Method for determining object type of reflective object on track

• US 20030174054A1
• Filed: 02/03/2003
• Published: 09/18/2003
• Est. Priority Date: 03/12/2002
• Status: Active Grant

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Nissan Motor Co., Ltd.
• Inventors: Shimomura, Noriko
• Granted Patent: US 6,888,622 B2
• Time in Patent Office: Days
• US Class Current: 340/435
• CPC Class Codes: G01S 13/522 using transmissions of interrupted pulse modulated waves
• CPC Class Codes: G01S 13/72 for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar
• CPC Class Codes: G01S 13/867 Combination of radar systems with cameras
• CPC Class Codes: G01S 13/931 of land vehicles
• CPC Class Codes: G01S 17/86 Combinations of lidar systems with systems other than lidar, radar or sonar, e.g. with direction finders
• CPC Class Codes: G01S 2013/9325 for inter-vehicle distance regulation, e.g. navigating in platoons
• CPC Class Codes: G01S 2013/9329 cooperating with reflectors or transponders
• CPC Class Codes: G01S 7/412 based on a comparison between measured values and known or stored values
• CPC Class Codes: G01S 7/415 Identification of targets based on measurements of movement associated with the target
• CPC Class Codes: G08G 1/167 Driving aids for lane monitoring, lane changing, e.g. blind spot detection

Abstract

A plurality of reflective points in front of one'"'"'s vehicle in a traveling direction are detected by a radar device. The plurality of detected reflective points are grouped based on a predetermined standard, and the grouped reflective points are regarded as detection objects. Relative speeds of the detected detection objects are measured with respect to one'"'"'s vehicle. Types of the detection objects are determined based on the measured relative speeds.

Claims

1. A method for determining an object type of a reflective object on a track, comprising:

a reflective point detecting step of detecting a plurality of reflective point in front of one'"'"'s vehicle in a traveling direction;

a first detecting step of grouping the plurality of reflective points detected in the reflective point detecting step by a predetermined standard and regarding the grouped reflective points as detection objects;

a relative speed measuring step of measuring relative speeds of the detection objects detected by the first detecting step with respect to one'"'"'s vehicle; and

a first determining step of determining types of the detection objects based on the relative speeds detected in the relative speed measuring step.

2. The method for determining an object type according to claim 1,

wherein the reflective point detecting step includes scanning ahead of one'"'"'s vehicle in a traveling direction by a predetermined detection wave output device, detecting reflective points reflecting detection waves outputted by the detection wave output device, and measuring relative positions of the reflective points with respect to one'"'"'s vehicle,

the first detecting step includes detecting the detection objects by determining reflective points detected at relative positions proximal to one another among the reflective points to be points on the same detection object,

the method for determining an object, further comprises a first labeling step of giving the same labels to detection objects existing within a first predetermined distance among the detection objects, and

the first determining step determines a set of the detection objects being given the same labels in the first labeling step to be any of a group of objects belonging to the same object and a group of the same type of objects.

3. The method for determining an object type according to claim 2, further comprising:

a first data storing step of identifiably storing first data including labeling results by the first labeling step for each of the detection objects detected by the first detecting step after performing steps up to the first labeling step prior to performing the first determining step;

a second detecting step of further performing steps up to the first detecting step;

a second labeling step of attaching the same labels to detection objects existing within the first predetermined distance among detection objects detected by the second detecting step;

a comparing step of mutually comparing the labeling results for a set of the detection objects being given the same labels by the second labeling step; and

a third labeling step of giving the same labels again to the detection objects which are given the same labels by the first labeling step and deleting labels being given to detection objects which are given no labels by the first labeling step among the set of the detection objects, in a case of determining that the labeling results are not entirely the same based on a comparison result by the comparing step.

4. The method for determining an object type according to claim 2,

wherein, prior to performing the first determining step,

a shape detecting step of detecting a relative position of a reference line provided on a lane in front of one'"'"'s vehicle with respect to one'"'"'s vehicle is further performed,

in the first determining step, distances from respective sets of the detection objects being given the same labels to the reference line provided on the lane in front of one'"'"'s vehicle are measured and a variance value of the distances is measured based on a detection result by the shape detecting step, and

the sets of the detection objects are determined to be groups of road signs in a case where the variance value is smaller than a predetermined value and the sets of the detection objects are detected at relative positions out of the lane in front of one'"'"'s vehicle by the first detecting step.

5. The method for determining an object type according to claim 2,

wherein, prior to performing the first determining step,

a size measuring step of measuring sizes of the detection objects detected by the first detecting step is further performed, and

in the first determining step, the set of the detection objects is determined to be a group of objects belonging to the same vehicle in a case where detection objects larger than a predetermined size are included in the set of the detection objects which are given the same labels by the first labeling step.

6. The method for determining an object type according to claim 4, further comprising:

after performing steps up to the first determining step,

a second data storing step of identifiably storing second data for each of first detection objects constituting the sets of the detection objects, the second data including labels given to the sets of the detection objects and a determination result to an effect that the sets of the detection objects are the groups of the road signs for at least the sets of the detection objects determined to be the groups of the road signs by the first determining step;

a third detecting step of further performing steps up to the first detecting step;

a retrieving step of retrieving the second data for detection objects detected by the third detecting step;

a fourth labeling step of giving the same labels to the first detection objects detected by the third detecting step in a case where a predetermined number or more of the first detection objects are detected by the third detecting step based on a retrieving result by the retrieving step, and of further giving the same labels as those of the first detection objects to other detection objects in a case where the other objects exist within a second predetermined distance longer than the first predetermined distance from any of the first detection objects detected by the third detecting step; and

a second determining step of determining the other detection objects to be the road signs.

7. The method for determining an object type according to claim 5,

wherein the size measuring step is performed prior to the first labeling step, and

in the first labeling step, a second detection object, which is larger than a predetermined size, is detected by the first detecting step, and further, the same label as those of the second detection objects is given to the third detection object in a case where the third detection object is detected within a range located deeper comparing the second detection object seeing the traveling direction of the detection wave from the detection wave output device and within a third predetermined distance longer than the first predetermined distance from the second detection object, and further, the same label as those of the second detection objects is given to a fourth detection object in a case where the fourth detection object exists within the third predetermined distance from any of the second detection object and the third detection object.

8. The method for determining an object type according to claim 7,

wherein steps up to the first labeling step are performed a certain number of times in the first determining step, and the second detection object and a detection object being given the same label as that of the second detection object are determined to be a group of objects belonging to the same large-sized vehicle in the first determining step in a case where a detection object being same as the second detection object and the third detection object is detected in each time of the step.

9. The method for determining an object type according to claim 1,

wherein the first detecting step includes grouping in time series groups of reflective points existing proximally to one another among the reflective points detected by the reflective point detecting step,

the relative speed measuring step includes regarding the grouped groups of the reflective points as the detection objects, and obtaining relative speeds of one'"'"'s vehicle with respect to the detection objects based on time-serial distance data between the detection objects and one'"'"'s vehicle, and

the method for determining an object type further comprises:

a third determining step of determining whether or not the detection objects detected by the first detecting step have a possibility of being road relating structural objects based on sizes thereof obtained from types determined for the detection objects and a grouping result in the past;

a fourth determining step of determining a degree at which the reflective points relating to the detection objects determined to have the possibility of being the road relating structural objects are crowded in a depth direction with respect to one'"'"'s vehicle in the third determining step;

a first setting step of setting a region for the detection objects determined to have high crowdedness in the fourth determining step;

a fifth determining step of determining that the detection objects corresponding to the region where the crowdedness of the detection objects is high in the first setting step are the road relating structural objects irrespective of relative speeds thereof; and

a sixth determining step of determining types of detection objects other than the detection objects determined to be the road relating structural objects in the fifth determining step based on relative speeds and positions of the detection objects and types thereof detected in the past.

10. The method for determining an object type according to claim 9, further comprising:

a first storing step of storing the types of the detection objects detected by the first determining step and the fifth determining step;

a second setting step of extracting only data of the detection objects determined to be the road relating structural objects from data stored in the first storing step, and of resetting the region for the detection objects high in crowdedness by use of the extracted data; and

a second storing step of storing the region set in the second setting step,

wherein the second setting step sets the region where the crowdedness of the objects is high by superposing a region where the crowdedness of the objects is high, which is set by the first setting step this time, and a region where the crowdedness of the objects was high, which was set by setting processing the previous time.

11. The method for determining an object type according to claim 10,

wherein the second setting step includes setting the region where the crowdedness of the objects is high after adding the detection objects determined to be the road relating structural objects in the region where the crowdedness thereof is high in the previous processing to a grouping result obtained this time.

12. The method for determining an object type according to claim 10,

wherein the second setting step includes determining that a detection object determined to be a running object in a region other than the region where the crowdedness of the objects was high by the past determination processing for an object type is determined to be the running object, even if the detection object is detected in the region where the crowdedness of the objects is high by a determination for an object type this time.

13. The method for determining an object type according to clam 10, further comprising:

a selecting step of obtaining a shape of a running lane in front of one'"'"'s running vehicle, of extracting only detection objects having widths equal to/less than a predetermined width and existing outside of the running lane from detection objects existing in a region determined to have road relating structural objects in the first setting step, and of selecting the detection objects having the width equal to/less than the predetermined width as the same road relating structural objects belonging to the region;

an operating step of operating to calculate distances from the detection objects to the running lane and of obtaining a center of gravities of the operated distances in a case where the number of the detection objects selected in the selecting step is equal to/more than a predetermined number; and

a third setting step of setting the region after the object determination in the second setting step at a position moved from the running lane to an outside thereof by a distance to the center of gravities, the center being obtained in the operating step.

14. The method for determining an object type according to claim 13, further comprising:

a seventh determining step of determining a group of detection objects to be an array of reflective plates provided along a road and of determining a line connecting points where the reflective plates provided along a road are detected to one another to be a shape of a road in front of one'"'"'s vehicle in a case where crowdedness of the detection objects in the region set in the third setting step is larger than a predetermined value when the region is set along a shape of the running lane.