JP2023015621A - Vehicle position specification device, vehicle position specification method, light distribution control device, and vehicular lighting fixture system - Google Patents

Abstract

To improve the accuracy in tracking using a time series of positions of a preceding vehicle.SOLUTION: In a vehicle position specification device comprising a controller, the controller obtains a duration period of a state in which a mutual change amount between successive positional data in time series is equal to a prescribed reference value or less, for each data group, with regard to a plurality of data groups respectively containing positional data in time series related to a preceding vehicle. Further, when a plurality of new positional data related to the preceding vehicle are generated, the controller obtains the strength of correlation between the latest positional data in time series of each data group and the plurality of new positional data, in order to add the plurality of new positional data respectively to any of the respective data groups, on the basis of the correlation strength, in order starting with data groups with relatively long duration periods.SELECTED DRAWING: Figure 6

Description

The present disclosure relates to a vehicle position specifying device, a vehicle position specifying method, a light distribution control device, and a vehicle lamp system.

Light distribution control technology that illuminates high beams by setting a dimming range (or light blocking range) within the high beam irradiation range according to the position of the vehicle in front of the vehicle (preceding vehicle, oncoming vehicle, etc.) is known (see Patent Document 1, for example). The position of the forward vehicle can be obtained, for example, by performing image recognition processing on an image of the space in front of the vehicle captured by a camera and detecting light spots (positions of headlights and taillights).

By the way, when tracking the position of the forward vehicle in chronological order, did the originally existing forward vehicle disappear due to noise data and was replaced by a new forward vehicle that did not originally exist? Misplacement of data may occur like this.

JP 2015-58802 A

One of the purposes of the specific aspects of the present disclosure is to improve the accuracy when tracking the position of the preceding vehicle in chronological order.

[1] A vehicle position specifying device according to one aspect of the present disclosure is a vehicle position specifying device including a controller, wherein the controller includes (a) a plurality of data groups each including time-series position data relating to a forward vehicle; for each data group, the duration of a state in which the amount of change between the position data that is chronologically back and forth is equal to or less than a predetermined reference value; (b) a plurality of new position data regarding the preceding vehicle occurs, the strength of correlation between the most recent position data in the time series of each data group and the plurality of new position data is obtained, and (c) the data group having a relatively long duration and adding each of the plurality of new position data to any one of the data groups based on the strength of the correlation.
[2] A vehicle position identification method of one aspect according to the present disclosure is a vehicle position identification method executed by a controller, the controller comprising: (a) a plurality of vehicle positions, each of which includes time-series position data relating to a forward vehicle; For each data group, determining the duration of a state in which the amount of change between the position data that is chronologically back and forth is equal to or less than a predetermined reference value; (c) determining the strength of correlation between the most recent position data in the time series of each data group and the plurality of new position data when such position data is generated; adding each of the plurality of new position data to any one of the data groups based on the strength of the correlation in order from the longer data group.
[3] A light distribution control device according to one aspect of the present disclosure controls a light distribution pattern of a vehicle lamp using each of the data groups obtained by the vehicle position specifying device of [1]. is.
[4] A vehicle lighting system according to one aspect of the present disclosure is a vehicle lighting system including the light distribution control device of [3] and a vehicle lighting device connected to the light distribution control device.

According to the above configuration, it is possible to improve the accuracy when tracking the position of the forward vehicle in chronological order.

FIG. 1(A) is a diagram for explaining a state of erroneous detection of a forward vehicle assumed in this embodiment. FIG. 1(B) is a diagram showing changes in the position of each object in the situation assumed in FIG. 1(A). FIG. 2 is a diagram for explaining the Euclidean distance. FIG. 3 is a diagram showing the configuration of a vehicle lamp system according to one embodiment. FIG. 4 is a diagram showing a configuration example of a computer that implements the controller. FIG. 5 is a diagram for explaining the survival time of position data. FIG. 6A is a diagram showing an example of transition of position data. FIG. 6B is a diagram for explaining the tracking processing of this embodiment. FIG. 6C is a diagram for explaining the tracking process of the comparative example. FIG. 7 is a flow chart showing the operation procedure of the controller of the vehicle lamp system. FIG. 8 is a diagram schematically showing an example of light distribution control using position data corresponding to each object.

FIG. 1(A) is a diagram for explaining a state of erroneous detection of a forward vehicle assumed in this embodiment. For example, at time (t-3), there is one forward vehicle and its position (right angle, left angle) is detected. The right angle and left angle are, for example, the positions of taillights arranged on the left and right sides of the forward vehicle. In this embodiment, it is assumed that there are a plurality of vehicles in front to be detected, and a data group including chronological position data corresponding to each object is managed by labeling them, for example, as object A and object B. . In the illustrated example, object A has not been detected and the position of object B has been detected. The position data of the object is specified by a relative angle with respect to the own vehicle. In the illustrated example, the position data of object B has a left angle of −1.2° and a right angle of +0.8°.

At the next time (t-2), if an erroneous detection occurs even though there is only one forward vehicle, their positions are detected as two forward vehicles. At this time, each position data is obtained in association with each of the objects A and B. FIG. Since the erroneous detection is for only one vehicle in front, the position data of object A and the position data of object B are very close to each other. In the illustrated example, the position data for object A is specified as a left angle of −0.9° and a right angle of +1.0°, and the position data for object B is specified as a left angle of −1.2° and a right angle of −1.2°. is specified as +0.8°. Assume that a similar erroneous detection also occurs at the next time (t-1). In the example shown, the position of object A is specified as a left angle of −0.9° and a right angle of +0.9°, and the position of object B is specified as a left angle of −1.2° and a right angle of +0°. .8°.

At the next time (t), the erroneous detection is resolved, object A is not detected, and the position of object B is detected. In the illustrated example, the position data of object B specifies a right angle of −1.2° and a left angle of +0.8°.

FIG. 1(B) is a diagram showing changes in the position of each object in the situation assumed in FIG. 1(A). In order to make the explanation easier to understand, the transition of the right angle is shown here, but the same applies to the left angle. The data group of the right angle of object B continues to be included in the normal detection range and changes gradually. On the other hand, the data group of the right angle of object A, which should not exist originally, was outside the normal detection range (no detection), but suddenly entered the normal detection range at time (t-2). It is included in the normal detection range until the next time (t-1).

Here, even if position data corresponding to object A suddenly appears as shown in the figure, the position data of objects A and B can be exchanged if the data group of position data relating to each object is managed while maintaining continuity. no. However, for example, if the continuity (strength of correlation) of the position data of each object between one time and the next time is determined based only on the Euclidean distance, the positions associated with each object A and B Data may be replaced. If such a change occurs, for example, a problem arises in a system that controls a light distribution pattern in accordance with a change in the position of an object.

Note that the Euclidean distance in the present embodiment means, as shown in FIG. It is represented by a distance D between points and is an index representing the strength of correlation between position data. In the example shown in FIG. 1B, the Euclidean distance is obtained between two position data at time (t-2) and two position data at the next time (t-1), and the position where the Euclidean distance is the minimum If the data are determined to be continuous, the position data may be replaced as described above.

Hereinafter, an embodiment will be described in which it is possible to determine the strength of correlation only by the Euclidean distance between position data and prevent the replacement of position data that may occur when the position data is added to the data group of each object.

FIG. 3 is a diagram showing the configuration of a vehicle lamp system according to one embodiment. The vehicle lighting system shown in the figure irradiates light to the space around (here, in front of) the own vehicle. A range is set, and selective light irradiation is performed with the range other than the range set as the light irradiation range. Such light distribution control is also referred to as ADB (Adaptive Driving Beam) control, and contributes to improved visibility around the vehicle.

The illustrated vehicle lamp system includes an imaging device 10, a controller (control device) 20, and a pair of headlight units (vehicle lamps) 30L and 30R. The imaging device 10 and the controller 20 are connected by a predetermined communication means. Further, the controller 20 and the headlamp units 30L and 30R are connected via predetermined wiring. In this embodiment, the imaging device 10 and the controller 20 correspond to the "vehicle position specifying device" and the "light distribution control device".

The imaging device 10 includes a camera 11 and an image processor 12. The image processor 12 performs predetermined image processing based on an image captured by the camera 11, thereby detecting a forward vehicle existing in front of the own vehicle. position is detected. In this embodiment, the imaging device 10 detects the left side position and the right side position of each forward vehicle as the position of each forward vehicle. The left side position and the right side position here correspond to the position of the taillight when the vehicle in front is the preceding vehicle, and to the position of the headlight when the vehicle in front is the oncoming vehicle. The position data of the forward vehicle is represented by an angle relative to the own vehicle position. Further, when there are a plurality of vehicles ahead, the position data (or signal) is output to the controller 20 with labeling data attached to each vehicle ahead.

The controller 20 specifies the position of the forward vehicle based on the position data of each forward vehicle output from the imaging device 10, and determines the dimming range and the light irradiation range corresponding to each forward vehicle according to the specified position. A light distribution pattern is set, and the headlight units 30L and 30R are controlled so that light irradiation is performed according to this light distribution pattern. The controller 20 is realized by using a computer having, for example, a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), etc., and causing the computer to execute a predetermined operation program. In order to facilitate understanding of the functions realized by the controller 20, functional blocks will be used for explanation. The controller 20 has a vehicle position acquisition unit 21, a survival time calculation unit 22, a tracking processing unit 23, a light distribution pattern setting unit 24, and a control signal generation unit 25 as functional blocks. The survival time calculation unit 22 corresponds to the "duration calculation unit", and the tracking processing unit 23 corresponds to the "correlation calculation unit" and the "data addition unit".

The vehicle position acquisition unit 21 acquires the position data (or signal) of the preceding vehicle output from the imaging device 10 . The position data of each forward vehicle is obtained in chronological order every predetermined time (every 50 milliseconds, for example). Further, the position data is added with data on the type of vehicle ahead (preceding vehicle, oncoming vehicle, etc.).

The survival time calculation unit 22 determines that, for a plurality of data groups each including position data relating to a forward vehicle obtained in time series, the amount of change between position data that precedes and follows in time series is equal to or less than a predetermined reference value. Survival time, which is continuous time, is calculated. The details of the survival time calculation method will be described later.

The tracking processing unit 23 obtains the strength of the correlation between each data group and the most recent position data in the time series when a plurality of position data regarding the preceding vehicle is newly obtained, and calculates the relative survival time. Based on the strength of this correlation, processing (tracking processing) of adding each new position data to any data group is performed in order from the data group with the longest .

The light distribution pattern setting unit 24 sets a predetermined range including the position of each vehicle in front of the high beam irradiable range as a dimming range and the other range as a light irradiation range, corresponding to the position of the vehicle ahead. Set the light pattern.

The control signal generator 25 generates a control signal corresponding to the light distribution pattern set by the light distribution pattern controller 24, and supplies it to each of the headlight units 30L and 30R.

Each of the headlight units 30L and 30R operates according to a control signal supplied from the controller 20 to irradiate the front of the vehicle with light, and includes an ADB unit 31 and a low beam unit 32, respectively.

The ADB unit 31 is a lamp unit capable of selectively irradiating light according to the position of the forward vehicle in the high beam irradiation range. Examples of such an ADB unit 31 include one that forms irradiation light by combining a light source and a liquid crystal element, one that forms irradiation light by individually turning on and off a plurality of light emitting elements (LEDs), and one that forms irradiation light by individually turning on and off a plurality of light emitting elements (LEDs). Various known units such as those that form irradiation light by scanning the emitted light with a MEMS device or the like, and those that form irradiation light by partially shielding the light from the light source using a shielding plate (shade) can be used.

The low beam unit 32 is a lamp unit capable of illuminating a region relatively closer to the vehicle than the high beam. Various known units can be used as such a low beam unit 32 . The low beam may be formed by one unit having both the function of the low beam unit 32 and the function of the ADB unit 31 .

FIG. 4 is a diagram showing a configuration example of a computer that implements the controller. The illustrated computer includes a CPU 201, a ROM 202, a RAM 203, a storage device 204, and an external interface (I/F) 205 which are communicatively connected. The CPU 201 operates based on a basic control program read from the ROM 202, reads a program (application program) 206 stored in the storage device 204, and executes the program to implement the functions of the controller 20 described above. A RAM 203 temporarily stores data to be used when the CPU 201 operates. The storage device 204 is a nonvolatile storage device such as a hard disk or solid state drive, and stores various data such as the program 206 . An external interface 205 is an interface that connects the CPU 201 and an external device, and is used to connect the imaging device 10 and the CPU 201, for example.

FIG. 5 is a diagram for explaining the survival time of position data. As shown in FIG. 1B, the position data of the data group corresponding to object A due to erroneous detection immediately after appearing suddenly becomes a value outside the detection range (for example, a certain upper limit value). The duration of detection of data is short. On the other hand, since the position data of the data group corresponding to object B is continuously detected, the detection of the position data continues for a long time. Considering such characteristics, in the present embodiment, a data group of position data corresponding to each object is managed in chronological order, and the survival time, which is the time during which the detection continues, is obtained. Although the transition of the right angle is shown in the upper part of the drawing, the transition of the left angle is the same. The lower part of the figure shows the change in survival time.

As shown in the illustrated example, the position data (here, the right angle) of the data group corresponding to the object does not fluctuate significantly over time and stays within a certain threshold range (below the reference value). In some cases, the survival time is gradually increased. A long survival time means that the position data of the data group has high continuity and high reliability (accuracy). This survival time is longer in a data group containing location data that has been detected normally instead of false positives. On the other hand, when the position data greatly fluctuates outside the threshold range, as at points a and b in the drawing, the survival time is reset and recounted from the initial value (for example, 0). A short survival time means that the data group containing the position data has low continuity and low reliability (certainty). This survival time is shortened because the location data changes rapidly when an erroneous detection occurs.

A method of counting the survival time will be specifically described. In each data group, if the right angle at a certain time is θa and the right angle at the immediately preceding time is θb, is the absolute value (change amount) of the difference |θa−θb| It is classified according to whether or not. Here, the threshold value N may be a single value regardless of the type of the preceding vehicle, or may be set individually according to the type. In this embodiment, the setting is made individually according to the type of the preceding vehicle. As an example, the threshold N1 is set to 0.7° when the type is a preceding vehicle, the threshold N2 is set to 1.5° when an oncoming vehicle is selected, and the threshold N3 is set to 1.0° when the type is other than the preceding vehicle/oncoming vehicle. set. Similarly, the left angle is also classified according to whether it is equal to or less than the threshold. Then, if the absolute value of the difference for both the right angle and the left angle is equal to or less than the threshold, the survival time is increased by a constant value. For example, increase survival time by +1. On the other hand, if the absolute value of the difference exceeds the threshold for either the right angle or the left angle, the survival time is reset to the initial value because the amount of change is too large and the continuity is low. As a result, the survival time of the data group of object B becomes longer, and the survival time of the data group of object A becomes shorter. By using a combination of the survival time obtained in this way and the Euclidean distance between the position data described above, it is possible to prevent the replacement of the position data as described below.

FIG. 6A is a diagram showing an example of transition of position data. This figure corresponds to the transition example of the position data shown in FIG. 1B. To facilitate understanding, the right angle is also shown as the position data here, but the left angle is assumed to be the same. Also, the Euclidean distance is obtained by associating the right angle and the left angle with the two axes as described above. As shown in the figure, in a situation where the position data (indicated by black circles in the figure) corresponding to the data group of object B has been continuously detected from before time (t-2), time (t-2), time Assume that position data corresponding to the data group of object A (indicated by black triangles in the drawing) is generated due to an erroneous detection at (t−1). At this time, the Euclidean distance between the position data (black circle) at the time (t) and each position data at the time (t−1) one before is x1 and x2, respectively, and the relationship between them is x2 Suppose that <x1. It is also assumed that the data group corresponding to object B has a longer survival time than the data group corresponding to object A. FIG.

At this time, in the present embodiment, position data is associated (that is, added to the data group) based on the Euclidean distance between the data group with the longest survival time and the position data at the next time in order. Specifically, as shown in FIG. 6B, position data (black circles) of the data group of the object B at time (t−1), which is the most recent time, and position data at the next time (t). The distance between is required. It is assumed that there is other position data outside the drawing range of the drawing (see FIG. 1B). Comparing the obtained Euclidean distances x1 and x3, x1 is shorter, so the position data of the Euclidean distance x1 is associated as the tracking destination of the object B data group.

Next, the distance between the data group (black triangle) of object A, which is a data group with a short survival time, and the position data at the next time (t) is obtained. In this case, comparing the Euclidean distances x2 and x4, x2 is shorter. Some other location data is associated as a tracking destination for object A's data group.

In this way, by performing association (addition to the data group) based on the Euclidean distance with priority given to the position data with the longest survival time, the continuity of the position data in the data group of the object B is maintained. Misplacement of position data is prevented. On the other hand, as shown in a comparative example in FIG. If the association is made, the tracking destination will be misplaced. Specifically, after time (t−1), object B disappears (is not detected), and at time (t), only object A is present (detected).

FIG. 7 is a flow chart showing the operation procedure of the controller of the vehicle lamp system. The operation procedure shown here is assumed to be repeated at regular intervals (for example, every 50 ms). It should be noted that the order of the illustrated processing blocks may be changed as long as there is no contradiction or inconsistency in the information processing results, or other processing blocks (not illustrated) may be added, and such embodiments are also excluded. not.

The vehicle position acquisition unit 21 of the controller 20 acquires the position data (right angle, left angle) of the forward vehicle from the imaging device 10 (step S11). Note that when there are a plurality of vehicles ahead, label data for distinguishing each vehicle ahead may be added to the position data output from the imaging device 10. The controller 20 side manages the preceding vehicle (object) corresponding to each position data. This is because the label data added by the imaging device 10 does not necessarily correspond uniquely to each preceding vehicle, and the correspondence relationship may change at each time.

The survival time calculator 22 of the controller 20 calculates the survival time of each data group, and stores the survival time in the memory in association with the data group of each object (step S12). Note that the previous position data does not exist at the time of the first processing, but in that case, for example, a predetermined initial value may be used to calculate the survival time, or the survival time may be uniformly set to 0, or the like. can.

Next, the tracking processing unit 23 of the controller 20 performs tracking processing of the new position data acquired in step S11 in descending order of survival time based on the survival time of the data group of each object obtained in the previous processing opportunity. (step S13). For example, if the survival time of the data group of object B is relatively long, the previous position data (most recent position data) corresponding to this object B is first subjected to matching processing based on the Euclidean distance, that is, tracking processing. After that, the previous position data corresponding to another object A is subjected to tracking processing based on the Euclidean distance.

When the tracking process is completed, the tracking processing unit 23 causes the memory to store a data group to which new position data is added for each object (step S14). The position data are stored in the memory in such a way that the chronological order of the data before the previous time can be recognized, and in association with the survival time.

When the position data corresponding to each object is determined, a light distribution pattern is set by the light distribution pattern setting unit 24 using the position data corresponding to each object (step S15), and control corresponding to the set light distribution pattern is performed. A signal is generated by the control signal generator 25 and output to the vehicle lamps 30L and 30R. After that, the process returns to step S11.

FIG. 8 is a diagram schematically showing an example of light distribution control using position data corresponding to each object. In the illustrated example, there is a forward vehicle 100, which is a preceding vehicle, and a certain range including the forward vehicle 100 in the so-called high beam irradiation range is set as a dimming range (or non-irradiation range) 110a, and the rest is light irradiation. Light irradiation is performed according to the light distribution pattern defined as the range 110b. At this time, the vehicle ahead 100 is changing lanes to the right in the drawing, and the brightness of a partial range 100c (a partial range close to the vehicle ahead 100) of the light irradiation range 110b corresponding to the moving direction is adjusted in advance. control is performed to reduce This control is performed for the purpose of preventing glare from being given to the forward vehicle 100 even when the control for switching from the light irradiation range 110b to the dimming range 110a is delayed due to, for example, a sudden movement of the forward vehicle 100. . During such control, transition of the position data corresponding to the preceding vehicle 100 is required. By preventing the object corresponding to the forward vehicle 100 from being replaced and obtaining continuous position data, such light distribution control can be performed with high accuracy.

According to the embodiment as described above, it is possible to improve the accuracy when tracking the position of the preceding vehicle in chronological order.

It should be noted that the present disclosure is not limited to the contents of the above-described embodiments, and various modifications can be made within the scope of the gist of the present disclosure. For example, in the above-described embodiment, the controller 20 exemplifies the case where two pieces of position data of the forward vehicle are generated, but three or more pieces of position data may be generated. Even in this case, each position data is individually associated with an object to obtain their survival time, and the tracking destination of each position data at the next time is determined based on those survival times and Euclidean distances.

In the above-described embodiment, as an example of light distribution control using position data associated with each object, the moving direction of the forward vehicle is predicted, and the brightness of a partial range of the light irradiation range is adjusted according to the moving direction. Although control for lowering has been mentioned, the content of light distribution control is not limited to this.

10: imaging device, 11: camera, 12: image processor, 20: controller, 21: vehicle position acquisition unit, 22: survival time calculation unit, 23: tracking processing unit, 24: light distribution pattern setting unit, 25: control Signal generator, 30L, 30R: vehicle lamp, 31: ADB unit, 32: low beam unit

Claims (8)

A vehicle locating device comprising a controller,
The controller is
For a plurality of data groups, each of which includes time-series position data relating to a vehicle ahead, the duration of a state in which the amount of change between the position data that precedes and follows in time series for each data group is equal to or less than a predetermined reference value. seeking
when a plurality of new position data relating to the preceding vehicle is generated, obtaining the strength of correlation between the most recent position data in the time series of each of the data groups and the plurality of new position data;
adding each of the plurality of new position data to any one of the data groups based on the strength of the correlation in order from the data group having the relatively longer duration;
Vehicle locator. A vehicle locating device comprising a controller,
The controller is
For each of a plurality of data groups each containing time-series position data relating to a vehicle ahead, a duration of a state in which the amount of change between the position data preceding and succeeding in time-series is equal to or less than a predetermined reference value is determined for each data group. a duration calculation unit;
When a plurality of new position data relating to the preceding vehicle is obtained, correlation calculation for obtaining strength of correlation between the most recent position data in the time series of each of the data groups and each of the plurality of new position data. Department and
a data addition unit that adds each of the plurality of new position data to any one of the data groups based on the strength of the correlation, in order from the data groups having the relatively long duration;
A vehicle locating device, comprising: It is defined that the smaller the Euclidean distance between each of the plurality of new position data and the most recent position data in each of the data groups, the stronger the correlation.
The vehicle position specifying device according to claim 1 or 2. The duration time increases as the state in which the amount of change between the position data chronologically preceding and succeeding in each data group continues to be equal to or less than the reference value, and when the amount of change exceeds the reference value, reset to initial values,
A vehicle position specifying device according to any one of claims 1 to 3. The reference value is set to a different value depending on the type of the preceding vehicle,
A vehicle position specifying device according to any one of claims 1 to 4. A method of vehicle localization performed by a controller comprising:
The controller is
For a plurality of data groups, each of which includes time-series position data relating to a vehicle ahead, the duration of a state in which the amount of change between the position data that precedes and follows in time series for each data group is equal to or less than a predetermined reference value. to seek
when a plurality of new position data relating to the preceding vehicle is generated, determining the strength of correlation between the most recent position data in the time series of each of the data groups and the plurality of new position data;
Adding each of the plurality of new position data to any one of the data groups based on the strength of the correlation in order from the data groups with the relatively longer durations;
A vehicle localization method that performs a Controlling a light distribution pattern of a vehicle lamp using each of the data groups obtained by the vehicle position specifying device according to any one of claims 1 to 5,
Light distribution control device. The light distribution control device according to claim 7 and a vehicle lamp connected to the light distribution control device,
Vehicle lighting system.

Images