JP2008143447A - Vehicle safety control device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle safety control device for actualizing a proper combination of radar output and image recognizing output. <P>SOLUTION: The vehicle safety control device comprises a radar unit 2 for detecting the position and relative speed of an object in front of a vehicle, a camera 3a for picking up an image in front of the vehicle, a processing region setting means for setting a processing region on the image obtained by the camera 3a, an object recognizing part 3b for recognizing the object on the image in the processing region, and an overlapping amount calculating part 4a for calculating the amount of overlapping the object with one's own vehicle in accordance with the detection result of the radar unit 2 and the recognition result of the object recognizing part 3b. The radar unit 2 and the object recognizing part 3b start detection and recognition, respectively, with the input of the same trigger signal generated by a trigger signal generating part 4b, and the overlapping amount calculating part 4a controls a safety device in accordance with the detection result of the radar unit 2 and the recognition result of the object recognizing part 3b synchronized by the trigger signal. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a vehicle safety control device.

In an image processing system for an image in front of a vehicle, an object in front of the host vehicle is detected by a radar and the distance to the object is detected, the front of the host vehicle is imaged by a camera, and an image is captured by the camera based on the distance data from the radar It is known that a processing area to be subjected to image recognition processing is set for a processed image, thereby reducing the amount of image processing (see, for example, Patent Document 1).
JP-A-9-264554

By the way, generally, the object detection processing time by the radar and the image recognition processing time for the image captured by the camera are different.
However, as described above, in the image recognition processing, since the processing area is set based on the distance data that is the radar object detection result output (hereinafter abbreviated as radar output), the object detection processing time by the radar, the camera If the image recognition processing times for the images picked up in are different, the time from the output of the radar output to the start of the image recognition processing is not constant.

  FIG. 7 is a time chart showing the processing timing and output timing of conventional radar detection and image recognition. As described above, the image recognition processing time t2 for the camera image is longer than the object detection processing time t1 by the radar, and both are set to a fixed time. In this time chart, when the detection result of the first radar detection is output as the radar output a1, the first image recognition processing is started based on the object position information of the radar output a1, and the image recognition result Is output as the first image recognition output b1. Similarly, when the detection result by the second radar detection is output as the radar output a2, the second image recognition processing is started based on the object position information of the radar output a2, and the image recognition result is It is output as the second image recognition output b2. This series of processing is repeated.

  As a result, as described above, the time from the output of the radar output to the start of the image recognition process is not constant. Therefore, the amount of deviation between the position of the object and the actual position of the object when the processing area is set at the start of the image recognition process varies. Conventionally, the processing area is set in anticipation of the variation of the shift amount. For this reason, it is necessary to set the processing area to a certain extent, but there is a problem in that the amount of image processing increases.

Also, when controlling a vehicle safety device using radar output and image recognition output, the safety device can be controlled based on a combination of radar output and image recognition output measured at different timings. There is sex.
For example, in the time chart of FIG. 7, the radar output a3 can be combined with the image recognition output b1 or the image recognition output b2, and the radar output a4 can be combined with the image recognition output b2 or the image recognition output b3. However, since the object information (position, relative speed, size) varies depending on the combination, there is a problem that the control content executed based on the object information also varies.
There is also an idea of combining radar output and image recognition output that are close in time, but in that case, it is necessary to wait until the slower output is output, which increases the waiting time and processing time. There is a problem of slowing down. For example, even if the radar output a4 and the image recognition output b2 are combined as a result, if it does not wait until the image recognition output b3 is output, it is determined which of the image recognition outputs b2 and b3 is closer in time. It cannot be judged.

  Therefore, the present invention provides a vehicle safety control device that can appropriately set a processing region and can appropriately combine a radar output and an image recognition output.

The vehicle safety control device according to the present invention employs the following means in order to solve the above-described problems.
The invention according to claim 1 is an object detection means (for example, a radar unit 2 in an embodiment to be described later) mounted on a vehicle to detect the position and relative speed of the object, and an imaging means mounted on the vehicle for imaging an object ( For example, a camera 3a in an embodiment to be described later, and a processing area setting unit (for example, S203 in the embodiment to be described later) that sets a processing area on an image obtained by the imaging unit based on a detection result of the object detection unit. ), An object recognition unit that recognizes an object on the image based on the luminance value of the pixel included in the processing region set by the processing region setting unit (for example, an object recognition unit 3b in an embodiment described later), Based on the detection result of the object detection means and the recognition result of the object recognition means, the vehicle safety device (for example, a motor driver in an embodiment described later) Safety device control means (for example, an overlap amount calculation unit 4a in an embodiment to be described later) that controls the unit 5, the actuator 6, and the airbag 7), and trigger signal generation means (for example, a trigger signal generated every predetermined time) A trigger signal generation unit 4b) in an embodiment to be described later, and the object detection means and the object recognition means start detection or recognition by input of the same trigger signal generated by the trigger signal generation means, The safety device control means controls the safety device based on the detection result of the object detection means and the recognition result of the object recognition means synchronized by the trigger signal.

  According to a second aspect of the present invention, in the first aspect of the present invention, the position of the object after the predetermined time is predicted based on the position and relative speed of the object detected by the object detection means according to the previous trigger signal. Object position prediction means (for example, step S105 in the embodiment described later), the processing area setting means sets a processing area on the image based on the position of the object predicted by the object position prediction means, The safety device control means controls the safety device based on a detection result detected by the object detection means in response to a current trigger signal and a recognition result recognized by the object recognition means in response to a current trigger signal. It is characterized by doing.

  According to a third aspect of the present invention, in the invention of the second aspect, the safety device control means uses the current position and relative speed of the object detected by the object detection means in response to the current trigger signal, and the current trigger signal. Accordingly, the contact determination between the vehicle and the object is performed based on the size of the object recognized by the object recognition means, and the safety device is controlled based on the determination result.

  The invention according to claim 4 is the generation interval of the trigger signal according to any one of claims 1 to 3 according to the output status of the object detection means or the object recognition means. The predetermined time is changed.

  The invention according to claim 5 is the invention according to claim 4, wherein the detection result is not output from the object detection means between the generation of the previous trigger signal and the generation of the current trigger signal, or the object recognition. When the recognition result is not output from the means, the predetermined time is changed to a large value.

  The invention according to claim 6 is the invention according to any one of claims 1 to 5, wherein the safety device is a brake (for example, an actuator 6 in an embodiment described later) provided in a vehicle or an airbag. (For example, an airbag 7 in an embodiment described later).

  According to the first aspect of the invention, the detection result by the object detection unit and the recognition result by the object recognition unit are synchronized by inputting the same trigger signal as the start timing of the object detection process and the object recognition process. And the safety device can be controlled based on an appropriate combination of these results.

According to the second aspect of the present invention, since the predetermined time that is the generation interval of the trigger signal is specified, the position of the object can be predicted with high accuracy. Further, since the processing area is set based on the position of the object predicted according to the previous trigger signal, the processing area can be set simultaneously with the start of the image recognition process. In addition, since the position prediction of the object is highly accurate, the processing region can be set appropriately.
In addition, since the safety device is controlled based on the combination of the object detection result by the object detection means according to the current trigger signal and the recognition result by the object recognition means according to the current trigger signal, the safety device is between the detection result and the recognition result. No shift on the time axis occurs.

  According to the invention of claim 3, the vehicle and the object are based on the combination of the high-accuracy position and relative speed detected by the object detection means and the high-precision object size recognized by the object recognition means. Can be determined, so that the determination accuracy of the contact possibility is improved.

  According to the invention which concerns on Claim 4, the generation | occurrence | production timing of a trigger signal can be set appropriately.

  According to the invention of claim 5, when the detection processing time by the object detection means or the recognition processing time by the object recognition means is increased, the predetermined time that is the trigger signal generation interval is increased. Thus, synchronization between the detection result by the object detection unit and the recognition result by the object recognition unit can be ensured.

  According to the invention which concerns on Claim 6, a brake and an airbag can be controlled appropriately.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a vehicle safety control device according to the present invention will be described below with reference to the drawings of FIGS.
As shown in the functional block diagram of FIG. 1, the vehicle safety control device 1 includes a radar unit (object detection means) 2, a camera unit 3, an electronic control device 4, a motor driver unit 5, an actuator 6, , And an airbag 7.

  The radar unit 2 includes a radar 2a and an object detection unit 2b. The radar 2a transmits an electromagnetic wave such as a laser beam or a millimeter wave toward the front in the traveling direction of the host vehicle, and when the transmitted electromagnetic wave is reflected by an object outside the host vehicle (for example, another vehicle), the radar 2a The reflected wave is received, the transmitted electromagnetic wave and the received electromagnetic wave (reflected wave) are mixed to generate a beat signal, and output to the object detection unit 2b.

  The object detection unit 2b calculates the position of the object reflecting the electromagnetic wave, the relative speed, the approximate size of the object, and the like based on the beat signal input from the radar 2a, and the calculated object information is electronically controlled. 4 is output. Further, the object detection unit 2 b calculates a predicted position of the object after a predetermined time based on the calculated relative speed, and outputs the predicted position information to the camera unit 3.

The camera unit 3 includes a camera (imaging unit) 3a including a CCD camera or a CMOS camera, and an object recognition unit (object recognition unit) 3b.
The camera 3a performs predetermined image processing such as filtering and binarization processing on each image of the outside world in the traveling direction of the own vehicle obtained by photographing, and generates image data composed of pixels of a two-dimensional array. And output to the object recognition unit 3b.

Image data is input from the camera 3a to the object recognition unit 3b, and predicted position information of the object is input from the object detection unit 2b of the radar unit 2. The object recognition unit 3b sets a region having a predetermined size (hereinafter referred to as a processing region) on the image data input from the camera 3a based on the input predicted position information. For example, as illustrated in FIG. 2, a rectangular processing region S <b> 2 having an area that can surround the preceding vehicle is set for the captured image S <b> 1 by the camera 3 a.
Further, the object recognizing unit 3b recognizes an object on the image based on the luminance value of the pixel included in the set processing region, and sends information such as the size and characteristics (shape, etc.) of the recognized object to the electronic control unit 4. Output to.

The electronic control unit 4 includes an overlap amount calculation unit (safety device control unit) 4a and a trigger signal transmission unit (trigger signal generation unit) 4b.
The overlap amount calculation unit 4 a includes object information (object position, relative speed, rough size of the object, etc.) detected by the object detection unit 2 b of the radar unit 2 and an object recognition unit 3 b of the camera unit 3. The detected object information (object size, characteristics, etc.) is input. Based on the input information and the like, the overlap amount calculation unit 4a calculates a time (hereinafter referred to as a predicted contact time) until the vehicle and the object are in contact with each other when the current running state is maintained. Estimate and calculate the amount of overlap between the vehicle and the object in the vehicle width direction after the predicted time (hereinafter referred to as overlap amount), and determine whether or not there is a possibility of contact between the vehicle and the object The determination result is output to the motor driver unit 5 and the airbag 7. Further, the overlap amount calculation unit 4a outputs an execution completion status signal to the trigger signal generation unit 4b when these processes are normally executed.

  The trigger signal generator 4b generates a trigger signal for synchronizing the start timing of the object detection process by the radar unit 2 and the image recognition process by the camera unit 3 at a constant generation interval Δt1 (seconds), and the generated trigger signal Are output to the radar 2a, the object recognition unit 3b, and the overlap amount calculation unit 4a. The trigger signal generation interval Δt1 can be changed, and is changed as necessary as will be described later.

The motor driver unit 5 includes a steering actuator control unit that controls the steering assist amount, a brake actuator control unit that controls the brake assist amount, and the like. The motor driver unit 5 automatically determines according to the determination result input from the overlap amount calculation unit 4a. The steering assist amount and the brake assist amount are controlled so as to avoid the occurrence of contact between the vehicle and the object or reduce the damage at the time of the contact occurrence, and a control signal corresponding to the assist amount is output to the actuator 6.
The actuator 6 includes a steering actuator, a brake actuator, and the like, and is driven in accordance with a control signal input from the motor driver unit 5, whereby the vehicle's steering control or deceleration control for avoiding contact or reducing damage or Both are executed. In this embodiment, the motor driver unit 5 and the actuator 6 constitute a first safety device.

  The airbag 7 includes a driver seat airbag, a passenger seat airbag, a side curtain airbag, and the like, and is controlled to be activated or deactivated according to the determination result input from the overlap amount calculation unit 4a. In this embodiment, the airbag 7 constitutes a second safety device.

  In the vehicle safety control device 1 configured as described above, the conventional problem caused by the difference between the object detection processing time by the radar unit 2 and the image recognition processing time by the camera unit 3, that is, an excessive processing area. In order to prevent a reduction in detection accuracy based on an inappropriate combination of the setting and the object detection result by the radar unit 2 and the image recognition result by the camera unit 3, the radar unit 2 and the camera unit 3 are provided with a trigger signal generation unit. When the trigger signal output simultaneously from 4b (that is, the same trigger signal) is received, the object detection process by the radar unit 2 is started and the image recognition process by the camera unit 3 is started. Added synchronization with image recognition output.

The processing procedure will be specifically described below.
First, the object detection process executed in the radar unit 2 will be described with reference to the flowchart of FIG.
First, in step S101, it is determined whether or not the trigger signal TSx is input from the trigger signal generator 4b. The trigger signal TSx in step S101, the trigger signal TSx in step S201 (see FIG. 4) in the image recognition process described later, and the trigger signal TSx in step S301 (see FIG. 5) in the overlap determination process are generated as trigger signals. It is assumed that the trigger signals are simultaneously output from the unit 4b, that is, the same trigger signal.
If the determination result in step S101 is “NO” (no TSx input), the execution of this routine is temporarily terminated.

If the determination result in step S101 is “YES” (with TSx input), the process proceeds to step S102, where an electromagnetic wave (transmission signal) is transmitted from the radar 2a toward the front of the vehicle. Thereby, the object detection process by the radar unit 2 is started.
Next, the process proceeds to step S103, where a reflected signal with a high risk of collision is selected from a plurality of reflected signals reflected by the object and received by the radar 2a, and this object is set as a target, and the position and relative velocity of the target are selected. The approximate size (hereinafter, these may be collectively referred to as object information detected by the radar unit 2) is calculated.

Next, the process proceeds to step S104, and a target movement trajectory is calculated based on the position change of the target and the relative speed calculated in step S103.
Next, proceeding to step S105, the predicted position of the target after the trigger signal generation interval Δt1 is calculated based on the relative speed calculated in step S103 and the movement trajectory calculated in step S104.
Next, the process proceeds to step S106, and the object information calculated in step S103 is output to the overlap amount calculation unit 4a of the electronic control unit 4, and the predicted position information of the target after Δt1 calculated in step S105 is stored in the camera unit 3. It outputs to the object recognition part 3b, and execution of this routine is once complete | finished.
In this embodiment, the object position prediction means is realized by the electronic control unit 4 executing the process of step S105.

Next, image recognition processing executed in the object recognition unit 3b of the camera unit 3 will be described with reference to the flowchart of FIG.
First, in step S201, it is determined whether or not the trigger signal TSx is input from the trigger signal generator 4b.
If the determination result in step S201 is “NO” (no TSx input), the execution of this routine is temporarily terminated.

If the determination result in step S201 is “YES” (with TSx input), the process proceeds to step S202, and image data captured by the camera 3a is captured.
Next, the process proceeds to step S203, and a processing region S2 is set in the captured image S1 based on the predicted target position calculated in step S105 in the previous object detection process executed with the previous trigger signal TSx-1 as the start timing. . That is, the process of step S105 in the previous object detection process predicts the position of the target when the current trigger signal TSx is received, and in step S203, the processing region S2 is set based on the predicted position of the target. To do.

In step S204, an edge pattern (object) on the processing region S2 is extracted based on the luminance value of the pixel included in the set processing region S2.
In step S205, it is determined whether the edge pattern extracted in step S204 is similar to the registered edge pattern. When this routine is executed for the first time, there is no registered edge pattern, so the determination result is NO.

If the determination result in step S205 is “YES” (there is a similar shape), the process proceeds to step S206, and the target size is calculated based on the extracted edge pattern.
In step S207, the extracted edge pattern is stored as a registered edge pattern.

On the other hand, if the determination result in step S205 is “NO” (no similarity), the process proceeds to step S208, and background separation processing is performed on the image data in the processing area S2.
In step S209, 1 is added to the count value n of the background separation counter (n ← n + 1). In step S210, it is determined whether the added count value n is greater than the threshold value N. .

  If the determination result in step S210 is “NO” (n ≦ N), the process returns to step S205, and it is determined whether or not the edge pattern after the background separation process has a similar shape to the registered edge pattern. When n = 2 or more and the process returns to step S205, the edge pattern after the previous (n−1) background separation process is the second registered edge pattern, and this time (n) after the background separation process. It is determined whether or not there is an edge pattern similar to the registered edge pattern stored in step S207 or the second registered edge pattern.

  If the determination result in step S210 is “YES” (n> N), the process proceeds to step S211, the count value of the background separation counter is reset to 0, the process proceeds to step S207, and the last (n = N) The edge pattern after background processing is stored as a registered edge pattern. In other words, by limiting the maximum number of repetitions of the background separation processing to N times, the image recognition processing time is limited so that the overlap determination processing described later is not hindered.

  It is also possible to change the time limit for the image recognition process by changing the maximum number of repetitions N according to the environmental conditions of the outside world. For example, it may take a long time for image recognition processing at night or in bad weather, and the output of object information by image recognition may be delayed. In such a case, the maximum number of repetitions N is changed to a large value and the image is processed. Increase the time limit for recognition processing. Then, the trigger signal generation interval Δt is changed according to the change in the time limit of the image recognition processing, and before the object information output from the radar unit 2 or the object information output from the camera unit 3 is output, the trigger signal is output. The next trigger signal is prevented from being transmitted from the generator 4b. For example, when the maximum number of repetitions N is increased to increase the time limit for image recognition processing, the trigger signal generation interval Δt is increased. In this way, it is possible to always ensure the synchronism between the output of object information by the object detection of the radar unit 2 and the output of object information by the image recognition of the object recognition unit 3b.

  As a specific method, for example, when object information is not output from the radar unit 2 between the generation of the previous trigger signal and the generation of the current trigger signal, or when object information is not output from the camera unit 3. The maximum number of repetitions N is increased and the trigger signal generation interval Δt is increased.

After performing the process of step S207, the process proceeds to step S212, the target size information calculated in step S206 is output to the overlap amount calculation unit 4a of the electronic control unit 4, and the execution of this routine is temporarily terminated.
In this embodiment, the processing area setting means is realized by the electronic control unit 4 executing the process of step S203.

Next, the overlap determination process executed in the overlap amount calculation unit 4a of the electronic control unit 4 will be described with reference to the flowchart of FIG.
First, in step S301, it is determined whether or not the trigger signal TSx is input from the trigger signal generator 4b.
If the determination result in step S301 is “NO” (no TSx input), the execution of this routine is temporarily terminated.

In step S302, it is determined whether object information is input from the radar unit 2. The object information from the radar unit 2 here is object information (target position, relative speed, size) obtained by the radar unit 2 executing object detection processing using the trigger signal TSx as a start trigger.
If the determination result in step S302 is “NO” (no input), the execution of this routine is temporarily terminated.

  If the determination result in step S <b> 302 is “YES” (with input), the process proceeds to step S <b> 303 to determine whether object information has been input from the camera unit 3. The object information from the camera unit 3 here is object information (target size) obtained by the camera unit 3 executing image recognition processing using the trigger signal TSx as a start trigger.

  If the determination result in step S303 is “YES” (with input), the process proceeds to step S304, where the target position and relative speed in the object information input from the radar unit 2 and the object input from the camera unit 3 are detected. Based on the size of the target as information, the overlap amount is estimated and calculated, and it is determined whether or not there is a possibility of contact between the own vehicle and the object. Thus, the possibility of contact between the vehicle and the object can be determined based on the position and relative velocity detected with high accuracy by the radar unit 2 and the size of the target detected with high accuracy by the camera unit 3. Since it can do, the determination accuracy of contact possibility improves.

  On the other hand, if the determination result in step S303 is “NO” (no input), the process proceeds to step S305, and the overrun is based only on the object information (target position, relative speed, size) from the radar unit 2. The lap amount is estimated and calculated, and it is determined whether or not there is a possibility of contact between the vehicle and the object. Thereby, even when the size of the target is not detected by the camera unit 3, it is possible to determine the possibility of contact between the own vehicle and the object.

FIG. 6 is a time chart showing processing timing and output timing of object detection processing by the radar unit 2 and image recognition processing by the camera unit 3 in the vehicle safety control device 1 of the embodiment.
The trigger signal TS is transmitted at regular intervals Δt1. The radar unit 2 starts the object detection process with the trigger signal as the start timing, and the camera unit 3 starts the image recognition process at the same time. Since the object detection processing time t1 of the radar unit 2 is longer than the image recognition processing time t2 of the camera unit 3, the object information (position, relative speed, approximate size) detected by the object detection processing is used as the radar output R1. The predicted position information of the object after the trigger signal generation interval Δt is output from the radar unit 2 as the radar output R2 before the image recognition output C of the image recognition process.

  The radar output R2 is a predicted position of the object at the start of the image recognition process (hereinafter abbreviated as the next image recognition process) that is started by the next trigger signal, and the next image recognition process is performed based on the predicted position. Therefore, the processing area can be set simultaneously with the start of the next image recognition process. Further, since the generation interval Δt of the trigger signal is specified, the position prediction can be performed with high accuracy, and the processing region can be set appropriately.

  The object information (object size) detected by the image recognition processing is output from the camera unit 3 as the image recognition output C with a delay from the radar outputs R1 and R2 of the radar unit 2. After the image recognition output C is output, the overlap amount between the vehicle and the object is calculated based on the radar output R1 and the image recognition output C.

Therefore, the radar output R1 and the image recognition output C can always be synchronized, and the radar output R1 of the object detection process started at the same timing and the image recognition output C of the image recognition process can be combined. The combination will not shift. For example, in this time chart, radar output R1 _-1 and image recognition output C _-1 , radar output R1 _-2 and image recognition output C _-2 , radar output R1 _-3 and image recognition output C _-3 , radar output R1 _-4 and the image recognition output C _-4 may be combined.
In other words, a process on the same event can be performed in parallel without causing a shift on the time axis between the object detection process by the radar unit 2 and the image recognition process by the camera unit 3.
As a result, the overlap amount can be estimated and calculated based on the combination of the radar output R1 and the image recognition output C that are always synchronized, and it can be determined whether or not there is a possibility of contact between the vehicle and the object.

It is a functional block diagram in the Example of the vehicle safety control apparatus which concerns on this invention. It is explanatory drawing of the process area | region in an image recognition process. It is a flowchart which shows the object detection process in an Example. It is a flowchart which shows the image recognition process in an Example. It is a flowchart which shows the overlap determination process in an Example. It is a time chart which shows the process timing and output timing of the object detection process and image recognition process in the vehicle safety control apparatus of an Example. It is a time chart which shows the processing timing and output timing of the conventional object detection processing and image recognition processing.

Explanation of symbols

1 Vehicle safety control device 2 Radar unit (object detection means)
3a Camera (imaging means)
3b Object recognition unit (object recognition means)
4a Overlap amount calculation unit (safety device control means)
4b Trigger signal generator (trigger signal generator)
5 Motor driver unit (safety device)
6 Actuator (safety device)
7 Airbag (safety device)
S105 Object position prediction means S203 Processing area setting means

Claims (6)

Object detection means for detecting the position and relative speed of an object mounted on a vehicle;
Imaging means mounted on the vehicle for imaging an object;
Processing area setting means for setting a processing area on an image obtained by the imaging means based on a detection result of the object detection means;
Object recognition means for recognizing an object on an image based on a luminance value of a pixel included in the processing area set by the processing area setting means;
Safety device control means for controlling the safety device of the vehicle based on the detection result of the object detection means and the recognition result of the object recognition means;
Trigger signal generating means for generating a trigger signal every predetermined time; and
The object detection means and the object recognition means start detection or recognition by the input of the same trigger signal generated by the trigger signal generation means, and the safety device control means is synchronized with the trigger signal. A vehicle safety control device that controls the safety device based on a detection result of a detection unit and a recognition result of the object recognition unit. An object position prediction means for predicting the position of the object after the predetermined time based on the position and relative speed of the object detected by the object detection means in response to the previous trigger signal;
The processing area setting means sets a processing area on the image based on the position of the object predicted by the object position prediction means,
The safety device control means controls the safety device based on a detection result detected by the object detection means in response to a current trigger signal and a recognition result recognized by the object recognition means in response to a current trigger signal. The vehicle safety control device according to claim 1.   The safety device control means is based on the position and relative velocity of the object detected by the object detection means according to the current trigger signal and the size of the object recognized by the object recognition means according to the current trigger signal. The vehicle safety control device according to claim 2, wherein contact determination between the vehicle and the object is performed and the safety device is controlled based on the determination result.   The said predetermined time which is the generation | occurrence | production interval of the said trigger signal is changed according to the output condition of the said object detection means or the said object recognition means, The Claim 1 characterized by the above-mentioned. Vehicle safety control device.   If the detection result is not output from the object detection unit or the recognition result is not output from the object recognition unit between the generation of the previous trigger signal and the generation of the current trigger signal, the predetermined time is changed to a large value. The vehicle safety control device according to claim 4.   The vehicle safety control device according to any one of claims 1 to 5, wherein the safety device is a brake or an airbag provided in the vehicle.

Images