JP2005157564A - On-vehicle information providing device for supporting traveling - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent the presence/absence of a pedestrian from being frequently switched due to the location of an object in the neighborhood of the edges of a pedestrian crossing when deciding whether or not an object present on the pedestrian crossing or in its neighborhood is a pedestrian. <P>SOLUTION: The threshold of the continuous detection time of an object, the size threshold of the object, and the threshold of reflective intensity are set according to the presence position of the object (steps S11, S13, S15), and whether or not any pedestrian is present is decided in terms of the continuous detection time, the size of the object and the reflective strength on the bais of this (steps S12, S14, S16), and when it is decided that any pedestrian is present from any point, it is decided that the object is a pedestrian (a step S17). At that time, the threshold of the continuous detection time is set so as to be longer, and the thresholds of the size of the object and the reflective intensity are set so as to be larger at the edges of the pedestrian crossing than the middle of the pedestrian crossing, so that it can be less likely decided that the object is the pedestrian, according as the object is present at the edges of the pedestrian crossing. Thus, it is possible to decide that the object is a pedestrian only when it is possible to clearly decide that the object is the pedestrian. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an on-vehicle information providing apparatus for driving support that detects the presence of a pedestrian crossing and a crossing person in the vicinity of the pedestrian crossing and notifies a passenger of this.

Conventionally, a system has been proposed in which a driver is notified of the presence of a crossing person existing in a predetermined area of an intersection (see, for example, Patent Document 1). This system always provides monitoring information on objects such as vehicles, stopped vehicles, crossers, and two-wheeled vehicles that exist within a predetermined area of an intersection. Then, when this monitoring information is received on the vehicle side, information for determining a traffic vehicle that crosses the road in the predetermined area, that is, a crossing person, from this monitoring information and notifying that the determined crossing person exists in the area. By providing information to the occupant, the occupant can recognize that there is a crosser in a predetermined area.
JP 11-056393 A

By the way, in the above-described conventional method, when a passing body moving along a pedestrian crossing and a traveling body moving toward an area are detected, this is notified to the driver.
In other words, because the driver is notified according to the presence or absence of a vehicle, if it becomes impossible to detect these vehicles temporarily for some reason, depending on whether or not the vehicle is detected, Notification to driver and non notification are switched. Therefore, for example, when there is a pedestrian near the edge of the area, a problem on the camera or radar device side to detect a crossing person, or the presence of a pedestrian based on information from the camera or radar device When a pedestrian cannot be detected temporarily due to a problem or the like on the processing device side, the notification to the driver and the non-notification are switched each time.

In particular, since the vehicle is monitored by a camera, a radar device, etc., depending on the weather, brightness, clothing of the crossing, etc., there may be cases where it is not possible to identify the crossing in the pedestrian crossing. As a result, when there are frequent missing crossers during detection, notification and non-notification to the driver are frequently switched, which may cause the driver to feel bothered or uncomfortable.
Therefore, the present invention has been made paying attention to the above-mentioned conventional unsolved problems, and aims to provide an in-vehicle information providing apparatus for driving support capable of accurately detecting the presence or absence of a crossing person. Yes.

In order to achieve the above-mentioned object, the in-vehicle information providing apparatus for driving support according to the present invention acquires object information on an object existing in a monitoring area including a pedestrian crossing and its vicinity by an object information acquisition unit, and at least as object information Get location information. Then, based on the acquired object information, based on the presence position of the object and its detection status, it is determined whether or not the object is a crossing by means of crossing, and when the object is determined to be a crossing, The information providing means notifies the occupant of the presence of the object.
At this time, the determination condition for determining whether or not the object is a crossing person is changed according to the position where the object exists. In other words, the position of the object in the monitoring area, for example, whether it is in the middle of the pedestrian crossing or whether it is at the end of the pedestrian crossing, etc. I do.

  According to the in-vehicle information providing apparatus for driving support according to the present invention, object information relating to an object existing in a monitoring area including a pedestrian crossing and its vicinity is acquired, and based on this object information, the position of the object and its detection status are obtained. Whether or not this object is a crossing person is determined based on this, but at this time, since the determination condition for determining whether or not the object is a crossing person is changed according to the position of the object, monitoring It is possible to determine whether or not the vehicle is a crossing person under a determination condition suitable for the position of the object in the region, and it is possible to more accurately determine whether or not the object is in the monitoring region.

Embodiments of the present invention will be described below.
First, a first embodiment will be described.
FIG. 1 is a schematic configuration diagram showing an example of an in-vehicle information providing apparatus 100 to which the present invention is applied and an infrastructure system 200 that is arranged on the road side and provides road-side information to the in-vehicle information providing apparatus 100. is there.
In FIG. 1, reference numeral 1 denotes communication between the infrastructure system 200, and road-to-vehicle communication for acquiring object information relating to an object existing in a monitoring area composed of a pedestrian crossing and its neighboring area as infrastructure information. An in-vehicle wireless device 2 as an apparatus is an information provision processing apparatus. This information provision processing device inputs the infrastructure information acquired by the in-vehicle wireless device 1, detects the presence of a crossing person in the monitoring area based on the infrastructure information, and when there is a crossing person, near the driver's seat The information providing device such as the monitor 3 and the speaker 4 provided in the system is operated, and the presence and the position of the crossing person are determined by providing information by an image using the monitor 3, voice guidance or alarm sound using the speaker 4, etc. , Notify the driver.

  On the other hand, the infrastructure system 200 includes a crossing person detection camera 11, a crossing person detection laser radar 12, and a data processing device 13 for inputting detection information of the crossing person detection camera 11 and the crossing person detection laser radar 12. In this data processing device 13, based on detection information from the crossing person detection camera 11 and crossing person detection laser radar 12, if there is an object in the monitoring area and an object exists, The position, reflection intensity, and size are measured, and this is transmitted as object information via the roadside radio 14 such as a radio beacon.

As shown in FIG. 2, the crossing person detection camera 11 and the crossing person detection laser radar 12 are arranged for each pedestrian crossing at an intersection, for example, and include monitoring areas E1 to E1 including a pedestrian crossing and its vicinity. It is arranged at a position where each object in E4 can be detected.
The roadside radio 14 receives the object information of the monitoring areas E1 to E4 set at the intersection at a point before the vehicle entering the intersection enters the intersection. Is arranged at a position where possible.

  In other words, in the infrastructure system 200, the presence or absence of an object in the monitoring area composed of the pedestrian crossing at the intersection and the vicinity thereof is monitored by the crossing detection camera 11 and the crossing detection laser radar 12, and the detection information Is input to the data processing device 13, and object information including the presence and position of the object in the monitoring area is generated and transmitted from the data processing device 13 via the roadside radio 14.

On the other hand, when the in-vehicle information providing apparatus 100 receives the infrastructure information from the roadside radio 14 at the point before the intersection by road-to-vehicle communication between the in-vehicle radio 1 and the roadside radio 14, the infrastructure information, that is, the object information is received. Originally, in the information provision processing device 2, it is determined whether or not there is a crossing person in the monitoring area, and when it is determined that there is a crossing person, the monitor 3 and the speaker 4 are used to cross the presence and position of the crossing person. Provide information about passengers to passengers.
FIG. 3 is a flowchart illustrating an example of a processing procedure of information providing processing executed by the information providing processing device 2. In the information provision processing device 2, while the infrastructure information is received, the information provision processing is executed at a predetermined cycle set in advance.

  In this information providing process, first, in step S1, the object information notified as infrastructure information from the in-vehicle wireless device 1 is read. That is, the presence / absence of an object in the monitoring area, and the presence of an object, the position, size, and reflection intensity are read. The object information may be read from the object information of each monitoring area. For example, when the host vehicle is considered to go straight from the operation status of the direction indicator of the host vehicle, for example, The object information about the monitoring areas E1 and E3 crossing the traveling road is read, and when the host vehicle is considered to turn right based on the operation status of the direction indicator, for example, the object information on the monitoring areas E1 and E4 is read. Alternatively, the driver may be provided with information on the presence of a crosser only for a specific monitoring area.

Next, the process proceeds to step S2, and the presence / absence of a crossing person in the monitoring area is detected according to the process shown in FIG.
Subsequently, the process proceeds to step S3. If it is determined in the process of step S2 that there is a crossing person, the process proceeds to step S4, and notification of the crossing person information to the driver is started. That is, the monitor 3 and the speaker 4 are operated, and the driver is notified of the presence of a crossing person by monitor display and sound. Then, the process ends.
On the other hand, when it is determined in step S3 that there is no crossing, the process proceeds from step S3 to step S5, and the notification of crossing information to the driver is stopped. That is, the operation of the monitor 3 and the speaker 4 is stopped, and the notification of the presence of the traverser to the driver by the monitor display or voice is stopped. Then, the process ends.

The determination of the presence or absence of crossers in step S2 is performed according to the following procedure.
As shown in FIG. 4, first, in step S11, a time threshold value for determining the presence or absence of a crossing person, that is, a time threshold value Ton for determining that there is a crossing person based on the duration time during which an object is detected, A time threshold value Toff for determining that there is no crossing is set based on the duration from the point when the object being detected as a crossing is no longer detected. These time threshold values are set according to the characteristic diagram shown in FIG. 5, and the time threshold value Ton for determining that there is a crossing person is set according to the characteristic line LTon, and the time threshold value for determining that there is no crossing person. Toff is set according to the characteristic line LToff.

  In FIG. 5, the horizontal axis represents the distance from the pedestrian crossing end on the side where the camera 11 and the laser radar 12 are provided to the object, and the vertical axis represents the time threshold for determining the presence or absence of a crossing person. . The time threshold value Ton for determining that there is a crosser is set to a relatively large value in an area where the distance from the pedestrian crossing edge to the object is relatively small, and the distance from the pedestrian crossing edge to the object is relatively large. The area is set to a relatively small value, that is, the vicinity of the end of the pedestrian crossing is set to a larger value compared to the middle of the pedestrian crossing.

On the other hand, the time threshold Toff for determining that there is no crossing is set to a relatively small value in an area where the distance from the pedestrian crossing edge to the object is relatively small, and compared in an area where the distance from the pedestrian crossing edge is relatively large. In other words, the vicinity of the pedestrian crossing edge is set to a smaller value compared to the middle of the pedestrian crossing.
Then, the distance from the pedestrian crossing edge to the object is calculated for the object notified by the object information as being in the monitoring area, and the time threshold value Ton corresponding to the calculated distance is specified from FIG. In addition, the object information is notified that the object exists in the monitoring area, but the object that is not notified as the object information and is not currently detected is crossed from the last existing position when the object exists in the monitoring area. The distance to the sidewalk end is calculated, and the time threshold Toff corresponding to the calculated distance is specified from FIG.

The distance from the pedestrian crossing edge to the object is calculated based on the position of the pedestrian crossing edge detected based on the detection information of the camera 11 and the laser radar 12 and the position of the object, and crossed from the position where the object exists. Calculate the distance to the closest end of either side of the sidewalk.
In this way, if the time threshold value is calculated in the process of step S11, the process proceeds to step S12. Next, the presence or absence of a crossing person is determined based on the continuous detection time and the continuous non-detection time.

The continuous detection time is obtained by measuring the elapsed time from the time when the presence of an object is detected based on the detection information of the camera 11 and the laser radar 12, and by starting a timer when the object is detected. The elapsed time is measured in advance. When there are a plurality of objects, the elapsed time is measured for each object.
The continuous non-detection time is obtained by measuring the elapsed time from the point when the object being detected as a crossing person is no longer detected based on the detection information of the camera 11 and the laser radar 12. The elapsed time is measured by starting a timer when the detected object is no longer detected.

  The time threshold values Ton and Toff set according to the existence position of the object existing in the monitoring area notified as the object information and the object that has been notified as existing in the monitoring area but is no longer detected. Based on the continuous detection time and the continuous non-detection time, the presence / absence of a crossing person is determined. That is, for an object that has been newly detected as an object located in the monitoring area or has been continuously detected, the continuous detection time is not less than the time threshold value Ton for determining that there is a crossing person. When this object is crossing, it is determined that there is a crossing. For an object whose continuous detection time is less than the time threshold value Ton, it is not determined that there is a crossing.

  On the other hand, for an object that has been detected as having a crossing but is no longer detected or continues to be undetected, it is set according to the continuous non-detection time and the distance from the end of the pedestrian crossing. Further, the time threshold Toff for determining that there is no crossing is compared, and when the continuous non-detection time is equal to or longer than the time threshold Toff, it is determined that there is no crossing and there is no crossing. If the continuous non-detection time has not reached the time threshold value Toff, it is not determined that there is no crossing at this time.

When it is determined whether or not there is a crossing person based on the continuous detection time and the continuous non-detection time in this way, the process proceeds to step S13, and the pedestrian crossing edge of the object notified as an object existing in the monitoring area A size threshold value Mon for determining that there is a crossing person and a size threshold value Moff for determining that there is no crossing person are set according to the distance from the vehicle.
These magnitude threshold values are set according to the characteristic diagram shown in FIG. 6, and the magnitude threshold value Mon for determining that there is a crossing person is set according to the characteristic line LMon, and the magnitude threshold value Moff for determining that there is no crossing person. Is set according to the characteristic line LMoff.

  In FIG. 6, the horizontal axis represents the distance from the pedestrian crossing end to the object, and the vertical axis represents the object size threshold, as in FIG. The size threshold value Mon for determining that there is a crosser is set to a relatively large value in an area where the distance from the pedestrian crossing edge to the object is relatively small, and an area where the distance from the pedestrian crossing edge to the object is relatively large. Is set to a relatively small value. That is, the size threshold value Mon is set to a larger value near the pedestrian crossing end than in the middle of the pedestrian crossing.

  On the other hand, the characteristic line LMoff representing the size threshold Moff for determining that there is no crossing is the distance from the pedestrian crossing end to the object, like the characteristic line LMon indicating the size threshold Mon for determining that there is a crossing. Is set to a relatively large value in a relatively small area, and is set to a relatively small value in an area where the distance from the pedestrian crossing edge to the object is relatively large. That is, the size threshold value Moff is set to a value larger in the vicinity of the pedestrian crossing end than in the middle of the pedestrian crossing. Furthermore, the size threshold value Moff for determining that there is no crossing is set to a value smaller than the size threshold value Mon for determining that there is a crossing, and the difference between the size threshold value Mon and Moff is: It is set so that the vicinity of the pedestrian crossing edge is larger than the middle of the pedestrian crossing.

Also in this case, the distance from the pedestrian crossing edge to the object is calculated based on the position of the object detected based on the detection information of the camera 11 and the laser radar 12. The size of the object may be specified by, for example, the height of the object, or may be specified by only the width, or by the height and width.
When the object size threshold value is calculated based on the distance from the pedestrian crossing end to the object in this way, the process proceeds to step S14, the size threshold value set in step S13, the camera 11 and the laser radar 12 are set. The presence or absence of a crossing person is determined based on the size of the object specified based on the detected information.

  Specifically, the size threshold value Mon for determining that there is an object set according to the distance from the pedestrian crossing edge of the object is detected for each object notified as existing in the monitoring area. The size of the object is compared. If the size of the detected object is equal to or larger than the size threshold value Mon, it is determined that there is a crossing person. When the size of the detected object is not equal to or larger than the size threshold value Mon, it is not determined that there is a crossing person even if the object is detected.

On the other hand, when the size of the object being detected as a crossing person is smaller than the size threshold Moff, it is determined that this object is not a crossing person and that there is no crossing person.
Even if the size of the object being detected is below the threshold value Mon for determining that there is a crossing person, it is not determined that there is no crossing person as long as it does not fall below the size threshold value Moff.
If the presence or absence of a crossing person is determined based on the size of the detected object in this way, the process proceeds to step S15, and the object reflection intensity threshold is calculated based on the characteristic diagram shown in FIG.

In FIG. 7, the horizontal axis represents the distance from the pedestrian crossing end to the object, and the vertical axis represents the reflection intensity threshold, as in FIG.
As shown in FIG. 7, both the reflection intensity threshold value Ron for determining that there is a crossing person and the reflection intensity threshold value Roff for determining that there is no crossing person are relatively small in the distance from the end of the pedestrian crossing to the object. In a region, a relatively large value is set, and in a region where the distance from the crosswalk end to the object is relatively large, the reflection intensity threshold is set to a relatively small value. That is, the reflection intensity threshold values Ron and Roff are set to values larger near the pedestrian crossing end than in the middle of the pedestrian crossing. Further, the reflection intensity threshold Roff for determining that there is no crossing is set to a smaller value than the reflection intensity threshold Ron for determining that there is a crossing, and the difference between the reflection intensity thresholds Ron and Roff Is set to a larger value near the end of the pedestrian crossing than in the middle of the pedestrian crossing.

  If the reflection intensity threshold value is set in the process of step S15, the process proceeds to step S16, and the reflection intensity of the object specified based on the detection information of the camera 11 and the laser radar 12 is compared with the threshold value. , Determine the presence or absence of crossers. That is, when the reflection intensity of the object is greater than or equal to the reflection intensity threshold Ron, it is determined that there is a crossing person. On the other hand, when the reflection intensity of the object being detected as having a crossing falls below the reflection intensity threshold Roff, it is determined that there is no crossing. Even if the reflection intensity of the object being detected as having a crossing is below the reflection intensity threshold Ron, it is not determined that there is no crossing as long as it exceeds the threshold Roff for determining that there is no crossing. .

If it is determined whether or not there is a crossing person based on the reflection intensity in this way, the process proceeds to step S17, where the continuous detection time and continuous non-detection time determined in steps S12, S14, and S16 are detected. The final presence / absence of a crossing person is determined based on the size and the presence / absence of a crossing person based on the reflection intensity of the detected object.
Specifically, when it is determined that there is a crossing person based on any one of the continuous detection time and the continuous non-detection time, the size of the detection object, and the reflection intensity of the detection object, the detection object is the crossing person. If it is determined that there is no crossing by any means, it is determined that the detected object is not a crossing. If it is not determined that there is a crossing or no crossing, the previous determination result is continued. For the object that is determined not to be a crossing person, the measurement of the continuous non-detection time is ended.

Next, the operation of the first embodiment will be described.
Assume that the host vehicle M is approaching an intersection as shown in FIG.
The own vehicle M reaches an area where the infrastructure information from the infrastructure system 200 can be received, receives the infrastructure information by road-to-vehicle communication between the in-vehicle wireless device 1 and the roadside wireless device 14, and receives an object in the monitoring region. When data such as presence / absence, position, and magnitude reflection intensity are read, the process proceeds from step S1 to step S2 in FIG. 3, and based on these, the presence / absence of a crossing person is determined.

  At this time, for example, when a pedestrian X exists on the road in the monitoring area E1 including a pedestrian crossing that crosses the traveling path of the host vehicle, the pedestrian X is infrastructure information as an object located in the monitoring area E1. To be notified. And since the pedestrian X exists in the boundary vicinity of a pedestrian crossing and a running path, the distance from the pedestrian crossing edge becomes comparatively short. Therefore, as shown in FIG. 5, the time threshold value Ton for determining that there is a crossing person is set to a relatively large value as the threshold value of the determination time. Therefore, when this pedestrian X waits for a signal or starts to cross a pedestrian crossing, when the continuous detection time exceeds the time threshold value Ton, the detected object corresponding to the pedestrian X is crossed. It is determined that it is a person.

At this time, for example, when the pedestrian X only passes by the pedestrian crossing, the period during which the pedestrian X is continuously detected is shorter than the time threshold value Ton. It is not judged that there is.
As described above, the pedestrian X existing in the monitoring area E1 is determined to be a crosser only when the continuous detection time exceeds the time threshold value Ton and can be clearly regarded as a crosser. Thus, for example, it is possible to avoid erroneous detection of a pedestrian or the like passing only near a pedestrian crossing as a crossing person.

  In this way, when the presence or absence of a crossing person is detected based on the time threshold, there is a crossing person based on one of the continuous detection time, the continuous non-detection time, the size of the object, and the reflection intensity of the object. If it is determined, the process proceeds from step S3 to step S4, and the monitor 3 or speaker 4 notifies the driver of the presence of the crosser by information display or voice. Therefore, at this point, the driver can recognize the presence of the crossing person and can pay attention to the crossing person.

  At this time, when the pedestrian X enters the pedestrian crossing from the outside of the monitoring area E1 at a relatively high speed, the pedestrian X proceeds to the vicinity of the center of the pedestrian crossing in a relatively short period of time. At this time, the time threshold value Ton for determining that there is a crosser is shorter in the middle of the pedestrian crossing than in the vicinity of the end of the pedestrian crossing, and it is necessary for the pedestrian X to reach the central part of the pedestrian crossing. As the time becomes shorter, the time threshold value Ton for determining that there is a crossing person becomes shorter, and it is determined that there is a crossing person at an earlier time point. It can be judged that there is a person. Therefore, the presence of the pedestrian X can be detected at an earlier time point, that is, the pedestrian X entering the pedestrian crossing can be detected at an earlier time point, and this can be notified to the driver. Attention to crossers at the stage.

In addition, once the object detected as a crossing is no longer detected, it is determined that there is no crossing when the duration from the point of no longer detection becomes the time threshold Toff or more, It is judged that there is no crossing at a point when it can be clearly considered that there is no crossing.
Therefore, for example, depending on the weather and brightness of the camera 11 and the laser radar 12 and the clothes of the crossing person, that is, the material of the clothes, only a weak reflected wave may be obtained. However, even in such a case, it is not erroneously determined that there is no crossing person, and it can be determined that the person is still a crossing person. Therefore, when the information cannot be detected temporarily, the driver is not frequently switched to whether or not information is provided, and it is possible to avoid giving the driver a sense of incongruity and annoyance. Can do.

In addition, when a pedestrian stops at the pedestrian crossing and turns back without crossing the pedestrian crossing, it is no longer detected after it is determined that there is a pedestrian. Since the time threshold value Toff is set to a relatively small value, it is quickly determined that the person is not a crossing person from the time point when it is no longer detected.
Also, if there are pedestrians who are crossing the pedestrian crossing, it is determined that there is no crossing in the area where the distance from the end of the pedestrian crossing to the object is short, that is, near the pedestrian crossing. Since the time threshold value Toff is set to a small value, if a state in which the pedestrian who has finished crossing is removed from the monitoring area and is no longer detected as an object in the monitoring area continues, when the relatively short time threshold value Toff has elapsed. It is determined that there is no crossing immediately. Therefore, a pedestrian who has finished crossing a pedestrian crossing can be promptly excluded from information provision targets.

Further, at this time, even if the crossing person who has finished crossing is in a state where it is detected again near the end of the monitoring area, the time threshold value Ton for determining that there is a crossing person at this time is Since it is set to a relatively large value in the vicinity of the end of the sidewalk, it can be avoided that the pedestrian who has finished crossing is again detected as a crossing person.
At this time, depending on the weather, the brightness of the surroundings, the clothes of the pedestrian, etc., it becomes impossible to detect the pedestrian in the monitoring area, even if there is actually a pedestrian. In some cases, this pedestrian cannot be detected and may be frequently lost. For this reason, as long as the duration time during which the object is detected does not exceed the time threshold value Ton for determining that there is a crossing person, the determination of the presence or absence of a crossing person based on the continuous detection time will determine that there is no crossing person. .

However, at this time, the presence / absence of a crossing person is determined based not only on the continuous detection time and the continuous non-detection time but also on the size of the object and the reflection intensity of the object.
Therefore, for example, when a pedestrian X is newly detected in the monitoring area, this pedestrian X is not determined to have a crossing at this point in the determination of the presence or absence of a crossing based on the continuous detection time. However, if the object is a pedestrian X and its size and reflection intensity are equal to or greater than the threshold value for determining that there is a crosser set based on FIGS. 6 and 7, it is determined that there is a crosser. The Therefore, at this point, the driver can be notified of the presence of the crosser. If the detected object is not a pedestrian X but a plantation by the road, its size and its reflection intensity are below the threshold for determining that there is a crosser, and it is not determined that there is a crosser. .

  At this time, as shown in FIGS. 6 and 7, in the middle of the pedestrian crossing, the threshold for determining that there is a crossing person is set to a relatively small value. Therefore, since an object in the middle of a pedestrian crossing is more likely to be a crossing person, the pedestrians are present in the pedestrian crossing, but are not large enough due to the surrounding environment such as weather and brightness. Even when the reflection intensity cannot be obtained, it is possible to avoid erroneously determining that there is no crossing person.

  Further, as shown in FIGS. 6 and 7, the threshold for determining that there is no crossing based on the size and reflection intensity of the object is set to a value smaller than the threshold for determining that there is a crossing. . Therefore, when it falls below the threshold for judging that there is a crossing, it is not judged that there is no crossing, and when it falls below the threshold for judging that there is no crossing, it is judged that there is no crossing. Even when a sufficient size and reflection intensity cannot be obtained due to the above, it is possible to avoid determining that there is no crossing even though there are pedestrians on the pedestrian crossing.

  In addition, the difference between the threshold for determining the presence of a crosser and the threshold for determining the absence of a crosser based on the size and reflection intensity of the object is larger at the pedestrian crossing edge than in the middle of the pedestrian crossing. The pedestrian crossing end is more difficult to determine that there is a crossing and once it is determined that there is a crossing, it is difficult to determine that there is no crossing. Therefore, frequently determine whether there is a crossing or no crossing for an object that is located at the pedestrian crossing and near the boundary of the monitoring area and that is detected or not detected as an object in the monitoring area. It is possible to avoid the driver from feeling uncomfortable or bothering due to frequent changes in information provided to the driver.

Next, a second embodiment of the present invention will be described.
Since the second embodiment is the same as the first embodiment except that the processing procedure for determining the presence or absence of a crossing person in FIG. 3 is different, the same reference numerals are given to the same parts. Detailed description thereof will be omitted.
In the second embodiment, the presence / absence of a crossing person is determined according to the flowchart of FIG.
Specifically, first, in step S21, an object notified by the object information as being present in the monitoring area, an object that has been notified as existing in the monitoring area but is not currently detected, or a state that is not detected. For an object that continues, the weight related to the continuous detection time and continuous non-detection time of the object is obtained based on the position of the object.

  This weight is obtained from, for example, the characteristic diagram of FIG. In FIG. 9, the horizontal axis represents the distance from the end of the pedestrian crossing to the object, and the vertical axis represents the determination time weight. As indicated by the characteristic line Lon in FIG. 9, the determination time weight kTon for determining the presence of a crossing person is set for the object notified as being in the monitoring area, and from the pedestrian crossing edge to the object. It is set to a relatively small value in an area where the distance is relatively small, and is set to a relatively large value in an area where the distance from the pedestrian crossing edge to the object is relatively large. That is, the weight kTon is set to a value smaller in the vicinity of the pedestrian crossing edge than in the middle of the pedestrian crossing.

  On the other hand, the determination time weight kToff for determining that there is no crossing is set for an object that has been detected as a crossing but is no longer detected, or an object that has not been detected yet. As indicated by the characteristic line Loff in Fig. 9, in a region where the distance from the last detected object to the pedestrian crossing end is relatively small, a relatively large value is set. In a region where the distance to the road is relatively large, a relatively small value is set, that is, a value near the pedestrian crossing edge is set to a larger value than the middle of the pedestrian crossing.

Note that the distance from the pedestrian crossing edge to the object is calculated based on the position of the object detected by the camera 11 and the laser radar 12, as in the first embodiment.
If the weights related to the continuous detection time and the continuous non-detection time are calculated in this way, the process proceeds to step S22. Next, the weight related to the size of the object is calculated based on the characteristic diagram of FIG. .
In FIG. 10, the horizontal axis represents the distance from the pedestrian crossing end to the object, and the vertical axis represents the weight kM related to the size of the object. This weight kM is set to a relatively small value in an area where the distance from the pedestrian crossing edge to the object is relatively small, and is set to a relatively large value in an area where the distance from the pedestrian crossing edge to the object is relatively large. . That is, the weight kM is set to a larger value near the pedestrian crossing edge than in the middle of the pedestrian crossing.

As for the distance from the pedestrian crossing edge to the object, for the object that is notified as the object existing in the monitoring area, the distance to the position of the object is calculated, and the object exists in the monitoring area. For an object that is no longer notified or an object that has been in a state where it is no longer notified, the calculation is performed according to the position of the object at the time when the presence was last notified.
Next, the process proceeds to step S23, and the weight related to the reflection intensity of the object is calculated based on the characteristic diagram of FIG.

  In FIG. 11, the horizontal axis represents the distance from the pedestrian crossing edge to the object, and the vertical axis represents the weight kR related to the reflection intensity of the object. The reflection intensity weight kR of the object is set to a relatively small value in a region where the distance from the pedestrian crossing end to the object is relatively small, and is set to a relatively large value in a region where the distance from the pedestrian crossing end to the object is relatively large. In other words, the pedestrian crossing end is set to a smaller value than the middle of the pedestrian crossing.

As for the distance from the pedestrian crossing edge to the object, for the object that is notified as the object existing in the monitoring area, the distance to the position of the object is calculated, and the object exists in the monitoring area. For an object that is no longer notified or an object that is in a state where it is no longer notified, the calculation is performed according to the position of the object at the time when the presence was last notified.
Next, the process proceeds to step S24, and there is a crossing person based on the parameters of the continuous detection time, the size of the object, the reflection intensity of the object, and the weights at the positions of the objects specified in steps S21 to S23. Judgment whether it is possible to be considered.

Specifically, as shown in the following equation (1), a sum is calculated by multiplying each parameter by a corresponding weight, and this sum is used as a determination parameter Pon with a crossing person. The threshold value αon for determining that there is a set crossing person is compared. Then, when the determination parameter Pon with crossing exceeds the threshold value αon, it is determined that it can be considered that there is a crossing.
Pon =
(Continuous detection time x its weight) + (size x its weight) + (reflection intensity x its weight)> αon
...... (1)

Next, the process proceeds to step S25, and it is determined whether or not it is possible to consider that there is no crossing based on each parameter and the corresponding weight.
That is, as shown in the following equation (2), the respective parameters are multiplied by their corresponding weights to calculate a sum, and this sum is set as a parameter Poff for determining no traverser. The threshold value αoff for determining that there is no crossing is set. When the determination parameter Poff for no crossing is less than the threshold value αoff, it is determined that no crossing can be considered. The threshold value αoff is a value satisfying αoff <αon.

Poff = [− continuous non-detection time × (weight thereof)] + (size immediately before non-detection × weight thereof)
+ (Reflection intensity immediately before non-detection × its weight) <αoff (2)
Then, the process proceeds to step S26, and it is determined whether or not it is possible to specify that there is a crossing person or no crossing person based on the determination parameters Pon and Poff. If the determination parameter Pon or Poff satisfies the above expression (1) or (2) and it can be determined that there is a crossing person or no crossing person, the process is terminated.

On the other hand, if the determination parameter Pon or Poff does not satisfy the expression (1) or (2) and it cannot be specified that there is a crossing or no crossing, the process proceeds to step S27. The previous judgment result is maintained. Then, the process ends.
Therefore, in the second embodiment, when the object in the monitoring area is accurately detected by the camera 11 or the laser radar 12, the expression (1) is governed by the continuous detection time. At this time, as shown in FIG. 9, the time weight kTon for determining that there is a crosser is set to be a small value in the vicinity of the pedestrian crossing edge. Is present, the continuous detection time of this object is increased to some extent, and when it is clearly determined that the vehicle is a crosser, it is determined that the vehicle is a crosser.

Also, since the time weight kTon is set to be a large value at the middle of the pedestrian crossing, this object is used when the pedestrian enters the pedestrian crossing at a relatively fast speed and reaches the middle of the pedestrian crossing. Even if the continuous detection time is relatively short, it is determined that the person is a crossing person.
At this time, even if the camera 11 or the laser radar 12 cannot temporarily detect the exact size or reflection intensity of the object due to the influence of the weather, the clothes of the pedestrian, or the like, the (1 As shown in equation (5), the determination is made based on the continuous detection time, the size of the object, and its reflection intensity. Therefore, if the continuous detection time is long to some extent, the object is accurately determined to be a crossing person. Can do.

  In addition, at this time, the components related to the size and reflection intensity of the object are set so that the middle of the pedestrian crossing is larger than the vicinity of the pedestrian crossing end as shown in FIGS. Even if the size and reflection intensity of the object are somewhat insufficient, it tends to be judged as a crossing person, and in fact there is no crossing person even though there is a crossing person in the middle of the pedestrian crossing. An erroneous determination can be avoided.

Conversely, in the vicinity of the pedestrian crossing, the components related to the size and reflection intensity of the object are set to be smaller than those in the middle of the pedestrian crossing. Therefore, it is possible to avoid erroneously determining that an object such as a plant near the pedestrian crossing is a crossing person.
On the other hand, if the information related to the object detected as a crossing person is not notified as object information, that is, if this object is no longer detected in the monitoring area, the object immediately before this object is no longer detected In the state where the size and the reflection intensity are accurately detected, the determination that the object is not present is governed by the continuous non-detection time from the equation (2).

  At this time, as shown in FIG. 9, the time weight kToff for determining that there is no crossing is set to be larger in the vicinity of the end of the pedestrian crossing than in the middle of the pedestrian crossing. Since it tends to be easy to judge, if a crossing person who passes the pedestrian crossing crosses the pedestrian crossing and goes out of the monitoring area, it will immediately determine that this pedestrian is not a crossing person. Notifying the driver of pedestrians that do not need to be notified to the driver can be quickly avoided.

  In the middle of the pedestrian crossing, the time weight kToff is set to be relatively small, making it difficult to judge that the person is not a pedestrian, so the weather, pedestrian clothing, etc. Because of this, if this pedestrian cannot be detected temporarily, it is possible to prevent it from being mistakenly determined that it is not a traverser quickly. It can be determined that it is not a crossing person when it can be regarded as not a person.

  In addition, since the determination that there is no crossing is made based on the continuous non-detection time, the size of the object, and the reflection intensity thereof, the size of the object and the reflection intensity immediately before the object is no longer detected are determined. If it is an object that has been obtained enough and can be clearly regarded as a crossing person, it is determined that the object is not a crossing person when the continuous non-detection time is somewhat longer and it can be regarded as clearly not a crossing person. On the contrary, the size of the object just before the object is no longer detected and its reflection intensity are not sufficiently obtained, so it can not be judged as a traverser from the point of object size and its reflection intensity If the object is a simple object, it can be quickly determined that the person is not a crossing person when the continuous non-detection time is shorter.

  Also, at this time, the size of the object just before the object is no longer detected and its reflection intensity are set to a value smaller than the middle of the pedestrian crossing near the pedestrian crossing edge, The component related to the size and reflection intensity of the object immediately before it is no longer detected increases, and it tends to be difficult to judge that it is not a crossing person, and sufficient object size and reflection intensity cannot be obtained due to the weather etc. It can be avoided that it is determined that the person is not a crossing due to the object, and conversely, near the edge of the pedestrian crossing, the component related to the size and reflection intensity of the object is reduced, and it is easy to determine that the person is not a crossing person. If the tendency is such that a sufficient size or reflection intensity of the object cannot be obtained, it can be determined that the person is not a crossing person more quickly.

  As described above, in the second embodiment, the determination condition for determining that the object is a crossing person and the determination that the object is not a crossing person are made according to the position of the object in the monitoring area. Since the judgment condition has been changed, it is possible to avoid frequently repeating the determination of the presence or absence of a crossing at the end of a pedestrian crossing, and whether the person is a crossing according to the location of the object. It is possible to accurately determine whether or not, and in this case as well, it is possible to obtain the same operational effects as those of the first embodiment.

In each of the above embodiments, a case has been described in which each pedestrian crossing is set as a monitoring area at an intersection that branches in four directions. However, the present invention is not limited to this, and the present invention is applicable to a three-way road or the like. In addition, the present invention can be applied not only to a pedestrian crossing but also to a road having a crossable passage.
In each of the above embodiments, the in-vehicle wireless device 1 corresponds to the object information acquisition unit, and the process of determining the presence or absence of a crossing person in step S2 of FIG. The process of S5 corresponds to the information providing means.

It is a schematic block diagram which shows an example of the vehicle-mounted information provision apparatus 100 in this invention, and the infrastructure system 200 by the side of a road. 2 is an example of an infrastructure system 200; 5 is a flowchart illustrating an example of a processing procedure of information providing processing executed by the information providing processing device 2 of the in-vehicle information providing device 100. It is a flowchart which shows an example of the process sequence of the process which judges the presence or absence of a crossing person in step S2 of FIG. It is a characteristic figure of the time threshold with respect to the continuous detection time and continuous non-detection time of an object. It is a characteristic figure of the size threshold with respect to the size of an object. It is a characteristic view of the reflection intensity threshold with respect to the reflection intensity of an object. It is a flowchart which shows an example of the process sequence of the process which judges the presence or absence of a crossing person in 2nd Embodiment. It is a characteristic view of the time weight with respect to the continuous detection time and continuous non-detection time of an object. It is a characteristic view of the size weight with respect to the size of an object. It is a characteristic view of the reflection intensity weight with respect to the reflection intensity of an object.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 In-vehicle wireless device 2 Information provision processing device 3 Monitor 4 Speaker 11 Crossing person detection camera 12 Crossing person detection laser radar 13 Data processing device 14 Roadside radio 100 In-vehicle information provision device 200 Infrastructure system

Claims (8)

Object information acquisition means for acquiring object information including at least position information regarding an object existing in a monitoring area including a pedestrian crossing and its vicinity;
A traversing discrimination means for discriminating whether or not the object is a traversing person based on the position of the object acquired by the object information acquisition means and its detection status;
An information providing means for notifying an occupant of the presence of the crossing when the crossing determination means determines that the object is a crossing;
The crossing means determination means is configured to change the determination condition for determining whether or not the object is a crossing person according to the position of the object acquired by the object information acquisition means. An in-vehicle information providing apparatus for driving support characterized by the above. When the continuous detection time for continuously detecting the object is equal to or greater than a threshold value of a continuous detection time set according to the position of the object, the crossing person determining means determines that the object is a crossing person. Discriminate,
2. The in-vehicle information for driving support according to claim 1, wherein the threshold value of the continuous detection time is set to a larger value on the side of the side of the pedestrian crossing than in the middle of the pedestrian crossing. Providing device. The crossing person determining means sets a continuous non-detection time in which the object that has been determined to be a crossing person is not detected continuously from the point in time according to the position where the object was present. Determining that the object is not a traverser when the continuous non-detection time threshold is greater than or equal to
The driving support according to claim 1 or 2, wherein the threshold value of the continuous non-detection time is set to a smaller value on the side of the side of the pedestrian crossing than in the middle of the pedestrian crossing. In-vehicle information providing device. The object information acquisition means acquires the position information and size information related to the size of the object as the object information,
The crossing means discriminating means, when the size of the object specified by the size information is not less than a first size threshold set according to the position of the object, And when the size of the object is equal to or smaller than a second size threshold set according to the position of the object, the object is determined not to be a traverser,
The second magnitude threshold is less than the first magnitude threshold and the difference between the first magnitude threshold and the second magnitude threshold is a transverse The in-vehicle information providing apparatus for driving support according to any one of claims 1 to 3, wherein a value on the side of the sidewalk of the pedestrian crossing is set to a larger value than a middle part of the sidewalk. The object information acquisition means acquires the position information and size information related to the size of the object as the object information,
The crossing means discriminating means, when the size of the object specified by the size information is not less than a first size threshold set according to the position of the object, And when the size of the object is equal to or smaller than a second size threshold set according to the position of the object, the object is determined not to be a traverser,
The second magnitude threshold is set to a value smaller than the first magnitude threshold and smaller in the middle of the pedestrian crossing than on the side of the pedestrian crossing. The in-vehicle information providing apparatus for driving support according to any one of claims 1 to 4. The object information acquisition means acquires the position information and reflection intensity information related to the reflection intensity of the object as the object information,
The crossing means discriminating means, when the reflection intensity of the object specified by the reflection intensity information is equal to or more than a first reflection intensity threshold set according to the position where the object exists, And when the reflection intensity of the object is less than or equal to a second magnitude threshold set according to the position of the object, the object is determined not to be a crossing person,
The second reflection intensity threshold is smaller than the first reflection intensity threshold, and the difference between the first reflection intensity threshold and the second reflection intensity threshold is: The on-board information providing device for driving support according to any one of claims 1 to 5, wherein a value on the side of the sidewalk of the pedestrian crossing is set to a larger value than a middle of the pedestrian crossing. The object information acquisition means acquires the position information and reflection intensity information related to the reflection intensity of the object as the object information,
The crossing means discriminating means, when the reflection intensity of the object specified by the reflection intensity information is equal to or more than a first reflection intensity threshold set according to the position where the object exists, And when the reflection intensity of the object is less than or equal to a second magnitude threshold set according to the position of the object, the object is determined not to be a crossing person,
The second reflection intensity threshold value is set to a value smaller than the first reflection intensity threshold value and smaller in the middle of the pedestrian crossing than the side of the pedestrian crossing. The in-vehicle information providing apparatus for driving support according to any one of claims 1 to 6. The object information acquisition means acquires, as the object information, at least one of the position information, size information related to the size of the object, and reflection intensity information related to the reflection intensity of the object,
The traversing means is a detection status component weighted at least one of a continuous detection time component weighted by a continuous detection time for continuously detecting the object and size information of the object and the reflection intensity information. When the sum of and exceeds the threshold, the object is determined to be a traverser,
A continuous non-detection time component weighted to a continuous non-detection time from the point when the object determined to be a crosser is no longer detected, and its size information immediately before the object determined to be a crosser is no longer detected; and When the difference between the detection status component weighted on at least one of the reflection intensity information exceeds the threshold, it is determined that the object is not a crossing person,
The weighting for the continuous detection time is performed so that the case where the object is present on the side of the sidewalk of the pedestrian crossing is smaller than the case where the object is present in the middle of the pedestrian crossing, and the continuous Weighting for the non-detection time is performed so that the case where the object is present on the side of the pedestrian crossing is larger than the case where the object is present in the middle of the pedestrian crossing,
The weighting for the size information and the reflection intensity information is present when the object is present in the middle of the pedestrian crossing or when it is present at the end of the pedestrian crossing. The in-vehicle information providing device for driving support according to claim 1, wherein the in-vehicle information providing device is performed so as to be larger than that in the case of driving.

Images