects OBk, i having the flag Fc).
It identifies whether k and i are roadside reflectors.

Since the roadside reflector is stationary, it looks as if it is traveling in the Y direction at the speed (-vk) from the own vehicle traveling at the vehicle speed vk. The object to be detected OBk,
The roadside reflector can be identified using the Y component Vy, k, i of the movement vector Vk, i (Vx, k, i, Vy, k, i) of i and the vehicle speed vk.

The motion vector Vk, i of the detected object OBk, i
Y component Vy, k, i of (Vx, k, i, Vy, k, i) is negative,
When the absolute value of the difference between the absolute value and the vehicle speed vk is less than a predetermined threshold value Thr (for example, 20 [km / h] = 5.6 [m / s]), the detected object OBk, i is output from the roadside reflector. Is determined. That is, the roadside reflector determination condition is V
y, k, i <0 and | -Vy, k, i-vk | <Thr.

Hereinafter, the detected object determined to be the roadside reflector is referred to as a “roadside reflector candidate object”. The roadside reflector candidate object is identified by the flag Fr (for example, Fr is “100”).

In many cases, a vehicle or a motorcycle stopped on a road has already been identified as a vehicle candidate object (a detected object whose flag is Fc) as described above. When the motorcycle is stopped slightly away from the road, the motorcycle is mistaken for a roadside reflector. If such a stopped motorcycle is misidentified as a roadside reflector candidate object and is used in the processing of “3.6.1 Roadside shape recognition” described later, roadside shape recognition will be inaccurate. However, such inaccurate roadside shapes are eliminated by the process of “3.6.2 Roadside shape reliability evaluation” performed after the process of “3.6.1 Roadside shape recognition”.

Even if the motorcycle stopped near the center of the road is erroneously recognized as a roadside reflector candidate object, all the detected objects OBk, i detected will be described in “3.7 Relative Position and Movement Vector of Previous Vehicle” described later. There is no problem because it is determined that the vehicle is the preceding vehicle in the process of "Evaluation of the traveling lane of the preceding vehicle by the vehicle" and the process of "3.8 Evaluation of the traveling lane of the preceding vehicle based only on the relative position of the preceding vehicle".

FIG. 21 is a flow chart showing the roadside reflector identification processing. In this flowchart, Nk represents the total number of detected objects OBk, i stored in the evaluation data storage area of the memory.

The i is initialized to 1 (step 71), and the first
The identification process is started from the detected object OBk, 1.
At this time, the vehicle speed vk is stored in the vehicle speed data storage area of the memory.
Is read.

Whether the i-th detected object OBk, i is a candidate vehicle object is determined by the flag Fc (step 73).

In the example shown in FIG. 20, the detected objects OBk, 5 and OBk, 6 are determined to be vehicle candidate objects.

If the detected object OBk, i is not a vehicle candidate object (NO in step 73), the movement vector Vk, i (Vx, k, i, Vy, k, i) of the detected object OBk, i is evaluated. It is read from the data storage area, and it is determined whether or not the above-described roadside reflector determination condition is satisfied (step 74).

If the roadside reflector determination condition is satisfied (YES in step 74), the detected object OBk, i is determined to be a roadside reflector candidate object, and the flag “*” of the detected object OBk, i is set. It is changed to Fr (step 75).

In the example shown in FIG. 20, for the detected object OBk, 1, the movement vector Vk, 1 (Vx, k, 1, Vy, k, 1
) Is (-2, -22). The vehicle speed vk is vk = 21. Therefore, | −Vy, k, 1 −vk | = | − (− 22) −21
| = 1 (<5.6). It is determined that the detected object OBk, 1 is a roadside reflector candidate object. Similarly, the detected objects OBk, 2 and OBk, 3 are also determined to be roadside reflector candidate objects.

I is incremented (step 75),
Returning to step 73, steps 74-76 for the new i
Is repeated.

If there is a detected object OBk, i that does not satisfy the roadside reflector determination condition in step 74, the flag "*" for the detected object OBk, i is kept as it is.

The above processing (steps 73 to 76) is repeated while sequentially incrementing i. If i> Nk, the identification processing of the roadside reflector is completed for Nk detected objects OBk, i. Become. (Step 72).

In this manner, when the reflector identification processing is performed on the data shown in FIG. 20, data as shown in FIG. 22 is obtained. For the detected object OBk, 4,
Since there is no movement vector Vk, 4 (Vx, k, 4, Vy, k, 4), it cannot be determined whether the object is a roadside reflector candidate object, and the flag remains “*”.

3.5 Clustering of Roadside Reflector (FIG. 3; Step 24)

The relative position Qk, i of the roadside reflector candidate object (detected object OBk, i whose flag is Fr)
(Xk, i, yk, i) and the movement vector Vk, i (Vx, k,
i, Vy, k, i), it is determined whether the roadside reflector candidate object is provided on the left or right side of the road. This determination is the clustering of the roadside reflector.

When the roadside reflector candidate object is determined to be on the right side by clustering, the flag Fr of the roadside reflector candidate object is a flag Frr indicating that it is the right roadside reflector (for example, Frr is "101").
Is changed to If it is determined that the roadside reflector candidate object is the left roadside reflector, the flag Fr is set to a flag Frl representing the right roadside reflector (for example, the flag Fr is “1”).
10 ").

There are two types of clustering of the roadside reflector, and one of them is used. Hereinafter, these two types of processing will be described.

3.5.1 Clustering of Roadside Reflector Based on Relative Position and Movement Vector of Roadside Reflector (Part 1)

The roadside reflector provided on the right side of the road on which the vehicle is traveling relatively moves on the right side in the direction opposite to the traveling direction when viewed from the vehicle. The roadside reflector provided on the left side relatively moves on the left side in a direction opposite to the traveling direction. Since the position of the vehicle is the origin O, the left roadside reflector candidate object moves on the left side of the X axis, and the right roadside reflector candidate object moves on the right side of the X axis. The roadside reflector candidate objects are clustered by points (passing points) on the X-axis crossed by the roadside reflector candidate objects.

The passing point is determined by the relative position Qk, i (xk, i, yk, i) and the movement vector Vk, i of the roadside reflector candidate object.
(Vx, k, i, Vy, k, i). As shown in FIG. 23, when the movement vector Vk, i (Vx, k, i, Vy, k, i) of the roadside reflector candidate object is extended to the X-axis, as shown in FIG. It is an intersection. If the X coordinate xr, i of the passing point (intersection between the extension line and the X axis) is negative, that is, if xr, i <0, it is determined that the roadside reflector candidate object is provided on the left side of the road. . If the X coordinate xr, i of the passing point (the intersection of the extension line and the X axis) is positive, that is, if xr, i> 0, it is determined that the roadside reflector candidate object is provided on the right side of the road .

Moving vector V of roadside reflector candidate object
The X coordinate xr, i of the intersection between the extension line of k, i and the X axis is the relative position Qk, i (xk, i, yk, i) of the roadside reflector candidate object.
And the motion vector Vk, i (Vx, k, i, Vy, k, i). From the relative position Qk, i (xk, i, yk, i) of the roadside reflector candidate object to the movement along the movement vector Vk, i (Vx, k, i, Vy, k, i) to the X axis Since the time is Vy, k, i <0, [yk, i /
(-Vy, k, i)]. The travel distance in the X-axis direction during the time from the relative position Qk, i (xk, i, yk, i) of the roadside reflector candidate object to the X-axis is [Vx, k,
i · {yk, i / (− Vy, k, i)}]. Therefore, the X coordinate xr, i of the intersection is represented by the following equation.

[0213]

Xr, i = xk, i + Vx, k, i · {yk, i / (− Vy, k, i)} (23)

As described above, clustering is performed according to whether the X coordinate xr, i of the passing point calculated for the roadside reflector candidate object is positive or negative.

3.5.2 Roadside Reflector Clustering Based on Relative Position and Movement Vector of Roadside Reflector (Part 2)

Since the clustering process differs depending on whether the road on which the roadside reflector is provided is straight or curved, first, whether the road is straight or not,
It is determined whether the vehicle is curved.

When the roadside reflector is provided on a straight road, and the vehicle is traveling along the road, the movement of the roadside reflector with respect to the vehicle is small in the X direction. Also, when the roadside reflector is provided on a curved road and the vehicle is traveling along the road, the roadside reflector moves more in the X direction than when it is straight.

Therefore, whether the road is straight or curved is determined by the X-direction component Vx, k of the movement vector Vk, i (Vx, k, i, Vy, k, i) of the roadside reflector candidate object. , i.

Moving vector V of roadside reflector candidate object
The absolute value of the X component Vx, k, i of k, i (Vx, k, i, Vy, k, i) is equal to or less than a predetermined threshold (for example, 0.1 [m / s]), that is,
If | Vx, k, i | <0.1 [m / s], it is determined that the roadside reflector candidate object is a roadside reflector provided on a straight road. If | Vx, k, i |> 0.1 [m / s], the roadside reflector candidate object is determined to be a roadside reflector provided on a curved road.

(1) When the road is straight

Since the road on which the own vehicle is traveling is straight, as shown in FIG. 24, the roadside reflector provided on the left side of the road moves on the left side of the own vehicle and the roadside reflector provided on the right side of the road. The reflector moves on the right side of the vehicle.

Movement vector V of roadside reflector candidate object
Since the X component Vx, k, i of k, i (Vx, k, i, Vy, k, i) is very small, the passing point of the roadside reflector (intersection with the X axis) is represented by its relative position Qk, i ( xk, i, yk, i) x-coordinate xk, i
Can be represented by xk, i is positive, that is, x
If k, i> 0, it is determined that the roadside reflector candidate object is provided on the right side of the road. Further, if the X coordinate xk, i of the relative position Qk, i (xk, i, yk, i) of the roadside reflector candidate object is negative, that is, if xk, i <0, the detected object OBk, i is a road. Is determined to be provided on the left side of.

(2) When the road is curved

FIG. 25 shows an example of the relative movement of the roadside reflector candidate object. The movement vector Vk, i (Vx, k, i, Vy, k, i) of the roadside reflector candidate object is shown with its relative position Qk, i (xk, i, yk, i) as the starting point.
The movement vector Vk, i (Vx, k,
The direction of i, Vy, k, i) depends on its relative position Qk, i (xk, i).
, Yk, i) can be considered as the tangential direction of the circle representing the curve of the road.

When the vehicle is traveling on a curved road, it is assumed that the vehicle is traveling along the road.
When viewed from the host vehicle (origin O), the roadside reflector moves along a circle (the center is on the X axis) representing the approximate roadside shape. Since the position of the own vehicle is the origin O, the intersection of the circle and the X axis indicates the distance from the own vehicle to the road side.
Based on the X coordinate of the intersection of the circle and the X axis, it is determined whether the roadside reflector candidate object is provided on the right side or on the left side of the road.

Since the center of the above-mentioned circle is on the X-axis, if the center of the circle is (xro, i, 0) and the radius is Rr, i,
The equation of this circle is expressed by the following equation.

[0227]

(X−xro, i) ² + y ² = Rr, i ² (24)

In the equation (24), the unknowns xro, i and Rr,
i is obtained as follows.

It is determined whether the road has a right curve or a left curve. If the road is a right curve, the roadside reflector moves to the left of the host vehicle, and if the road is a left curve, the roadside reflector moves to the right of the host vehicle. Movement vector Vk, i of the roadside reflector candidate object (Vx, k, i
, Vy, k, i) are negative, that is, Vx, k, i
If k, i <0, it is determined that the road on which the roadside reflector candidate object is provided has a right curve. Movement vector Vk, i of the roadside reflector candidate object (Vx, k, i, V
y, k, i) has a positive X component Vx, k, i, that is, Vx, k, i>
If it is 0, it is determined that the road on which the roadside reflector candidate object is provided has a left curve. Hereinafter, a case where the road has a right curve will be described.

If the road has a right curve (Vx, k, i <
0), if the angle between the movement vector Vk, i of the roadside reflector candidate object and its X component Vx, k, i is φi, the following equation is established.

[0231]

[Equation 11] tanφi = Vx, k, i / Vy, k, i (25)

On the other hand, since the center of the circle on which the roadside reflector candidate object relatively moves is on the X axis, a straight line connecting the center of the circle and the roadside reflector candidate object (a straight line in the radial direction of the circle)
And the angle formed by the X axis is φi. Therefore, the following equation is established.

[0233]

Tanφi = yk, i / (xro, i−xk, i) (26)

When tanφi is deleted from equations (25) and (26), the X coordinate xro, i of the center of the circle is obtained as follows.

[0235]

Vx, k, i / Vy, k, i = yk, i / (xro, i−xk, i) ∴xro, i = xk, i + yk, i · (Vy, k, i / Vx, k, i)… (27)

The radius Rr, i of the circle is obtained by calculating the relative position Qk, i (xk, i, yk, i) of the roadside reflector candidate object by the formula (2)
It is obtained by substituting x and y in 4).

[0237]

(Xro, i−xk, i) ² + yk, i ² = Rr ² ∴Rr, i = sqrt [(xro, i−xk, i) ² + yk, i ² ] (28) sqrt [] represents a square root.

By substituting equation (27) for xro, i in equation (28), the radius Rr, i is expressed by the following equation.

[0239]

Equation 15] Rr, i = sqrt [{( xk, i + yk, i · (Vy, k, i / Vx, k, i) -xk, i} 2 + yk, i 2] = yk, i · sqrt [ (Vy, k, i / Vx, k, i) ² +1] (29)

Further, the X coordinate xr, i of the intersection of the circle and the X axis
Is expressed by the following equation from the relationship between the X coordinate xro, i of the center of the circle and its radius Rr, i.

[0241]

Xr, i = xro, i-Rr, i (30)

Equation (27) is replaced by xro, i of equation (30), and equation (28) is replaced by equation (3)
By substituting into Rr, i of (0), the X coordinate xr, i of the intersection is expressed by the following equation.

[0243]

Xr, i = xk, i + yk, i · {(Vy, k, i / Vx, k, i) −sqrt [(Vy, k, i / Vx, k, i) ² +1]} (31)

When the road has a left curve (Vx, k, i> 0),
The X coordinate xr, i of the intersection of the circle and the X-axis drawn by the relative movement of the roadside reflector candidate object is xr, i in equation (30).
Xr, i = xro, i + Rr, i instead of = xro, i-Rr, i.

[0245]

Xr, i = xk, i + yk, i · {(Vy, k, i / Vx, k, i) + sqrt [(Vy, k, i / Vx, k, i) ² +1]} ... ( 32)

The X coordinate xr, i of the intersection thus obtained
Is positive or negative, it is determined whether the roadside reflector exists on the left or right side of the road.

FIG. 28 is a flow chart showing the clustering processing of the roadside reflector (common to (part 1) and (part 2)).
It is a chart.

The i is initialized to 1 (step 81), and the first
The clustering process is started from the detected object OBk, 1.

Whether the i-th detected object OBk, i is a roadside reflector candidate object is determined by a flag in the evaluation data storage area (step 83). If the flag of the detected object OBk, i is Fr, it is determined that the object is a roadside reflector candidate object.

In the example shown in FIG. 22, the detected objects OBk, 1,.
Since Fr is stored in the flag storage area, OBk, 2 and OBk, 3 are determined to be roadside reflector candidate objects.

If the detected object OBk, i is a roadside reflector candidate object (NO in step 83), its relative position Qk, i
(Xk, i, yk, i) and the movement vector Vk, i (Vx, k,
i, Vy, k, i) are read from the evaluation data storage area, and the X coordinate xr, i at the intersection with the X axis is calculated according to equation (23) (or equations (31), (32)). (Step 84).

When the calculated X coordinate xr, i of the intersection is xr, i <
If 0 (YES in step 85), the detected object OB
k, i is registered as a left roadside reflector (step 86). That is, the flag Fr associated with the roadside reflector candidate object is changed to the flag Frl representing the left roadside reflector.
Is changed to

If the calculated X coordinate of the intersection is xr, i> 0 (NO in step 85), the roadside reflector candidate object is registered as the right roadside reflector (step 87). That is, the flag Fr attached to the roadside reflector candidate object is changed to the flag Frr representing the left roadside reflector.

In the example shown in FIG. 22, for the detected object OBk, 1 which is a roadside reflector candidate object, when xr, 1 is calculated according to the equation (23), xr, 1 = (− 4.7) + (− 1) ) ・
Since (47.8 / (-(-21))) = (-7.0) (<0), it is determined that the detected object OBk, 1 is the left roadside reflector. The flag Fr of the detected object OBk, 1 is changed to a flag Frl representing the left roadside reflector. Similarly, the detected objects OBk, 2 and OBk, 3 are also determined to be the left roadside reflectors, and the flag Fr is changed to the flag Frl.

I is incremented (step 88).
Returning to step 83, for a new i, steps 83 to 88
Is repeated. If the detected object OBk, i specified by i in the evaluation data storage area is not a roadside reflector candidate object (NO in step 83), the processing in steps 84 to 87 is skipped.

In the example shown in FIG. 22, the detected objects OBk, 4,
Since OBk, 5 and OBk, 6 are not roadside reflector candidate objects, steps 84 to 87 are skipped for these objects.

The above processing (steps 83 to 88) is repeated while sequentially incrementing i, and if i> Nk, the clustering processing of the roadside reflector has been completed for Nk detected objects. (Step 8
2).

In this way, clustering is performed on the roadside reflector candidate object. When the clustering process of the roadside reflector candidate object is performed on the data shown in FIG. 22, data as shown in FIG. 27 is obtained.

3.6 Recognition of Roadside Shape and Evaluation of Its Reliability (FIG. 3; Step 25)

As described above, the roadside shape is recognized based on the relative positions of the roadside reflector candidate objects belonging to one of the left and right groups by the roadside reflector clustering process. Roadside shapes are classified as either straight or curved.

When the roadside shape is recognized, the reliability of whether the recognized roadside shape is appropriate is evaluated.
A stationary motorcycle may be mistaken as a roadside reflector candidate object in the process described in “3.4.2 Roadside reflector identification”. In addition, the clustering of roadside reflector candidate objects may be incorrect in the process of “3.6 Roadside reflector clustering”. If there is such an error, the roadside shape will be incorrect. Therefore, the reliability of the recognized roadside shape is evaluated, and it is determined whether the roadside shape is appropriate.

3.6.1 Recognition of roadside shape

Of the two groups classified in the process of “3.6 Roadside Reflector Clustering”, the roadside reflector candidate object belongs to a group with a large number, that is, a group (right or left) with a large total number of flags Frl or Frr. The shape is recognized. Recognition of the roadside shape requires the presence of at least two or more roadside reflector candidate objects.

In a group having a large number of roadside reflector candidate objects, it is determined whether the roadside shape is recognized as a straight line or a curve. Among the roadside reflector candidate objects belonging to the selected group, the Y coordinate yk, i of the position closest to the host vehicle (relative position Qk, i (xk, i, yk, i))
Is the relative position Qk of the roadside reflector candidate object
Roadside reflector candidate at the X coordinate xk, i of i (xk, i, yk, i) and the farthest position (the maximum Y coordinate yk, i of the relative position Qk, i (xk, i, yk, i)) Object relative position Qk, j
The absolute value of the difference between (xk, j, yk, j) and the X coordinate xk, j is
When it is equal to or less than a predetermined threshold value ThL (for example, 1 [m]), that is, when | xk, i -xk, j | | Xk, i −xk, j |> ThL
In the case of, the roadside shape is recognized as a circle.

(1) When the road is straight

As shown in FIG. 28, the recognized roadside shape is the X coordinate xk, i of the relative position Qk, i (xk, i, yk, i) of the roadside reflector candidate object closest to the host vehicle.
And a straight line parallel to the Y axis (because the absolute value of the X coordinate value is small as described above). Therefore, the equation of the straight line representing the roadside shape is expressed by the following equation.

[0267]

X = xL = xk, i (33)

Instead of equation (33), the relative position Qk, i (xk, i) of the roadside reflector candidate object closest to the host vehicle
, Yk, i) and the relative position Qk, j (xk, j, yk, j) of the farthest roadside reflector candidate object may be a straight line equation representing the road shape. .

(2) When the road is curved

Generally, when a road is curved, the curvature of the curve is often represented by the radius of a circle. If the road is curved, the roadside shape is approximated as a circle. The center of this circle is on the X axis as shown in FIG.

Assuming that the center of the circle is (xLo, 0) and the radius is RL, the equation of the circle representing the roadside shape is represented by the following equation.

[0272]

(X−xLo) ² + y ² = RL ² (34)

Recognition of the roadside shape is to calculate the X coordinate xLo and the radius RL of the center of the circle. Hereinafter, a method of calculating the center X coordinate xLo and the radius RL will be described.

The relative position Qk, i (xk, i, yk, i) of the roadside reflector candidate object closest to the host vehicle and the relative position Qk, j (xk, j) of the roadside reflector candidate object farthest position , Yk, j) are on the circle represented by equation (34), so that the following equation holds.

[0275]

(Xk, i−xLo) ² + yk, i ² = RL ² (35) (xk, j−xLo) ² + yk, j ² = RL ² (36)

Eliminating RL from equations (35) and (36), the X coordinate xLo of the center of the circle is obtained.

[0277]

(Xk, i−xLo) ² + yk, i ² = (xk, j−xLo) ² + yk, j ² ∴ xLo = [xk, j ² −xk, i ² + yk, j ² −yk, i ² ] / [2 (xk, j−xk, i)] (37)

The radius RL of the circle is derived from equation (35) as follows.

[0279]

RL = sqrt [(xk, i−xLo) ² + yk, i ² ] (38)

As described above, the X coordinate xLo and the radius RL of the center of the circle are determined by the relative position Qk, i (xk, i, yk, i) and the farthest position of the roadside reflector candidate object closest to the own vehicle. Can be represented by the relative position Qk, j (xk, j, yk, j) of the roadside reflector candidate object at

Also, an equation representing the roadside shape may be obtained by least square approximation based on the relative positions of all roadside reflector candidate objects belonging to either the left or right clustered group. In addition, based on the relative positions Qk, i (xk, i, yk, i) of two arbitrarily selected roadside reflector candidate objects among the roadside reflector candidate objects belonging to either the left or right clustered group.
The roadside shape may be recognized. By using not only the roadside reflector candidate object in the current processing cycle k but also the time-series tracked roadside reflector candidate object, an accurate roadside shape can be recognized.

3.6.2 Roadside shape reliability evaluation

When the roadside shape is recognized, the reliability of the roadside shape is evaluated. It is possible to determine whether the recognized roadside shape is appropriate by this evaluation. As a result, in the process of “3.7 Evaluating the traveling lane of the preceding vehicle based on the distance from the preceding vehicle to the roadside” described later, it is possible to accurately evaluate the traveling lane.

[0284] An allowable range is set with a certain width for the recognized roadside shape, and a roadside reflector candidate object that is not used in the roadside shape recognition processing (for example, the detected object OBk closest to the host vehicle) is set within the allowable range. Roadside reflector candidate object excluding, i and the farthest detected object OBk, j)
The reliability of roadside shape recognition is evaluated based on whether or not there is an “unused roadside reflector candidate object”.

When the recognized roadside shape is a straight line, the allowable range is set in a range of ± Δx on both sides of the straight line represented by the equation x = xL representing the roadside shape. That is, the allowable range is a range from x = xL−Δx to x = xL + Δx.

[0286] If the roadside shape of the curve, the allowable range is set within a range of ± [Delta] R in the radial direction on both sides of a circle represented by a circle equation ^{(x-xLo) 2 + y} 2 = RL 2 representing the roadside shape Is done. That is, the allowable range is (x−xLo) ² +
From y ² = (RL−ΔR) ² to (x−xLo) ² + y ² =
(RL + ΔR) ² .

When the allowable range is set, it is determined whether or not the unused roadside reflector candidate object exists within the allowable range. When there is one unused roadside reflector candidate object, if it is within the allowable range, the recognized roadside shape is determined to be reliable, and the reliability is “1”.

When there are a plurality of unused roadside reflector candidate objects, among the roadside reflector candidate objects used for recognizing the roadside shape, a roadside reflector candidate object which is the closest to the host vehicle, If one unused roadside reflector candidate object located closest to the intermediate position with the farthest roadside reflector candidate object is within the allowable range, the recognized roadside shape is determined to be reliable. The reliability is “1”.

If the one unused roadside reflector candidate object described above does not exist within the allowable range but exists outside the allowable range, it is determined that there is no reliability, and the reliability is “0”. If there is no unused roadside reflector candidate object, the reliability is “0”.

FIG. 30 is a flowchart showing a roadside shape recognition process and a reliability evaluation process.

The flags are checked for all the detected objects OBk, i stored in the evaluation data storage area of the memory, the total number of flags Frr and Frl are counted, and the flag having the larger total number of flags Frr and Frl is determined. Selected (step 91).

For example, in the flags stored in the evaluation data storage area shown in FIG.
Since the number of flags Frr is 0, the flag Frl is selected. The selected flag Frl, that is, the left roadside shape is recognized.

[0293] Among the detected objects having the selected flag, the relative position Qk, i (xk, i, yk, i) of the detected object OBk, i having the minimum Y coordinate (the position closest to the host vehicle). And Y
The detected object OB having the maximum coordinates (the position farthest from the own vehicle)
The relative position Qk, j (xk, j, yk, j) of k, j is read from the evaluation data storage area (step 92).

In the example shown in FIG. 27, the roadside reflector candidate object having the minimum Y coordinate is the detected object OBk, 1, and the roadside reflector candidate object having the maximum Y coordinate is the detected object OBk, 3.
It is. The relative position Qk, 1 of these detected objects OBk, 1
(-4.7, 47.8) and relative position Q of detected object OBk, 3
k, 3 (-3.1, 54.9) is read.

The difference between the X coordinates of the relative positions of the two read roadside reflector candidate objects is compared with a threshold Th1 to determine whether the roadside shape is recognized as a straight line or a curve (step 93).

If the difference between the X coordinates is equal to or smaller than the threshold ThL, it is determined that the road is a straight line (YES in step 93), and the roadside shape is recognized as a straight line (step 95). The X coordinate xL of the straight line equation represented by the equation (33) is stored in the roadside shape data storage area (not shown).

If the difference between the X coordinates is equal to or larger than the threshold value ThL, it is determined that the curve is a curve (NO in step 93), the X coordinate xLo of the center of the circle is calculated according to the equation (37), and the equation (38) is used. The radius RL of the circle is calculated (step 94). The calculated center X coordinate xLo and radius RL are stored in the roadside shape data storage area.

In FIG. 27, the X-coordinate of the detected object OBk, 1 which is a roadside reflector candidate object is xk, 1 = -4.7, and the detected object OBk, 3 which is a roadside reflector candidate object.
Is xk, 1 = -3.1. The difference between these X coordinates is | xk, 1 −xj, j | = | −4.7 − (− 3.1) | = 1.6 (> 1)
Therefore, the roadside shape is recognized as a circle. Relative position Qk, 1 (-4.7,47.8) and relative position Qk, 3 (-3.1,54.9)
, The X coordinate xLo and the radius RL of the center of the circle are calculated according to the equations (37) and (38). That is,
xLo = [(− 4.7) ² − (− 3.1) ² +47.8 ² −54.9 ² ] / [2
{(-4,7)-(-3.1)}] = 234.0 [m], RL = sqrt [((-4,
7) −234) ² +47.8 ² ] = 243.4 [m]. These are stored in the roadside shape data storage area.

FIG. 31A is a diagram showing a road shape recognized in the processing cycle k based on the detected objects Bk, 1 and OBk, 3 which are the roadside reflector candidate objects. The center of the circle representing the recognized roadside shape (xLo,
0) is (234.0, 0) and the radius RL is 243.4.

An allowable range is set for the recognized roadside shape (step 96). For example, when an allowable range is set for the roadside shape shown in FIG. 31 (A), the result becomes as shown in FIG. 31 (B).

The relative position Qk, i (xk, i, yk, i) of the unused roadside reflector candidate object is read from the evaluation data storage area, and it is determined whether or not this is within the allowable range (step 97). .

If the unused roadside reflector candidate object falls within the set allowable range (YES in step 97), the reliability is set to "1" (step 98). If not within the allowable range (NO in step 97), the reliability is “0”.

In the example shown in FIG. 27, the relative position Qk, 2 (x
k, 2, yk, 2) are within the set allowable range as shown in FIG. 25 (C). Therefore, the reliability of the recognized roadside shape is “1”.

Thus, the roadside shape is recognized, and its reliability is evaluated.

3.7 Evaluation of Traveling Lane of Leading Vehicle Based on Distance from Leading Vehicle to Roadside (FIG. 4, Step 26)

When the reliability of the roadside shape is “1”, the lane in which the preceding vehicle is traveling is evaluated based on the distance from the own vehicle to the roadside and the distance from the preceding vehicle to the roadside. In this evaluation, all of the detected objects OBk, i (all including the roadside reflector candidate object and the vehicle candidate object) stored in the evaluation data storage area are regarded as the preceding vehicles, and the traveling lanes of these preceding vehicles are evaluated. Is performed. As described above, whether the stopped motorcycle is misidentified as a roadside reflector candidate object or the clustering of the roadside reflector candidate object is incorrect.

Using the recognized roadside shape, the distance D0 from the own vehicle to the roadside and the distance Di from the detected object OBk, i regarded as the preceding vehicle to the roadside are calculated. For all the detected objects OBk, i, the difference between the distance Di from the detected object OBk, i to the road and the distance D0 from the own vehicle to the road is defined as the evaluation value ri, and the difference is determined based on the evaluation value ri. The traveling lane of the vehicle will be evaluated.

The method of calculating the evaluation value ri differs depending on whether the roadside shape is a straight line or a circle. Hereinafter, each will be described.

(1) When the roadside shape is straight

[0310] If the recognized roadside shape is a straight line,
As shown in FIG. 32, the equation of the straight line is represented by X = xL, so that the X coordinate xL of the intersection of the straight line X = xL and the X axis
Is the distance D0 from the own vehicle to the road side. That is, D
0 = xL.

The distance Di from the detected object OBk, i to the straight line X = xL representing the roadside shape is equal to the detected object OBk, i.
X coordinate xk, i of relative position Qk, i (xk, i, yk, i) of i
And xL. That is, Di = xL
−xk, i.

Further, the evaluation value ri of the detected object OBk, i
Is represented by the following equation.

[0313]

Ri = | D0−Di | = | xL− (xL−xk, i) | = | xk, i | (39)

This is clear from FIG.

(2) When the road shape is curved

When the recognized roadside shape is a right curve (xL0 <
0), as shown in FIG. 33, the distance D0 from the host vehicle to the roadside is the X coordinate xL of the intersection of the circle representing the recognized roadside shape with the X axis. The coordinates xL of the intersection are calculated from the center (xLo, 0) of the circle and the radius RL by the following equation.

[0317]

XL = xLo-RL (40)

In equation (40), x Lo and RL are
(37) and calculated by equation (38). Since the absolute value of the X coordinate xL at this intersection is the distance D0 from the host vehicle to the road side, it is expressed by the following equation.

[0319]

D0 = xL = xLo-RL (41)

The distance D from the detected object OBk, i to the roadside
As shown in FIG. 33, i is the distance when traveling in parallel to the X axis from the relative position Qk, i (xk, i, yk, i) of the detected object OBk, i to the circle representing the roadside shape. . Object to be detected OB
The coordinates of the intersection that intersects the circle when traveling in parallel to the X coordinate from the relative position Qk, i (xk, i, yk, i) of k, i are (xL, i,
yk, i). This intersection is on the circle represented by the above equation (34). The X coordinate xL, i of the intersection is calculated by the following equation.

[0321]

(XL, i−xLo) ² + yk, i ² = RL ² ｘxL, i = xLo−sqrt [RL ² −yk, i ² ] (42)

The X coordinate xL, i of the intersection and the detected object O
Using the X coordinate xk, i of the relative position Qk, i of Bk, i, the distance Di from the detected object OBk, i to the roadside is calculated by the following equation.

[0323]

[Number 28] Di = xL, i -xk, i = xLo-sqrt [RL 2 -yk, i 2] + xk, i ... (43)

Further, the evaluation value ri of the detected object OBk, i
Is expressed by the following equation.

[0325]

Ri = | D0−Di | = | (xLo−RL) − ｛(xLo−sqrt [RL ² −yk, i ² ]) + xk, i | = | −RL + sqrt [RL ² −yk, i ² ] −xk, i |… (44)

The recognized roadside shape is a left curve (xL0>
0), similarly, the distance D0 from the vehicle to the roadside is expressed by the following equation.

[0327]

D0 = xLo + RL (45)

The distance Di from the detected object OBk, i to the road is expressed by the following equation.

[0329]

[Number 31] ^{Di = xLo + sqrt [RL 2} -yk, i 2] + xk, i ... (46)

Furthermore, the evaluation value ri of the detected object OBk, i
Is expressed by the following equation.

[0331]

Equation 32] ri = | D0 -Di | = | (xLo + RL) - {(xLo + sqrt [RL 2 -yk, i 2]) + xk, i} | = | RL -sqrt [RL 2 -yk, i 2] - xk, i |… (47)

When the evaluation value ri is calculated for the detected object OBk, i, the own lane abundance Jr, i indicating the degree of the detected object OBk, i being on the own lane based on the calculated evaluation value ri. And the other lane abundance Tr, i representing the degree of being on another lane is calculated.

The own lane existence degree Jr, i is represented by the degree of conformity of the evaluation value ri to the membership function representing the linguistic information “own lane” shown in FIG. Other lane abundance Tr, i
Is represented by the degree of conformity with the evaluation value ri for the membership function representing the linguistic information "other lanes" shown in FIG.

The own lane existence degree Jr, i and the other lane existence degree Tr, i calculated respectively for the detected object OBk, i.
Is described in “3.10 Comprehensive judgment of traveling lane of preceding vehicle”
Is used in the overall determination of the traveling lane of the preceding vehicle in the processing of.

FIG. 35 is a flowchart showing a procedure of processing for evaluating the traveling lane of the preceding vehicle performed based on the distance from the preceding vehicle to the road side.

It is determined whether the reliability of the recognized roadside shape is "1" (step 101). If the reliability is "0" (NO in step 101), the evaluation based on the distance from the preceding vehicle to the roadside is performed. The processing is terminated without being performed.

If the reliability is “1” (step 101
YES), the detected object OBk, i is regarded as the preceding vehicle, and the traveling lane is evaluated based on the distance from the detected object OBk, i to the roadside.

In the example shown in FIG. 27, since the reliability is “1”, the evaluation based on the distance to the road is performed.

The i is initialized to 1 (step 102), and the relative position Qk, 1 (xk, 1) of the first detected object OBk, 1
, Yk, 1) are read from the evaluation data storage area.

In the example shown in FIG. 27, the relative position Qk, 1 (-4.7, 47.8) of the detected object OBk, 1 is read.

The evaluation value ri is calculated for the detected object OBk, i (step 104).

In the example shown in FIG. 27, the recognized roadside shape is a road with a right curve, xLo = 234.0 [m], RL = 243.4
[m], the evaluation value ri is r1 according to equation (44).
= | 11.4-0 | = 11.4.

When the evaluation value ri is calculated, the evaluation value r
The own lane abundance Jr, i and the other lane abundance Tr, i are calculated using the membership function of FIG. 35 using i (step 105).

For example, the own lane existence degree Jr, 1 for the detected object OBk, 1 is “0”, and the other lane existence degree Tr,
1 becomes “1”.

For all the detected objects OBk, i, i
While incrementing (step 106), the evaluation value r
i and its own lane existence degree Jr, i and other lane matching Tr, i are calculated. Nk detected objects OBk, i
Means that the evaluation based on the distance to the road side has been completed. (Step 103).

FIG. 36 shows an example of the own lane existence degree Jr, i and the other lane existence degree Tr, i based on the evaluation value ri calculated for the detected object OBk, i shown in FIG.

[0347] In this manner, the traveling lane of the preceding vehicle is evaluated.

3.8 Evaluation of Traveling Lane of Leading Vehicle Based on Relative Position and Moving Vector of Leading Vehicle (FIG. 4; Step
27)

In this evaluation, the traveling lane of the preceding vehicle is determined by predicting the future route of the preceding vehicle and approaching the own vehicle based on the relative position and the movement vector of the preceding vehicle. Is what you do.

This evaluation is performed in the same manner as in the above-described “3.7 Evaluating the Traveling Lane of the Leading Vehicle Based on the Distance from the Leading Vehicle to the Roadside”, as described above for all the detected objects OBk, i stored in the evaluation data storage area. (All including the roadside reflector candidate object and the vehicle candidate object) are regarded as preceding vehicles. The magnitude of the moving vector of the preceding vehicle is small (strictly speaking, the absolute value of the Y component Vy, k, i of the moving vector Vk, i (Vx, k, i, Vy, k, i) is equal to or less than a predetermined threshold value. , For example, 5
[km / h] = 1.4 [m / s] or less), accurate evaluation results cannot be obtained, so such preceding vehicles are not evaluated.

The detected object OBk, i regarded as the preceding vehicle
Moves relatively from the relative position Qk, i (xk, i, yk, i) in the direction approaching the own vehicle, the detected object O
The traveling lane of the detected object OBk, i is determined based on the X coordinate xc, i of the point (intersection) where Bk, i is predicted to intersect the X axis. The absolute value of the X coordinate xc, i becomes the evaluation value ci for determining the traveling lane. That is, the evaluation value is c
i = | xc, i |.

A detected object OBk, i regarded as a preceding vehicle
Is different depending on whether it is on a straight road or a curved road. Hereinafter, each case will be described.

For the detected object OBk, i which is the preceding vehicle, the absolute value of the X component Vx, k, i of the movement vector Vk, i (Vx, k, i, Vy, k, i) is equal to the predetermined threshold Thv (For example,
0.1 [m / s]), that is, if | Vx, k, i | ≦ 0.1 [m / s], the detected object OBk, i is determined to be a preceding vehicle existing on a straight road. . Also, | Vx, k, i |> 0.1
If [m / s], the detected object OBk, i is determined to be a preceding vehicle existing on a curved road.

(1) When the road is straight

Since the preceding vehicle is on a straight line, it appears that the preceding vehicle is going straight ahead when viewed from the own vehicle. Therefore, the X coordinate xk, i of the relative position Qk, i (xk, i, yk, i) of the preceding vehicle (detected object OBk, i) is X
The X coordinate xc, i of the intersection with the axis. That is, xc, i =
xk, i.

In this case, the evaluation value ci is expressed by the following equation.

[0357]

(33) ci = | xc, i | = | xk, i | (48)

(2) When the road is curved

When the preceding vehicle is on a curved road, the preceding vehicle moves along a circle having a center on the X axis as viewed from the own vehicle. Therefore, X at the intersection with the X axis
The coordinates xc, i are given in “3.5.2 Roadside reflector clustering by relative position and movement vector of roadside reflector.
(2) When the road is a curve ".

If the road has a left curve (Vx, k, i> 0.1 [m / s]), the X coordinate xc, i of the intersection is represented by the following equation.

[0361]

Xc, i = xk, i + yk, i · {(Vy, k, i / Vx, k, i) −sqrt [(Vy, k, i / Vx, k, i) ² +1]} (49)

In this case, the evaluation value ci is expressed by the following equation.

[0363]

(35) ci = | xc, i | = | xk, i + yk, i · ｛(Vy, k, i / Vx, k, i) −sqrt [(Vy, k, i / Vx, k, i) ² +1]｝ |… (50)

The road has a left curve (Vx, k, i> 0.1 [m / s])
In the case of, the X coordinate xc, i of the intersection is expressed by

[0365]

Xc, i = xk, i + yk, i ｛{(Vy, k, i / Vx, k, i) + sqrt [(Vy, k, i / Vx, k, i) ² + 1]} ... ( 51)

In this case, the evaluation value ci is expressed by the following equation.

[0367]

(37) ci = | xc, i | = | xk, i + yk, i · ｛(Vy, k, i / Vx, k, i) + sqrt [(Vy, k, i / Vx, k, i) ² +1]｝ |… (52)

When the evaluation value ci is calculated for the detected object OBk, i, the own lane existence degree Jc, i indicating the degree of the detected object OBk, i being on the own lane based on the calculated evaluation value ci. And the other lane abundance Tc, i representing the degree of being on another lane is calculated.

The own lane existence degree Jc, i is represented by the degree of conformity of the evaluation value ci to the membership function representing the linguistic information "own lane" shown in FIG. Other lane abundance Tc, i
Is represented by the degree of conformity with the evaluation value ci for the membership function representing the linguistic information "other lane" shown in FIG.

The own lane abundance Jc, i and the other lane abundance Tc, i calculated for each detected object OBk, i, respectively.
Is described in “3.10 Comprehensive judgment of traveling lane of preceding vehicle”
Is used in the determination of the traveling lane of the preceding vehicle in the processing of.

FIG. 38 is a flowchart showing a procedure of processing for evaluating the traveling lane of the preceding vehicle performed based on the relative position and the movement vector of the preceding vehicle.

[0372] i is initialized to 1 (step 111), and the first detected object OBk, 1 relative position Qk, 1 (xk, 1,
yk, 1) and the movement vector Vk, 1 (Vx, k, 1, Vy, k,
1) is read from the evaluation data storage area, and the identification process is started from the detected object OBk, 1.

In the example shown in FIG. 27, the relative position Qk, 1 (−4.7, 47.8) of the detected object OBk, 1 and the movement vector Vk,
1 (-1, -21) is read.

The motion vector Vk, i of the detected object OBk, i
It is determined whether the Y component Vy, k, i of (Vx, k, i, Vy, k, i) is greater than 1.4 [m / s] (step 113).

The movement vector Vk, i (Vx, k, i, Vy, k, i
If the Y component Vy, k, i) is greater than 1.4 [m / s] (YES in step 113), an evaluation value ci is calculated for the detected object OBk, i (step 114).

In the example shown in FIG. 27, the relative position Qk, 1 of the detected object OBk, 1 is (-4.7,47.8) and the movement vector Vk, 1 (-1, -21). c1 = | xc, 1
| = | (-4.7) + (-47.8) · ｛((-21) / (-1)) + sqrt
[((-21) / (-1)) ² +1]｝ | = 5.3. Evaluation values are similarly calculated for the detected objects OBk, 2 and OBk, 3.

If the Y component Vy, k, i is equal to or less than 1.4 [m / s] (NO in step 113), the evaluation processing for the detected object OBk, i is not performed.

In the example shown in FIG. 27, the detected objects OBk, 4,
No evaluation process is performed on OBk, 5 and OBk, 6.

When the evaluation value ci is calculated, the evaluation value c
Based on i, the own lane existence degree Jc, i and the other lane existence degree Tc, i are calculated using the membership function of FIG. 37 (step 115).

For example, when the own lane abundance Jc, 1 and the other lane abundance Tc, 1 for the detected object OBk, 1 are calculated, the evaluation value c1 = 5.3.
1 becomes “0”, and the other lane existence degree Tc, 1 becomes “1”.

[0393] i is incremented (step 116).
), Returning to step 103, and repeating steps 113 to 115 for a new i.

The above processing (steps 113 to 116) is repeated while sequentially incrementing i. If i> Nk, the relative positions Qk, i (xk, i) for Nk detected objects OBk, i , Yk, i) and the movement vector Vk, i
This means that the evaluation based on (Vx, k, i, Vy, k, i) has been completed. (Step 112).

In this way, the traveling lane is evaluated based on the relative position and the movement vector of the preceding vehicle.

An example of the own lane abundance Jc, i and the other lane abundance Tc, i based on the evaluation value ci calculated for the detected object OBk, i shown in FIG. 27 is shown in FIG.

3.9 Evaluation of Traveling Lane of Leading Vehicle Only Based on Relative Position of Leading Vehicle (FIG. 4; Step 28)

Also in this evaluation, all of the detected objects OBk, i (all including roadside reflector candidate objects and vehicle candidate objects) stored in the evaluation data storage area are regarded as preceding vehicles, and their relative positions Qk, i (xk, i, yk, i)
The evaluation of the driving lane is performed based only on the driving lane.

[0387] Even if there is a curved road, there is always an area within the measurement angle range that is the own lane. In FIG. 40, the two-lane roads LR1 and LR2 of the right curve (the radius is 300 [m]; the radius is set to include a curved road and a straight road) are solid lines, Roads LL1 and LL2 with a left curve (having a radius of 300 [m]) are indicated by chain lines. When the own lane is the road LL1 or LR1, the common area between these roads and the measurement angle range is indicated by hatching.

As a basis of such a common area, a three-dimensional membership function indicating the degree (grade) of the own lane as shown in FIG. 41 is considered, and using this membership function, the driving lane of the preceding vehicle is determined. An evaluation is performed. This membership function is preset. However, the membership function can be created based on the recognized roadside shape. In FIG. 41, the grade is larger (maximum 1.0) as the vehicle is closer to the host vehicle (origin O), and the grade is smaller as the vehicle goes farther from the host vehicle and departs from the side.

Based on the relative position Qk, i of the preceding vehicle (detected object OBk, i), the degree of conformity to the three-dimensional membership function shown in FIG. 41 is calculated. This matching degree is the own lane existence degree Jp, i of the detected object OBk, i. In this evaluation, the other lane existence degree Tp, i is not calculated.

FIG. 42 is a flow chart showing a procedure of a process for evaluating the traveling lane of the preceding vehicle performed based on the relative position of the preceding vehicle.

[0391] i is initialized to 1 (step 121), and the relative position Qk, 1 (xk, 1) of the first detected object OBk, 1 is set.
, Yk, 1) are read from the evaluation data storage area,
The identification process starts from the detected object OBk, 1. For example, the relative position Qk, 1 of the detected object OBk, 1 (-4.7, 47.
8) is read.

Based on the relative position of the detected object OBk, i, a three-dimensional membership function shown in FIG.
The own lane existence degree Jp, i is calculated (step 123).

For example, for the detected object OBk, 1,
The calculated own lane existence degree Jp, 1 is “0”.

[0394] i is incremented (step 124).
), Returning to step 122, and repeating steps 123 to 124 for a new i.

The above processing (steps 123 to 124) is repeated while sequentially incrementing i. If i> Nk, the relative position Q is determined for Nk detected objects OBk, i.
This means that the process of evaluating the traveling lane of the preceding vehicle using only k, i (xk, i, yk, i) has been completed. (Step 122).

For the data shown in FIG.
When p, i is calculated, the data shown in FIG. 43 is obtained. The own lane abundance Jp, i of the detected objects OBk, 1, OBk, 2, OBk, 3 and OBk, 4 is all "0", and the own lane abundance of the detected objects OBk, 5 and OBk, 6. Are "0.1" and "0.3", respectively. Other lane abundance T
Since p and i are not calculated, "*" indicating that there is no data is shown.

[0397] In this manner, the traveling lane is evaluated based only on the relative position of the preceding vehicle.

3.10 Comprehensive judgment of traveling lane of preceding vehicle (FIG. 4; step 29)

The comprehensive judgment of the traveling lane of the preceding vehicle is described in “3.7
Evaluation of the driving lane of the preceding vehicle based on the distance from the preceding vehicle to the roadside, "3.8 Evaluation of the driving lane of the driving lane based on the relative position and the movement vector of the preceding vehicle," and "3.9
The processing is performed based on the own lane abundance and the other lane abundance calculated in the process of “Evaluation of traveling lane based on only relative position of preceding vehicle”.

[0400] For all the detected objects OBk, i regarded as preceding vehicles, the own lane abundance and other lane abundance calculated in each evaluation process are summed for each detected object OBk, i, and the own lane abundance is calculated. Is larger than the sum of the other lane abundances, it is determined that the detected object OBk, i is a preceding vehicle on the own lane. If the other lane abundance is greater than the own lane abundance, the detected object OBk, i is determined to be a preceding vehicle on the other lane.

FIG. 44 is a flowchart showing a procedure for comprehensively determining the traveling lane of the preceding vehicle.

[0402] i is initialized to 1 (step 121), and identification processing is started from the detected object OBk, 1.

The own lane existence degrees Jr, i, Jc, i and Jp, i of the i-th detected object OBk, i are read from the evaluation data storage area, and are integrated. (Step 13
3).

For example, since the detected lane abundance of the detected object OBk, 1 is Jr, 1 = 0, Jc, 1 = 0, Jp, 1 = 0, the sum is "0".

The other lane abundances Tr, i, Tc, i and Tp, i of the ith detected object OBk, i are read from the evaluation data storage area, and are integrated. (Step 13
Four ).

For example, since the other lane abundance of the detected object OBk, 1 is Tr, 1 = 1, Tc, 1 = 1, Tp, 1 = * (0), the total sum is “2”.

[0407] The total sum of the own lane abundances is compared with the total sum of the other lane abundances (step 135). If the total sum of the own lane abundances is larger than the total sum of the other lane abundances (YE at step 135)
S), the own lane flag is set on the detected object OBk, i (step 136).

If the sum of the own lane abundances is smaller than the total sum of the other lane abundances (NO in step 135), the own lane flag of the detected object OBk, i remains "*".

For example, the sum of the vehicle lane abundances of the detected object OBk, 1 is “0”, and the total lane abundance is “2”.
Therefore, the detected object OBk, 1 is not the own lane.

[0410] i is incremented (step 137).
), Returning to step 132, and repeating steps 133 to 137 for new i.

The above processing (steps 133 to 137) is repeated while sequentially incrementing i. If i> Nk, the evaluation of Nk detected objects OBk, i based on the distance to the road side has been completed. Become. (Step 132
).

The sum of the own lane existence degree and the sum of the other lane existence degrees are calculated for the data shown in FIG. 36, FIG. 39 and FIG. Can be Objects to be detected OBk, 1, OBk, 2, OB
k, 3, Bk, 4 and OBk, 5 are the preceding vehicles in the other lane. The detected object OBk, 6 is a preceding vehicle in the own lane.

4.11 Output of Inter-Vehicle Distance (FIG. 4; Step 3)
0)

The relative position Qk, i (xk, i, yk, i) of the detected object OBk, i determined to be the preceding vehicle on the own lane
The smallest one of the Y coordinates yk, i is output as the following distance.

In the example shown in FIG. 45, the only preceding vehicle on the own lane is the detected object OBk, 6.
The Y coordinate "35.9" of the relative position Qk, 6 (xk, 6, yk, 6) of Bk, 6 is output as the inter-vehicle distance to the preceding vehicle. This detected object OBk, 6 represents the vehicle 2 shown in FIG.

[0416] As described above, the inter-vehicle distance to the preceding vehicle is measured in the signal processing device 19.

It is needless to say that it may be determined whether or not the preceding vehicle is on the own lane based on one or two evaluation results of the traveling lanes of the above three types of preceding vehicles. Absent. If the own lane abundance is equal to or greater than a predetermined threshold, it may be determined that the preceding vehicle exists on the own lane.

FIG. 46 shows a state in which the inter-vehicle distance measuring device 10 is mounted on the vehicle 1. This inter-vehicle distance measuring device 10 is attached to a front portion of the vehicle 1 and is fixed to a frame (not shown) supporting the bumper 1A provided inside the bumper 1A.

[Brief description of the drawings]

FIG. 1 shows a state in which a vehicle is traveling on a road.

FIG. 2 is a block diagram showing an electrical configuration of the inter-vehicle distance measuring device.

FIG. 3 is a flowchart showing a processing procedure of one processing cycle of inter-vehicle distance measurement in the signal processing device.

FIG. 4 is a flowchart showing a processing procedure of one processing cycle of inter-vehicle distance measurement in the signal processing device.

FIG. 5 shows an example of a pair data storage area of a memory storing pair data representing a sweep angle and a distance.

FIG. 6 (A) shows a polar coordinate system indicating a detection position of a detected object, and FIG. 6 (B) shows a rectangular coordinate system when the position display of the detected object shown in FIG. .

FIG. 7 is a flowchart illustrating a procedure of a coordinate conversion process.

FIG. 8 shows an example of a relative position data temporary storage area that stores coordinate-converted relative position data.

FIG. 9 shows an example of a window used in the identity identification processing.

FIG. 10 is a flowchart illustrating a processing procedure of identity identification processing.

FIG. 11 is a flowchart illustrating a processing procedure of identity identification processing.

FIG. 12 illustrates an example of a relative position data storage area that stores relative position data obtained by the identity identification processing.

FIG. 13 illustrates a first example used in the movement vector calculation process.
An example of the window W1 is shown.

FIG. 14 shows a second example used in the movement vector calculation process.
An example of the window W2 is shown.

FIG. 15 illustrates an example of a movement vector.

FIG. 16 illustrates an example of a registered data storage area in a previous processing cycle.

FIG. 17 is a flowchart illustrating a procedure of a movement vector calculation process.

FIG. 18 illustrates an example of a registered data storage area in the current processing cycle.

FIG. 19 is a flowchart illustrating a process of identifying a vehicle candidate object.

FIG. 20 illustrates an example of an evaluation data storage area after the vehicle candidate object identification processing has been performed.

FIG. 21 is a flowchart showing a roadside reflector identification process.

FIG. 22 illustrates an example of an evaluation data storage area after the roadside reflector identification processing has been performed.

FIG. 23 illustrates a relative direction and a moving direction based on a movement vector of a roadside reflector candidate object.

FIG. 24 illustrates a relative movement of a roadside reflector candidate object based on a relative position and a movement vector when the road is a straight line.

FIG. 25 illustrates a relative movement based on a relative position and a movement vector of a roadside reflector candidate object when the road is a curve.

FIG. 26 is a flowchart illustrating a processing procedure of clustering processing of a roadside reflector.

FIG. 27 illustrates an example of an evaluation data storage area after the clustering process of the roadside reflector is performed.

FIG. 28 illustrates a recognized roadside shape when the road is a straight line.

FIG. 29 illustrates a recognized roadside shape when the road is a curve.

FIG. 30 is a flowchart illustrating a procedure of a roadside shape recognition process and a reliability determination process.

FIG. 31A shows a roadside reflector candidate object used for roadside shape recognition and a recognized roadside shape, FIG. 31B shows an allowable range set for the recognized roadside shape, C)
Indicates that a roadside reflector candidate object not used for recognition of the roadside shape exists within the allowable range.

FIG. 32 illustrates a positional relationship of a preceding vehicle when the road is straight.

FIG. 33 illustrates a positional relationship between a preceding vehicle and a curved road.

FIG. 34 is a graph illustrating an example of a membership function representing linguistic information “own lane” and “other lane” used for evaluating the traveling lane of the preceding vehicle based on the distance from the preceding vehicle to the roadside.

FIG. 35 is a flowchart illustrating a procedure of a process of evaluating a traveling lane of a preceding vehicle based on a distance from a preceding vehicle to a road side.

36 shows an example of the own lane abundance and the other lane abundance with respect to the evaluation value calculated based on the distance from the preceding vehicle to the roadside, and is calculated based on the data in the evaluation data storage area shown in FIG. 27. It is a thing. .

FIG. 37 is a graph showing an example of a membership function representing linguistic information “own lane” and “other lane” used for evaluating a traveling lane based on a relative position and a movement vector of a preceding vehicle.

FIG. 38 is a flowchart illustrating a procedure of processing for evaluating a traveling lane of a preceding vehicle based on a relative position and a movement vector of the preceding vehicle.

FIG. 39 shows an example of the own lane abundance and the other lane abundance with respect to the evaluation value based on the distance from the preceding vehicle to the road side.

FIG. 40 shows a measurement angle range of a right curve and a left curve on a road.

FIG. 41 shows an example of a three-dimensional membership function serving as the own lane, which is used for evaluating a traveling lane based only on the relative position of the preceding vehicle.

FIG. 42 is a flowchart illustrating a procedure of processing for evaluating a traveling lane of a preceding vehicle based only on a relative position of the preceding vehicle.

FIG. 43 shows own lane abundance and other lane abundance calculated based only on the relative position of the preceding vehicle.

FIG. 44 is a flowchart illustrating a procedure for comprehensively determining a traveling lane of a preceding vehicle.

FIG. 45 illustrates an example of the total sum of the own lane abundance and the other lane abundance obtained by each evaluation.

FIG. 46 illustrates a state in which the inter-vehicle distance measurement device is mounted on a vehicle.

[Explanation of symbols]

 10 Inter-vehicle distance measuring device 11 Emitter 12 Laser beam sweeping device 13 Sweep angle detecting device 14 Receiver 15 Distance calculating device 16 Vehicle speed sensor 17 Alarm device 18 Actuator 19 Signal processing device

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-5-113481 (JP, A) JP-A-61-23985 (JP, A) JP-A-5-34460 (JP, A) 205198 (JP, A) JP-A-5-342500 (JP, A) JP-A-5-288847 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01S 7/48-7 / 51 G01S 17/00-17/95 G01S 13/00-13/95

Claims (16)

(57) [Claims] 1. A laser beam is projected at a predetermined time interval at a predetermined angle interval in a predetermined measurement angle range, a projection timing signal indicating a projection time point of the laser beam, and a sweep angle signal indicating a projection angle. Light receiving means for receiving the reflected light of the laser light emitted from the light emitting means from the object to be detected, and outputting a light receiving timing signal indicating a point in time when the reflected light is received; Based on the time interval between the light emitting timing signal from the light means and the light receiving timing signal from the light receiving means,
Distance calculating means for calculating a distance to the detected object; and a relative position of the detected object based on the distance to the detected object calculated by the distance calculating means and a sweep angle signal obtained from the light projecting means. Is calculated, and the movement vector of the detected object is calculated based on the two relative positions at different times calculated for the same detected object, and the detected object is calculated based on the movement vector of the detected object. Identifies whether it is a roadside reflector provided on the right side of the road, a roadside reflector provided on the left side of the road, or a vehicle, and either the left or right one of the detected objects identified as the roadside reflector A roadside shape is recognized based on a relative position of at least two detected objects on one side, and the detected object identified as a vehicle based on the recognized roadside shape is driven by the vehicle. And to determine the lane that, the signal processing means the distance to the vehicle traveling the same lane to output the smallest of the inter-vehicle distance, the inter-vehicle distance measuring apparatus having a. 2. A laser beam is projected at a constant time interval at a constant angular interval over a predetermined measurement angle range.
A light projecting means for outputting a light projecting timing signal representing a time point at which the laser light is projected and a sweep angle signal representing the light projecting angle; A light receiving means for receiving light and outputting a light receiving timing signal indicating a point in time when the reflected light is received, based on a time interval between the light emitting timing signal given from the light emitting means and the light receiving timing signal given from the light receiving means, Distance calculating means for calculating the distance to the detected object; vehicle speed detecting means for detecting the vehicle speed of the vehicle; the distance to the detected object calculated by the distance calculating means; and a sweep angle signal given from the light projecting means Relative position calculating means for calculating the relative position of the detected object based on the above two different time points calculated by the relative position calculating means for the same detected object. Moving vector calculating means for calculating a moving vector of the detected object based on the paired position; for each detected object, a moving vector calculated by the moving vector calculating means and a vehicle speed detected by the vehicle speed detecting means. A roadside shape recognition means for detecting a roadside reflector provided on the roadside of the road and recognizing a roadside shape based on the detected relative position of the roadside reflector; each detected object is represented by the above roadside shape from the own vehicle. Lane determining means for determining whether the detected object is a preceding vehicle on the own lane based on the distance to the road side and the distance from the detected object to the road side represented by the roadside shape; The distance from the detected object whose relative position is the smallest among the detected objects determined to be the preceding vehicle on the own lane by the determining means. Vehicle distance output means for outputting as between distance, inter-vehicle distance measuring apparatus having a. 3. A laser beam is projected at a constant time interval at a constant angular interval over a predetermined measurement angle range.
A light projecting means for outputting a light projecting timing signal representing a time point at which the laser light is projected and a sweep angle signal representing the light projecting angle; A light receiving means for receiving light and outputting a light receiving timing signal indicating a point in time when the reflected light is received, based on a time interval between the light emitting timing signal given from the light emitting means and the light receiving timing signal given from the light receiving means, Distance calculating means for calculating the distance to the detected object; vehicle speed detecting means for detecting the vehicle speed of the vehicle; the distance to the detected object calculated by the distance calculating means; and a sweep angle signal given from the light projecting means Relative position calculating means for calculating the relative position of the detected object based on the above two different time points calculated by the relative position calculating means for the same detected object. Moving vector calculating means for calculating a moving vector of the detected object based on the paired position; for each detected object, a moving vector calculated by the moving vector calculating means and a vehicle speed detected by the vehicle speed detecting means. A roadside shape recognition means for detecting a roadside reflector provided on the roadside of the road and recognizing a roadside shape based on the detected relative position of the roadside reflector; each detected object is represented by the above roadside shape from the own vehicle. A first vehicle lane presence degree and another vehicle indicating the degree of presence of the detected object on the own lane based on the distance to the road side and the distance from the detected object to the road side represented by the road side shape. First evaluation means for calculating a first other lane presence degree indicating the degree of presence on the line, and a small amount calculated by the first evaluation means First
Lane determining means for determining whether or not the detected object is a preceding vehicle on the own lane based on the own lane abundance of the detected lane; and a detected vehicle determined to be a preceding vehicle on the own lane by the lane determining means. An inter-vehicle distance measuring device, comprising: inter-vehicle distance output means for outputting the distance to an object having the smallest relative position as an inter-vehicle distance. 4. A laser beam is projected at a constant time interval at a constant angular interval over a predetermined measurement angle range.
A light projecting means for outputting a light projecting timing signal representing a time point at which the laser light is projected and a sweep angle signal representing the light projecting angle; A light receiving means for receiving light and outputting a light receiving timing signal indicating a point in time when the reflected light is received, based on a time interval between the light emitting timing signal given from the light emitting means and the light receiving timing signal given from the light receiving means, Distance calculating means for calculating the distance to the detected object; vehicle speed detecting means for detecting the vehicle speed of the vehicle; the distance to the detected object calculated by the distance calculating means; and a sweep angle signal given from the light projecting means Relative position calculating means for calculating the relative position of the detected object based on the above two different time points calculated by the relative position calculating means for the same detected object. A moving vector calculating means for calculating a moving vector of the detected object based on the paired position; for each detected object, the detected object is present on the own lane based on a relative position and the moving vector of the detected object; Second evaluation means for calculating a second own lane existence degree representing the degree and a second other lane existence degree representing the degree existing on the other lane; at least a second evaluation means calculated by the second evaluation means
Lane determining means for determining whether or not the detected object is a preceding vehicle on the own lane based on the own lane abundance of the detected lane; and a detected vehicle determined to be a preceding vehicle on the own lane by the lane determining means. An inter-vehicle distance measuring device, comprising: inter-vehicle distance output means for outputting the distance to an object having the smallest relative position as an inter-vehicle distance. 5. A laser beam is projected at a constant time interval at a constant angular interval over a predetermined measurement angle range.
A light projecting means for outputting a light projecting timing signal representing a time point at which the laser light is projected and a sweep angle signal representing the light projecting angle; A light receiving means for receiving light and outputting a light receiving timing signal indicating a point in time when the reflected light is received, based on a time interval between the light emitting timing signal given from the light emitting means and the light receiving timing signal given from the light receiving means, Distance calculating means for calculating the distance to the detected object; vehicle speed detecting means for detecting the vehicle speed of the vehicle; the distance to the detected object calculated by the distance calculating means; and a sweep angle signal given from the light projecting means Relative position calculating means for calculating a relative position of the detected object based on the detected object; and for each detected object, the detected object is present on the own lane based on the relative position of the detected object. Third third evaluation means for calculating a third other lanes abundance indicating a degree existing in the own lane abundance and other lanes on, at least a third calculated by said third evaluation means for indicating a degree of
Lane determining means for determining whether or not the detected object is a preceding vehicle on the own lane based on the own lane abundance of the detected lane; and a detected vehicle determined to be a preceding vehicle on the own lane by the lane determining means. An inter-vehicle distance measuring device, comprising: inter-vehicle distance output means for outputting the distance to an object having the smallest relative position as an inter-vehicle distance. 6. A laser beam is emitted at a constant time interval at a constant angular interval over a predetermined measurement angle range.
A light projecting means for outputting a light projecting timing signal representing a time point at which the laser light is projected and a sweep angle signal representing the light projecting angle; A light receiving means for receiving light and outputting a light receiving timing signal indicating a point in time when the reflected light is received, based on a time interval between the light emitting timing signal given from the light emitting means and the light receiving timing signal given from the light receiving means, Distance calculating means for calculating the distance to the detected object; vehicle speed detecting means for detecting the vehicle speed of the vehicle; the distance to the detected object calculated by the distance calculating means; and a sweep angle signal given from the light projecting means Relative position calculating means for calculating the relative position of the detected object based on the above two different time points calculated by the relative position calculating means for the same detected object. Moving vector calculating means for calculating a moving vector of the detected object based on the paired position; for each detected object, a moving vector calculated by the moving vector calculating means and a vehicle speed detected by the vehicle speed detecting means. A roadside shape recognition means for detecting a roadside reflector provided on the roadside of the road and recognizing a roadside shape based on the detected relative position of the roadside reflector; each detected object is represented by the above roadside shape from the own vehicle. A first vehicle lane presence degree and another vehicle indicating the degree of presence of the detected object on the own lane based on the distance to the road side and the distance from the detected object to the road side represented by the road side shape. First evaluation means for calculating a first other lane abundance representing the degree of presence on a line; relative position of the detected object with respect to each detected object A second self-lane abundance indicating the degree of the detected object existing on the own lane and a second other lane abundance indicating the degree of existence on the other lane based on the movement vector. Means for evaluating each of the detected objects, based on the relative positions of the detected objects, a third self-lane abundance degree indicating the degree of existence of the detected object on the own lane and a degree of existence on another lane. Third evaluation means for calculating a third other lane abundance representing the first, second, and third own lane abundances, and first, second, and third other lane abundances for each detected object The lane determining means for determining that the detected object having the total sum of the lane abundances greater than the total lane abundance is the preceding vehicle on the own lane, and the lane determining means. Determined to be the preceding vehicle on the lane Have been among the object to be detected, the inter-vehicle distance measuring apparatus having inter-vehicle distance output means, the relative position to output a distance between the smallest as inter-vehicle distance. 7. A laser beam is projected at a constant time interval at a constant angular interval over a predetermined measurement angle range.
A light projecting means for outputting a light projecting timing signal representing a time point at which the laser light is projected and a sweep angle signal representing the light projecting angle; A light receiving means for receiving light and outputting a light receiving timing signal indicating a point in time when the reflected light is received, based on a time interval between the light emitting timing signal given from the light emitting means and the light receiving timing signal given from the light receiving means, Distance calculating means for calculating the distance to the detected object; vehicle speed detecting means for detecting the vehicle speed of the vehicle; the distance to the detected object calculated by the distance calculating means; and a sweep angle signal given from the light projecting means Relative position calculating means for calculating a relative position of the detected object based on the relative position calculated by the relative position calculating means. The lane determining means for determining whether the vehicle is a preceding vehicle on the lane, and the distance between the detected object whose relative position is smallest among the detected objects determined to be the preceding vehicle on the own lane by the lane determining means. An inter-vehicle distance output means for outputting as an inter-vehicle distance, wherein the relative position calculating means checks whether or not the detected objects whose relative positions have been calculated depend on the reflected light from the same detected object; Are performed, and the detected objects determined to be the same by the identity identification are collectively referred to as 1
An inter-vehicle distance measuring device for re-calculating relative positions of one detected object as a plurality of detected objects. 8. A laser beam is emitted at a constant time interval at a constant angular interval over a predetermined measurement angle range.
A light projecting means for outputting a light projecting timing signal representing a time point at which the laser light is projected and a sweep angle signal representing the light projecting angle; A light receiving means for receiving light and outputting a light receiving timing signal indicating a point in time when the reflected light is received, based on a time interval between the light emitting timing signal given from the light emitting means and the light receiving timing signal given from the light receiving means, Distance calculating means for calculating the distance to the detected object; vehicle speed detecting means for detecting the vehicle speed of the vehicle; the distance to the detected object calculated by the distance calculating means; and a sweep angle signal given from the light projecting means Relative position calculating means for calculating the relative position of the detected object based on the above two different time points calculated by the relative position calculating means for the same detected object. A moving vector calculating means for calculating a moving vector of the detected object based on the paired position; for each detected object, a relation between a relative position and a moving vector of one detected object and another detected object is a vehicle; It is checked whether or not the pair condition is satisfied. If there are two detected objects that satisfy the pair condition, the vehicle identification means that combines them into one detected object is calculated by at least the relative position calculating means. Lane determining means for determining whether or not the detected object combined by the vehicle identifying means is a preceding vehicle on the own lane based on the relative position, and a preceding lane on the own lane by the lane determining means. An inter-vehicle distance measuring device, comprising: inter-vehicle distance output means for outputting, as an inter-vehicle distance, a distance between a detected object determined to be a vehicle and an object having a smallest relative position. Measuring device. 9. The road-side shape recognizing means is provided on a road side of a road based on a moving vector calculated by the moving vector calculating means and a vehicle speed detected by the vehicle speed detecting means for each detected object. The roadside reflectors are detected, and the detected roadside reflectors are divided into a group to which the right side of the road belongs and a group to which the left side of the road belongs. 7. The inter-vehicle distance measuring apparatus according to claim 3, wherein the apparatus recognizes a roadside shape. 10. The roadside shape recognizing means includes a group to which a roadside reflector is provided on the right side of the road and a group to which the roadside reflector is provided on the left side of the road, according to a relative position of the roadside reflector and a moving direction based on a moving vector. The inter-vehicle distance measuring device according to claim 9, which is divided into groups. 11. The road-side shape recognizing means includes a group to which the road-side reflector is provided on the right side of the road and a group to which the road-side reflector is provided on the left side of the road by a relative movement based on a relative position and a movement vector of the road-side reflector. The inter-vehicle distance measuring apparatus according to claim 9, wherein the apparatus is divided into a group to which the object belongs. 12. The inter-vehicle distance measuring device according to claim 3, wherein the roadside shape recognition means recognizes the roadside shape as a straight line or a circle. 13. The road-side shape recognizing means further performs reliability evaluation as to whether the recognized road-side shape is appropriate, and the first evaluating means is recognized by the road-side shape recognizing means. 7. The inter-vehicle distance measuring device according to claim 6, wherein the self-lane abundance and the other lane abundance are not calculated for all the detected objects when the roadside shape is not reliable. 14. The method according to claim 1, wherein when the movement vector of the detected object is equal to or less than a predetermined threshold, the second evaluating means determines the second own lane abundance and the second other lane abundance for the detected object. 7. The inter-vehicle distance measuring device according to claim 6, wherein the calculation is not performed. 15. The third evaluation means increases the third vehicle lane abundance of a detected object located in a short distance in front of the vehicle, and increases as the distance from the vehicle increases and the distance to the side increases. 6. The method according to claim 5, wherein the third own lane abundance is reduced.
Or the inter-vehicle distance measuring device according to 6. 16. A vehicle equipped with the inter-vehicle distance measuring device according to claim 1.