JP2017021435A - Vehicle drive assist apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle drive assist apparatus capable of selecting a control target at timing matching driver's feeling without complicated setting operation or the like.SOLUTION: A control unit sets (S102) a selection region at a region width different among types of a three-dimensional objects recognized (S101) in front of a subject vehicle along a subject vehicle travel road; and corrects (S105) a three-dimensional object width of the three-dimensional object selected (S103) in response to a distance from the subject vehicle so that the three-dimensional object width can converge into a fixed value set in advance (S104) at a distant location. When the three-dimensional object assumed to have the corrected three-dimensional object width overlaps the selection region (YES in S106), the three-dimensional object is extracted (S107) as a control target candidate and the control target is selected (S110) from among the control target candidates extracted (S109).SELECTED DRAWING: Figure 7

Description

  The present invention relates to a vehicle driving support device that determines the possibility of contact with a front three-dimensional object detected by a stereo camera, a laser radar, or the like and performs driving support control.

  In recent years, various driving assistance devices have been proposed and put into practical use in vehicles, in which a driving environment is photographed by a camera mounted on the vehicle, a three-dimensional object is recognized, and the possibility of a collision with the host vehicle is estimated. Yes. In an apparatus for estimating the possibility of a collision between the front three-dimensional object and the host vehicle, a front three-dimensional object such as a preceding vehicle existing on the predicted traveling path (own vehicle traveling path) of the own vehicle is detected at an appropriate timing. There is a need.

  As a technique for performing such forward three-dimensional object detection, for example, in Patent Document 1, a radar vehicle detects a preceding vehicle in the traveling direction of the host vehicle, and sets a determination region (selection) set along the predicted course of the host vehicle. Calculate the overlap amount where the preceding vehicle overlaps the (region), and select it as the control target that may cause the host vehicle to contact the preceding vehicle when the overlap amount exceeds the predetermined value Techniques to do this are disclosed.

JP 2004-330950 A

  By the way, since there are various types of solid objects recognized in front of the host vehicle, there is a possibility that the driver may feel uncomfortable when selecting a control target by uniform processing for each solid object. . For example, even when a vehicle is present in front of the host vehicle, the recognized three-dimensional object width (vehicle width) differs depending on whether the vehicle is a normal vehicle or a large truck. The timing for starting the lap in the selected area on the road is different, and as a result, the timing for selecting the control target is different. Moreover, about a pedestrian etc., selecting as a control object at a timing earlier than a vehicle etc., and starting control agree | coincide by a driver's feeling. Therefore, in this type of driving support device, it is desirable to appropriately tune the timing for selecting the recognized three-dimensional object as a control target according to the type of the three-dimensional object.

  On the other hand, there is a case where the three-dimensional object width cannot be accurately recognized in the distance from the host vehicle. For example, when a curve exists in front of the host vehicle, a vehicle or the like located far away may be erroneously recognized as a vehicle width (solid object width) not only on the back but also on the side. Therefore, in selecting the control target, it is desirable to perform tuning in consideration of such misrecognition of the three-dimensional object width. On the other hand, three-dimensional objects that are far away from the host vehicle have less need for emergency evacuation control than those in the vicinity, and the timing for selecting such a three-dimensional object as a control target is strict. Tuning or the like causes complicated setting work.

  The present invention has been made in view of the above circumstances, and provides a driving support device for a vehicle that can select a control target at a timing that matches a driver's feeling without performing complicated setting work or the like. Objective.

  A vehicle driving support device according to an aspect of the present invention detects a three-dimensional object ahead of the host vehicle, and recognizes a three-dimensional object information including a type and a three-dimensional object width for each detected three-dimensional object; The vehicle traveling path estimation means for estimating the traveling path of the vehicle, the selection area setting means for setting a selection area having a different area width for each type of the three-dimensional object along the traveling path of the vehicle, and the three-dimensional object width. The three-dimensional object width correcting means for correcting the three-dimensional object width according to the distance from the host vehicle so as to converge to the set three-dimensional object width at a distance, and the three-dimensional object assumed to have the corrected three-dimensional object width are Control object candidate extraction means for extracting the three-dimensional object as a control object candidate when wrapped with a selection area, and control object selection means for selecting a control object from the extracted control object candidates It is.

  According to the vehicle driving support device of the present invention, it is possible to select a control target at a timing that matches the feeling of the driver without performing complicated setting work or the like.

Schematic configuration diagram of the driving support device Explanatory drawing which shows the selection field when a solid thing is a standard car Explanatory drawing which shows the selection field when a solid thing is a large-sized car Explanatory drawing which shows the selection area when a solid thing is a pedestrian Explanatory drawing which shows the selection area when a solid thing is other solid objects Map showing an example of the relationship between the forward gaze distance and the coefficient Flow chart showing control target selection routine

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. The drawings relate to an embodiment of the present invention, FIG. 1 is a schematic configuration diagram of a driving support device, FIG. 2 is an explanatory diagram showing a selection region when a three-dimensional object is a normal vehicle, and FIG. FIG. 4 is an explanatory diagram showing a selection area when a solid object is a pedestrian, FIG. 5 is an explanatory diagram showing a selection area when the solid object is another solid object, FIG. 6 is a map showing an example of the relationship between the forward gaze distance and the coefficient, and FIG. 7 is a flowchart showing a control object selection routine.

  In FIG. 1, reference numeral 1 denotes a vehicle such as an automobile (own vehicle), and a driving support device 2 is mounted on the vehicle 1. The driving support device 2 includes, for example, a stereo camera 3, a stereo image recognition device 4, a control unit 5, and the like.

  The host vehicle 1 is connected to a vehicle speed sensor 11 that detects the host vehicle speed V, a yaw rate sensor 12 that detects the yaw rate γ, a main switch 13 that performs ON-OFF switching of various functions of driving support control, and the like, and a steering wheel. A steering angle sensor 14 that detects the steering angle θst and that is opposed to the steering shaft, an accelerator opening sensor 15 that detects an accelerator pedal depression amount (accelerator opening) θacc by the driver, and the like are provided.

  The stereo camera 3 is constituted by a set of CCD cameras using a solid-state imaging device such as a charge coupled device (CCD) as a stereo image optical system. These left and right CCD cameras are each mounted at a certain distance in front of the ceiling in the vehicle interior, take a stereo image of an object outside the vehicle from different viewpoints, and output image data to the stereo image recognition device 4. In the following description, one image (for example, the right image) of the stereo images is referred to as a reference image, and the other image (for example, the left image) is referred to as a comparative image.

  First, the stereo image recognition device 4 divides the reference image into small regions of 4 × 4 pixels, compares the luminance or color pattern of each small region with the comparison image, finds the corresponding region, and creates the entire reference image. Find the distance distribution across. Further, the stereo image recognition device 4 examines the luminance difference between each pixel on the reference image and an adjacent pixel, extracts those whose luminance difference exceeds a threshold value as an edge point, and extracts the extracted pixel (edge A distance image (a distribution image of edge points having distance information) is generated by giving distance information to (point). Then, the stereo image recognition device 4 performs a well-known grouping process on the generated distance image and compares it with a pre-stored three-dimensional frame (window), so that a white line and a side wall in front of the vehicle Recognize three-dimensional objects.

  Here, the white line to be recognized in the present embodiment is, for example, a white line (single white line) configured with a single lane line or a white line (double line with an auxiliary lane line inside the lane line) (White line) is a general term for lines extending on the road and defining the vehicle traveling lane, and the form of each line includes a solid line, a broken line, and the like, and further includes a yellow line and the like. In the white line recognition of the present embodiment, even if the white line that exists on the road is a double white line, it is recognized by approximating it with a single approximate line on each of the left and right sides.

  Further, the stereo image recognition device 4 can detect the three-dimensional object in a three-dimensional object recognition based on, for example, a comparison result with a window or the like. And the like, and the solid object width Wo is recognized. When the stereo image recognition device 4 determines that the three-dimensional object detected in the distance cannot be distinguished as a normal vehicle, a large vehicle, a pedestrian, or the like, the stereo image recognition device 4 recognizes the three-dimensional object as another three-dimensional object. Thus, in this embodiment, the stereo image recognition apparatus 4 implement | achieves the function as a front three-dimensional object recognition means with the stereo camera 3. FIG.

  The control unit 5 receives driving environment information in front of the host vehicle 1 recognized by the stereo image recognition device 4. Further, the vehicle speed V from the vehicle speed sensor 11, the yaw rate γ from the yaw rate sensor 12, and the like are input to the control unit 5 as travel information of the host vehicle 1, and the operation input information by the driver from the main switch 13. An operation signal, a steering angle θst from the steering angle sensor 14, an accelerator opening θacc from the accelerator opening sensor 15, and the like are input.

  When traveling environment information including three-dimensional object information ahead of the host vehicle 1 is input from the stereo image recognition device 4, the control unit 5 selects a three-dimensional object to be controlled by various driving support controls. When selecting the three-dimensional object, the control unit 5 first estimates the own vehicle traveling path, and sets a three-dimensional object selection area As along the estimated own vehicle traveling path. Then, the control unit 5 extracts the three-dimensional object wrapped in the set selection area As, and selects the three-dimensional object closest to the host vehicle 1 from the extracted three-dimensional object as a control target. In this case, more specifically, the control unit 5 sets a selection region As having a different width for each type of three-dimensional object, for example, as shown in FIGS. In addition, in order to reduce the influence of the detection error of the three-dimensional object width Wo in the distance of the host vehicle 1, the control unit 5 moves away from the fixed value Wf (for example, Wf = 2m) in which the three-dimensional object width Wo is set in advance. The three-dimensional object width Wo is corrected according to the distance from the host vehicle 1 so as to converge. Then, when the three-dimensional object assumed to have the corrected three-dimensional object width Wc is overlapped with the selected region As, the control unit 5 extracts the three-dimensional object as a control target candidate. As described above, in the present embodiment, the control unit 5 realizes each function as the own vehicle traveling path estimation means, the selection area setting means, the three-dimensional object width correction means, the control target candidate extraction means, and the control target selection means. To do.

  Further, for example, when an execution of an ACC (Adaptive Cruise Control) function which is one of the functions of the driving support control is instructed through the operation of the main switch 13 by the driver, the control unit 5 is controlled on the own vehicle traveling path. It is determined whether or not a preceding vehicle (ordinary vehicle, large vehicle, etc.) is selected.

  As a result, when the preceding vehicle is not selected as the control target, the constant speed traveling control is performed to maintain the vehicle speed V of the host vehicle 1 at the set vehicle speed set by the driver through the opening / closing control (engine output control) of the throttle valve 16. Is executed.

  On the other hand, when a preceding vehicle is selected as a control target and the vehicle speed of the preceding vehicle is equal to or lower than the set vehicle speed, follow-up running control is performed in which the following distance is converged to the target inter-vehicle distance. In this follow-up running control, the control unit 5 basically converges the inter-vehicle distance to the target inter-vehicle distance through opening / closing control of the throttle valve 16 (engine output control).

  Furthermore, when the control unit 5 determines that there is a high possibility of a collision with the control target (preceding vehicle, pedestrian, or other three-dimensional object such as a side wall), the control unit 5 warns the driver by an alarm or a meter display. After the prompt, if there is no operation for collision avoidance, collision avoidance control is executed. That is, the control unit 5 performs the collision avoidance control with the control object using, for example, the control of the output hydraulic pressure from the active booster 17 (the automatic intervention control of the brake).

  Next, the control object selection process executed in the control unit 5 will be described in detail according to the control object selection routine shown in FIG.

  This routine is repeatedly executed every set time. When the routine is started, the control unit 5 first starts the operation based on the driving environment information input from the stereo image recognition device 4 in step S101. It is checked whether or not there is a three-dimensional object recognized in front of the vehicle 1.

  If it is determined in step S101 that there is no three-dimensional object in front of the host vehicle 1, the control unit 5 directly exits the routine.

  On the other hand, when it is determined in step S101 that there is a three-dimensional object in front of the host vehicle 1, the control unit 5 proceeds to step S102 and estimates the host vehicle traveling path. That is, for example, the control unit 5 estimates the traveling direction of the host vehicle 1 based on the host vehicle speed V detected by the vehicle speed sensor 11 and the yaw rate γ detected by the yaw rate sensor 12, and based on the host vehicle traveling direction. Estimate the travel route of the vehicle.

  In step S103, the control unit 5 selects any one of the three-dimensional objects recognized in front of the host vehicle 1.

  In subsequent step S104, the control unit 5 sets a selection region corresponding to the type of the three-dimensional object currently selected along the own vehicle traveling path estimated in step S102. Specifically, for example, the control unit 5 sets different selection areas As depending on whether the currently selected three-dimensional object is a normal car, a large vehicle, a pedestrian, or another three-dimensional object. To do.

  For example, as shown in FIG. 2, when the currently selected three-dimensional object is a normal vehicle, the control unit 5 uses a preset width W1 as a reference in a section of a predetermined forward gaze distance Zt from the own vehicle 1. The area to be the width is set as the selection area As1 to be controlled. In this case, for example, a region having a uniform width W1 along the own vehicle traveling path is set as the selection region As1 for the ordinary vehicle. The forward gaze distance Zt is, for example, a distance that is variably set according to the current host vehicle speed V, and is selected based on a product of a preset front gaze time T and the host vehicle speed V (Zt = T × V). Is done.

  Further, for example, as shown in FIG. 3, when the currently selected three-dimensional object is a large vehicle, the control unit 5 sets a predetermined width W2 in the section of the predetermined forward gaze distance Zt from the own vehicle 1. Is set as the control target selection region As2. In this case, the reference width W2 for the large vehicle is narrower than the reference width W1 for the normal vehicle in consideration of the fact that the solid object width Wo of the large vehicle is larger than the solid object width Wo of the normal vehicle. Is set. The selection region As2 for the large vehicle is, for example, a region that extends toward the far side with the reference width W2 as the minimum width, and a predetermined distance (for example, half the front gaze distance (Zt / 2)). ) A region that coincides with the selected region As1 for ordinary vehicles is set along the own vehicle traveling path.

  For example, as shown in FIG. 4, when the currently selected three-dimensional object is a pedestrian, the control unit 5 sets a predetermined width W3 in a section of a predetermined forward gaze distance Zt from the own vehicle 1. Is set as a control target selection region As3. In this case, the reference width W3 intended for pedestrians is set wider than the reference width W1 intended for ordinary vehicles in order to select pedestrians as control targets at an earlier timing than ordinary vehicles and the like. As the selection area As3 for the pedestrian, for example, the reference width W3 is an area that is reduced toward the far side with the maximum width, and a predetermined distance (for example, a half value (Zt / 2) of the forward gaze distance) ) A region that coincides with the selected region As1 for ordinary vehicles is set along the own vehicle traveling path.

  Further, for example, as shown in FIG. 5, when the currently selected three-dimensional object is another three-dimensional object, the control unit 5 is set in advance in a section of a predetermined forward gaze distance Zt from the own vehicle 1. A region having the width W4 as a reference width is set as a selection region As4 to be controlled. In this case, since the reference width W4 for other three-dimensional objects is selected as a control object at a timing earlier than that of a normal vehicle and later than that of a pedestrian, the reference width W4 targets a normal vehicle. It is set wider than the reference width W1 and narrower than the reference width W3 intended for pedestrians. The selection region As4 for other three-dimensional objects is, for example, a region that is reduced toward the far side with the reference width W4 as the maximum width, and a predetermined distance (for example, half the forward gaze distance (Zt / 2 )) A region that coincides with the selected region As1 for ordinary vehicles is set along the own vehicle traveling path.

When proceeding from step S104 to step S105, the control unit 5 corrects the three-dimensional object width Wo of the currently selected three-dimensional object, for example, by weighted average calculation using the following equation (1). That is, when the corrected three-dimensional object width is Wc, the control unit 5
Wc = (Wo · α + Wf · (100−α)) / 100 (1)
Thus, the three-dimensional object width Wo is corrected.

  Here, α in the equation (1) is a coefficient that changes according to the distance from the vehicle 1 to the three-dimensional object, and is set with reference to a preset map or the like. For example, as shown in FIG. 6, the coefficient α is linearly decreased according to the ratio of the distance z from the host vehicle 1 to the three-dimensional object in the forward gaze distance Zt. It is set to be “0” when the distance z to the object coincides with the forward gaze distance Zt.

  When the process proceeds from step S105 to step S106, the control unit 5 wraps with the selected area As set in step S104 when it is assumed that the width of the currently selected three-dimensional object is the corrected three-dimensional object width Wc. Check if it is in a state.

  If it is determined in step S106 that at least a part of the three-dimensional object is wrapped in the selection area As, the control unit 5 proceeds to step S107 and extracts the currently selected three-dimensional object as a control target candidate. Then, the process proceeds to step S109.

  On the other hand, if it is determined in step S106 that the three-dimensional object is not in the state of being wrapped in the selected region As, the control unit 5 proceeds to step S108, and after excluding the currently selected three-dimensional object from the control target candidates, The process proceeds to S109.

  When the process proceeds from step S107 or step S108 to step S109, the control unit 5 determines whether or not the determination process in step S106 has been performed on all three-dimensional objects in front of the host vehicle 1 that are currently recognized by the stereo image recognition device 4. Check out.

  If it is determined in step S109 that determination processing for all three-dimensional objects has not yet been performed, the control unit 5 returns to step S103.

  On the other hand, if it is determined in step S109 that determination processing has been performed for all three-dimensional objects, the control unit 5 proceeds to step S110, and the control target candidate closest to the host vehicle 1 from the currently extracted control target candidates. Is selected as a control target, and then the routine is exited.

  According to such embodiment, while setting the selection area | region of a different area | region width for every classification of the solid object recognized ahead of the own vehicle 1 along the own vehicle advancing path, the solid object width Wo of the recognized three-dimensional object. 3D object width Wo is corrected according to distance z from host vehicle 1 so that it converges to a preset fixed value Wf in the distance, and the 3D object assumed to have corrected 3D object width Wc is the selected region. The three-dimensional object is extracted as a control target candidate when it is wrapped, and by selecting a control target from the extracted control target candidates, it matches the feeling of the driver without performing complicated setting work etc. The control target can be selected at the timing.

  That is, the actual three-dimensional object width Wo can be recognized with high accuracy, and the three-dimensional object width Wo actually recognized in the vicinity of the host vehicle 1 is likely to require emergency evacuation control. By extracting control target candidates based on the solid object width Wc that is strongly reflected and the selection area As that has been tuned for each type of three-dimensional object, at a timing that matches the feeling of the driver A control target can be selected. On the other hand, in the distance of the own vehicle 1, the three-dimensional object width Wo is corrected without performing complicated setting work by correcting the three-dimensional object width so that the three-dimensional object width Wo converges to a preset fixed value Wf. When the recognition error is large, compensation or the like can be realized, and the control target can be selected at a timing without any sense of incongruity.

  In this case, by setting the fixed value Wf for converging the three-dimensional object width Wo to a value common to various other three-dimensional objects, it is possible to further simplify the setting work for the vehicle 1 far away.

  Further, by converging the area width of the selected area set for each type to a common area width (for example, the area width W1 set corresponding to the ordinary vehicle) in the distance of the host vehicle 1, the three-dimensional object width Wo In combination with the correction, a three-dimensional object located far away can be extracted as a control target candidate at a timing without a sense of incongruity by a simplified setting.

  In addition, this invention is not limited to each embodiment described above, A various deformation | transformation and change are possible, and they are also in the technical scope of this invention. For example, in the above-described embodiment, an example of a configuration for recognizing three-dimensional object information based on an image captured by a stereo camera has been described. However, the present invention is not limited to this, and for example, a monocular camera and a millimeter It is also possible to recognize the three-dimensional object information based on information acquired by a combination with a wave radar or a laser radar. Of course, the types of three-dimensional objects to be classified are not limited to those described above.

1 ... Vehicle (own vehicle)
2 ... Driving support device 3 ... Stereo camera (front three-dimensional object recognition means)
4 ... Stereo image recognition device (front three-dimensional object recognition means)
5... Control unit (own vehicle traveling path estimation means, selection area setting means, three-dimensional object width correction means, control object candidate extraction means, control object selection means)
DESCRIPTION OF SYMBOLS 11 ... Vehicle speed sensor 12 ... Yaw rate sensor 13 ... Main switch 14 ... Steering angle sensor 15 ... Accelerator opening sensor 16 ... Throttle valve 17 ... Active booster

Claims (3)

A front three-dimensional object recognition means for detecting a three-dimensional object in front of the host vehicle and recognizing three-dimensional object information including a type and a three-dimensional object width for each detected three-dimensional object;
Own vehicle traveling path estimation means for estimating the own vehicle traveling path;
A selection area setting means for setting a selection area of a different area width for each type of the three-dimensional object along the own vehicle traveling path;
Three-dimensional object width correcting means for correcting the three-dimensional object width according to the distance from the host vehicle so that the three-dimensional object width converges to a preset fixed value at a distance;
Control object candidate extraction means for extracting the three-dimensional object as a control object candidate when the three-dimensional object assumed to have a corrected three-dimensional object width is wrapped with the selected region;
A vehicle driving support apparatus, comprising: a control object selection unit that selects a control object from the extracted control object candidates.   The vehicle driving support device according to claim 1, wherein the fixed value is a value common to the various three-dimensional objects.   3. The vehicle according to claim 1, wherein the selection area setting unit sets the setting area for each type so that the area width converges to a common area width in the distance of the host vehicle. Driving assistance device.

Images