Cruise control device and cruise control method

Abstract

To provide a travel control device that can perform appropriate brake control.SOLUTION: A travel control device according to one embodiment of the present disclosure includes: a generation unit that generates gradient data by detecting the gradient of pixel values in a second image region set according to a first image region including an image of a preceding vehicle in a captured image; a comparison unit that compares first gradient data with second gradient data generated as gradient data by the generation unit based on captured images different from each other; and a control unit that can suppress brake control of a vehicle based on a comparison result of the comparison unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a driving control device and a driving control method for controlling driving of a vehicle. [Background technology]

[0002] Among image processing devices mounted on vehicles such as automobiles, there are those that detect an image of water vapor in a captured image obtained by an image sensor (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-186668 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the meantime, there is a technology for vehicles to detect a preceding vehicle based on a captured image and perform driving control according to the detected preceding vehicle. When a vehicle is running following a preceding vehicle, if a smoke-like substance such as water vapor or exhaust gas is generated near the preceding vehicle, it is desirable to appropriately perform brake control of the vehicle.

[0005] It is desirable to provide a cruise control device and a cruise control method that can appropriately control the brakes of a vehicle. [Means for solving the problem]

[0006] A cruise control device according to an embodiment of the present disclosure includes a generating unit, a comparing unit, and a control unit. The generating unit is configured to generate gradient data by detecting a gradient of pixel values ​​in a second image area in a captured image that is set according to a first image area including an image of a leading vehicle. The comparing unit is configured to generate gradient data by detecting a gradient of pixel values ​​in a second image area in a captured image that is set according to a first image area including an image of a leading vehicle. Imaged at timeThe control unit is configured to compare the first gradient data and the second gradient data generated as gradient data by the generation unit based on the captured image. , so as not to suddenly stop the vehicle. The vehicle brake control is suppressed.

[0007] A cruise control method according to an embodiment of the present disclosure includes: generating gradient data by detecting a gradient of pixel values ​​in a second image area set in accordance with a first image area including an image of a leading vehicle in a captured image; Imaged at time comparing first gradient data and second gradient data generated as gradient data based on the captured image; and based on a comparison result of the first gradient data and the second gradient data, To avoid sudden stops of your vehicle, and suppressing brake control of the host vehicle. Effect of the Invention

[0008] According to the cruise control device and the cruise control method according to an embodiment of the present disclosure, brake control can be performed appropriately. [Brief description of the drawings]

[0009] [Figure 1] 1 is a block diagram illustrating an example configuration of a driving control device according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is an explanatory diagram illustrating an example of a stereo image shown in FIG. [Diagram 3] 2 is an explanatory diagram illustrating an example of an operation of the three-dimensional object detection unit illustrated in FIG. 1. [Figure 4] 2 is an explanatory diagram illustrating an operation example of a vehicle search unit illustrated in FIG. 1. [Diagram 5] 2 is an explanatory diagram illustrating an example of an operation of the gradient data generating unit 32 shown in FIG. 1. [Figure 6] 1. FIG. 4 is another explanatory diagram illustrating an example of an operation of the gradient data generating unit 32 shown in FIG. [Figure 7A] 2 is a flowchart illustrating an example of an operation of the driving control device illustrated in FIG. 1. [Figure 7B]10 is another flowchart illustrating an example of an operation of the driving control device illustrated in FIG. [Figure 8A] 10 is a flowchart illustrating an example of an operation of a driving control device according to a modified example. [Figure 8B] 10 is another flowchart illustrating an example of an operation of the driving control device according to the modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

[0011] <Embodiment Mode> [Configuration example] 1 shows an example of the configuration of a cruise control device (cruise control device 1) according to an embodiment. Note that a cruise control method according to an embodiment of the present disclosure is embodied in this embodiment, and will be described together. The cruise control device 1 includes a stereo camera 11 and a processing unit 20. The cruise control device 1 is mounted on a vehicle 10 such as an automobile.

[0012] The stereo camera 11 is configured to capture an image of the front of the vehicle 10, thereby generating a pair of images (a left image PL and a right image PR) having a parallax between them. The stereo camera 11 has a left camera 11L and a right camera 11R. Each of the left camera 11L and the right camera 11R includes a lens and an image sensor. In this example, the left camera 11L and the right camera 11R are disposed in the vehicle 10, near the upper part of the windshield of the vehicle 10, and spaced apart by a predetermined distance in the width direction of the vehicle 10. The left camera 11L and the right camera 11R perform imaging operations in synchronization with each other. The left camera 11L generates a left image PL, and the right camera 11R generates a right image PR. The left image PL and the right image PR constitute a stereo image PIC. The stereo camera 11 performs imaging operations at a predetermined frame rate (for example, 60 [fps]) to generate a series of stereo images PIC.

[0013] 2 shows an example of a stereo image PIC, where (A) shows an example of a left image PL and (B) shows an example of a right image PR. In this example, a leading vehicle 90 is traveling ahead of the vehicle 10 on a road on which the vehicle 10 is traveling. The left camera 11L captures an image of the leading vehicle 90 to generate a left image PL, and the right camera 11R captures an image of the leading vehicle 90 to generate a right image PR. The stereo camera 11 is adapted to generate a stereo image PIC including such a left image PL and a right image PR.

[0014] The processing unit 20 (FIG. 1) is configured to recognize the environment ahead of the vehicle 10 based on the stereo image PIC supplied from the stereo camera 11, and to control the traveling of the vehicle 10 based on the recognition result. The processing unit 20 is configured, for example, with a CPU (Central Processing Unit) that executes a program, a RAM (Random Access Memory) that temporarily stores processing data, and a ROM (Read Only Memory) that stores the program. The processing unit 20 has a distance image generating unit 21, a three-dimensional object detecting unit 22, a vehicle searching unit 23, a smoke-like object detecting unit 30, and a traveling control unit 25.

[0015] The distance image generating unit 21 is configured to generate a distance image PZ by performing a predetermined image processing including a stereo matching process and a filtering process based on the left image PL and the right image PR included in the stereo image PIC. The distance image generating unit 21 performs a stereo matching process by identifying corresponding points including two image points that correspond to each other based on the left image PL and the right image PR. The pixel value of each pixel of the distance image PZ is a parallax value, and is, for example, the difference between the coordinate value of the horizontal coordinate of the image point in the left image PL and the coordinate value of the horizontal coordinate of the image point in the right image PR. This parallax value corresponds to a distance value to a point corresponding to each pixel in a three-dimensional real space. Note that this is not limited to this, and for example, each of the multiple pixel values ​​may be a distance value indicating the distance to a point corresponding to each pixel in a three-dimensional real space. The distance image generating unit 21 is configured to supply the generated distance image PZ to the three-dimensional object detection unit 22.

[0016] The three-dimensional object detection unit 22 is configured to detect a three-dimensional object based on the distance image PZ. The three-dimensional object detection unit 22 is configured to set a three-dimensional object region Robj in an image region in the distance image PZ that corresponds to the detected three-dimensional object.

[0017] FIG. 3 shows an example of a three-dimensional object region Robj in the distance image PZ detected by the three-dimensional object detection unit 22. Although the distance image PZ is an image of parallax values, in FIG. 3, the three-dimensional object itself is depicted for convenience of explanation. The distance image PZ has continuous parallax values ​​in an image region corresponding to a three-dimensional object such as a vehicle, a building, or a person. The three-dimensional object detection unit 22 detects a three-dimensional object by using such continuous parallax values ​​included in the distance image PZ. Then, the three-dimensional object detection unit 22 sets a three-dimensional object region Robj in an image region corresponding to the detected three-dimensional object in the distance image PZ. Then, the three-dimensional object detection unit 22 supplies information about the set three-dimensional object region Robj and information about the parallax value in the three-dimensional object region Robj to the vehicle search unit 23 and the driving control unit 25.

[0018] The vehicle search unit 23 is configured to search for a vehicle based on the image P, which is one of the left image PL and the right image PR, by using a machine learning technique.

[0019] FIG. 4 is a schematic diagram showing an example of an operation of the vehicle search unit 23. The image P includes an image of a leading vehicle 90 traveling ahead of the vehicle 10. The vehicle search unit 23 sequentially sets a plurality of rectangular processing target regions Rp in the image P, for example, by gradually changing the position and size, based on the three-dimensional object region Robj set by the three-dimensional object detection unit 22. Then, the vehicle search unit 23 calculates a vehicle score SC in each processing target region Rp using a machine learning technique. The vehicle score SC indicates a higher value as the image of the processing target region Rp includes vehicle features. For example, when the vehicle score SC of the processing target region Rp having the highest vehicle score SC among the plurality of processing target regions Rp is equal to or higher than a predetermined score, the vehicle search unit 23 sets the processing target region Rp as the vehicle region Rv. The image P is, for example, an image having a high vehicle score SC among the left image PL and the right image PR. The vehicle search unit 23 supplies information about the set vehicle region Rv and the vehicle score SC to the smoke-like object detection unit 30 and the travel control unit 25.

[0020] The smoke-like object detection unit 30 is configured to detect a smoke-like object such as water vapor or exhaust gas based on the image P. The smoke-like object detection unit 30 has a pixel value analysis unit 31, a gradient data generation unit 32, and a similarity calculation unit 33.

[0021] The pixel value analysis unit 31 is configured to determine whether or not a smoky object is present based on the pixel value of each pixel in the image P. Specifically, the pixel value analysis unit 31 determines whether or not a smoky object is present by performing image analysis based on information about the color and brightness indicated by the pixel value of each pixel. For example, when the image P includes an image area of ​​a moderate size whose color is white or gray indicating a smoky object and whose brightness is a brightness indicating a smoky object, the pixel value analysis unit 31 can determine that a smoky object is present. Then, the pixel value analysis unit 31 is configured to supply the determination result of the smoky object to the driving control unit 25.

[0022] The gradient data generating unit 32 is configured to generate gradient data DG of pixel values ​​in the image P using HOG (Histogram of Oriented Gradients).

[0023] FIG. 5 shows an example of gradient data DG in an image P. In this example, the image P includes an image of a preceding vehicle 90. The gradient data generating unit 32 sets a calculation area Rg including the vehicle area Rv based on the vehicle area Rv set by the vehicle searching unit 23, and generates gradient data DG by calculating the gradient direction of pixel values ​​and the gradient value of pixel values ​​using HOG in each of a plurality of cell areas (35 (=7×5) in this example) set in the calculation area Rg. This FIG. 5 shows data on the gradient direction of pixel values. Specifically, the direction of the line in each cell area shown in FIG. 5 indicates the direction of the edge of the image. The gradient data DG corresponds to, for example, an image of the preceding vehicle 90. Therefore, when the vehicle 10 is traveling following the preceding vehicle 90, the gradient data DG obtained based on a series of images P captured by the stereo camera 11 are expected to be substantially the same as each other.

[0024] FIG. 6 shows another example of the gradient data DG. In this image P, a smoky object 91 is generated at the position of the preceding vehicle 90. Such an image P is obtained, for example, when the preceding vehicle 90 passes over a manhole from which water vapor is generated, or when exhaust gas is discharged from the preceding vehicle 90. The gradient data generating unit 32 generates gradient data DG of pixel values ​​based on the vehicle region Rv set immediately before by the vehicle searching unit 23, as in the case of FIG. 5. Since the image of the smoky object 91 is different from the image of the preceding vehicle 90, the gradient data DG obtained based on the image P of FIG. 6 is different from the gradient data DG obtained based on the image P of FIG. 5. In addition, since the shading of the smoky object 91 changes over time, the gradient data DG corresponding to the smoky object 91 obtained based on the series of images P captured by the stereo camera 11 may change over time.

[0025] The gradient data generating unit 32 generates such gradient data DG, and supplies the generated gradient data DG to the similarity calculating unit 33.

[0026] The similarity calculation unit 33 is configured to calculate the similarity A by comparing the gradient data DG obtained based on the latest image P of the series of images P captured by the stereo camera 11 with the gradient data DG obtained based on the image P immediately before. Specifically, the similarity calculation unit 33 calculates the cosine similarity in this example based on the gradient data DG (gradient data DG1) obtained based on the latest image P and the gradient data DG (gradient data DG0) obtained based on the image P immediately before. This similarity A can take a value of, for example, "0" or more and "1" or less. The more similar the gradient data DG1 and the gradient data DG0 are to each other, the closer the similarity A is to "1", and the more different the gradient data DG1 and the gradient data DG0 are to each other, the closer the similarity A is to "0". The similarity calculation unit 33 calculates the similarity A by comparing the gradient data DG1 and the gradient data DG0, and supplies data on the calculated similarity A to the driving control unit 25.

[0027] The driving control unit 25 is configured to perform driving control of the vehicle 10 based on the detection result of the three-dimensional object detection unit 22, the search result of the vehicle search unit 23, the detection result of the smoke-like object detection unit 30, and vehicle information INF acquired from a vehicle control device (not shown) of the vehicle 10. The vehicle information INF includes, for example, driving information such as the driving speed and yaw angle of the vehicle 10, and status information indicating the operation status of the windshield wipers and turn signals of the vehicle 10. The driving control unit 25 supplies driving control information CTL that instructs the driving control of the vehicle 10 to the vehicle control device. For example, as shown in FIG. 5, when the vehicle 10 is driving following the preceding vehicle 90, if the preceding vehicle 90 to be detected decelerates, the driving control unit 25 performs brake control so that the vehicle 10 decelerates. Furthermore, when the vehicle 10 is traveling following a preceding vehicle 90, for example, if a smoke-like object 91 such as water vapor or exhaust gas is generated as shown in FIG. 6, the traveling control unit 25 performs appropriate brake control so as not to suddenly stop the vehicle 10. That is, generally, when a smoke-like object 91 is generated as shown in FIG. 6, the detection target may be changed from the preceding vehicle 90 to the smoke-like object 91. In this case, since the smoke-like object 91 is generally almost stationary, brake control is performed so as to suddenly stop the vehicle 10. In the traveling control device 1, in such a case, appropriate brake control is performed so as not to suddenly stop the vehicle 10.

[0028] Here, the gradient data generating unit 32 corresponds to a specific example of a "generating unit" in the present disclosure. The similarity calculating unit 33 corresponds to a specific example of a "comparison unit" in the present disclosure. The driving control unit 25 corresponds to a specific example of a "control unit" in the present disclosure. The pixel value analyzing unit 31 corresponds to a specific example of an "analysis unit" in the present disclosure. The image P corresponds to a specific example of a "captured image" in the present disclosure. The vehicle area Rv corresponds to a specific example of a "first image area" in the present disclosure. The calculation area Rg corresponds to a specific example of a "second image area" in the present disclosure. The similarity A corresponds to a specific example of a "similarity" in the present disclosure.

[0029] [Actions and Actions] Next, the operation and function of the cruise control device 1 of the present embodiment will be described.

[0030] (Overview of overall operation) First, referring to FIG. 1, an overview of the overall operation of the driving control device 1 will be described. The stereo camera 11 captures an image in front of the vehicle 10 to generate a stereo image PIC including a left image PL and a right image PR having a parallax from each other. The distance image generating unit 21 generates a distance image PZ based on the left image PL and the right image PR included in the stereo image PIC. The three-dimensional object detecting unit 22 detects a three-dimensional object based on the distance image PZ to set a three-dimensional object area Robj. The vehicle searching unit 23 searches for a vehicle using a machine learning technique based on an image P that is one of the left image PL and the right image PR to set a vehicle area Rv. The smoke-like object detecting unit 30 detects a smoke-like object such as water vapor or exhaust gas based on the image P. The driving control unit 25 performs driving control of the vehicle 10 based on the detection result of the three-dimensional object detecting unit 22, the search result of the vehicle searching unit 23, the detection result of the smoke-like object detecting unit 30, and the vehicle information INF acquired from a vehicle control device (not shown) of the vehicle 10.

[0031] (Details of operation) In the smoky object detection unit 30, the pixel value analysis unit 31 performs image analysis based on the pixel value of each pixel in the image P to determine whether or not there is a smoky object. The gradient data generation unit 32 generates gradient data DG of pixel values ​​in the image P using HOG. The similarity calculation unit 33 calculates the similarity A by comparing the gradient data DG (gradient data DG1) obtained based on the latest image P with the gradient data DG (gradient data DG0) obtained based on the image P immediately before. The driving control unit 25 performs appropriate brake control based on the detection result of the smoky object detection unit 30 and the vehicle information INF, for example, when the vehicle 10 is traveling following the preceding vehicle 90, and when a smoky object such as water vapor or exhaust gas occurs, so as not to suddenly stop the vehicle 10. This operation will be described in detail below.

[0032] 7A and 7B show an example of the operation of the driving control device 1. The driving control device 1 performs the following process each time the stereo camera 11 generates a stereo image PIC.

[0033] First, the smoky object detection unit 30 judges whether or not a smoky object is present based on the pixel value of each pixel in the image P (step S101). Specifically, the pixel value analysis unit 31 judges whether or not a smoky object is present based on information about the color and brightness indicated by the pixel value of each pixel. For example, as shown in FIG. 6, when a smoky object 91 occurs, if the smoky object 91 is relatively dense, the pixel value analysis unit 31 can judge that a smoky object is present. If it is judged that no smoky object is present ("N" in step S102), the process proceeds to step S104.

[0034] When it is determined that a smoky object is present ("Y" in step S102), the driving control unit 25 increments the count value CNT (step S103). This count value CNT is an evaluation value of a smoky object, which indicates that the larger the value, the more likely it is that the image P contains an image of a smoky object. In this way, when the pixel value analysis unit 31 determines that a smoky object is present, the driving control unit 25 increments the count value CNT, which is the evaluation value of the smoky object, by "1".

[0035] Next, the driving control unit 25 judges whether both the left image PL and the right image PR are normal based on the search result of the vehicle search unit 23 (step S104). Specifically, the driving control unit 25 can judge that the left image PL is normal when the vehicle score SC obtained based on the left image PL exceeds a predetermined value, and can similarly judge that the right image PR is normal when the vehicle score SC obtained based on the right image PR exceeds a predetermined value. If at least one of the left image PL and the right image PR is not normal ("N" in step S104), the process proceeds to step S110.

[0036] In step S104, if both the left image PL and the right image PR are normal ("Y" in step S104), the driving control unit 25 checks whether the surroundings of the vehicle 10 are bright based on the image P (step S105). Specifically, the driving control unit 25 checks whether the external environment around the vehicle 10 is bright based on the pixel value of each pixel. If the surroundings are not bright ("N" in step S105), the process proceeds to step S110.

[0037] In step S105, if the surroundings are bright ("Y" in step S105), the driving control unit 25 checks whether the operation state of the wipers for wiping the windshield on which the stereo camera 11 is located is off based on the vehicle information INF acquired from a vehicle control device (not shown) of the vehicle 10 (step S106). If the wipers are on ("N" in step S106), the process proceeds to step S110.

[0038] The three conditions in steps S104 to S106 are conditions for performing processing based on the similarity A. That is, for example, when at least one of the left image PL and the right image PR is not normal (step S104), when the surroundings are not bright (step S105), or when the wipers are on (step S106), the obtained similarity A may not be appropriate. Therefore, when both the left image PL and the right image PR are normal (step S104), when the surroundings are bright (step S105), and when the wipers are off (step S106), the driving control device 1 performs processing based on the similarity A as described below.

[0039] In step S106, if the wiper is in the off state ("Y" in step S106), the driving control unit 25 checks whether the similarity A calculated by the similarity calculation unit 33 maintains a threshold value TH0 or more during a period T having a predetermined time length in the past (step S107). The threshold value TH0 is "0.9" in this example. As shown in FIG. 5, when the vehicle 10 is traveling following the preceding vehicle 90, the gradient data DG obtained based on the series of images P captured by the stereo camera 11 are expected to be almost the same, so that the similarity A is expected to maintain a threshold value TH0 or more during such a period. Therefore, in this step S106, by checking that the similarity A maintains a threshold value TH0 or more during a period T having a predetermined time length in the past, it is possible to check that the similarity A has been calculated normally. If the similarity A does not maintain a threshold value TH0 or more during the period T ("N" in step S107), the process proceeds to step S110.

[0040] In step S107, if the similarity A remains equal to or greater than the threshold value TH0 during the period T ("Y" in step S107), the driving control unit 25 checks whether the latest similarity A calculated by the similarity calculation unit 33 is smaller than any of the multiple threshold values ​​TH (step S108). In this example, the multiple threshold values ​​TH include threshold values ​​THA and THB, where the threshold value THA is "0.84" and the threshold value THB is "0.75".

[0041] For example, as shown in FIG. 6, when a smoky object 91 occurs, the similarity A may be smaller than any one of the multiple threshold values ​​TH. For example, when the smoky object 91 is relatively thick, the similarity A may be smaller than the threshold value THB. Also, when the smoky object is thin and the preceding vehicle 90 is seen thinly behind the smoky object, the similarity A may be larger than when the smoky object 91 is relatively thick, but may be smaller than the threshold value THA. That is, even when the smoky object is thin and the pixel value analysis unit 31 determines in step S101 that there is no smoky object, the similarity A may be smaller than the threshold value THA. When the similarity A is not smaller than any one of the multiple threshold values ​​TH ("N" in step S108), the process proceeds to step S110.

[0042] In step S108, if the similarity A is smaller than any one of the multiple threshold values ​​TH ("Y" in step S108), the driving control unit 25 increases the count value CNT according to the similarity A (step S109). Specifically, for example, the driving control unit 25 can increase the count value CNT by an amount corresponding to the number of threshold values ​​TH that the similarity A has fallen below among the multiple threshold values ​​TH. For example, the driving control unit 25 can increase the count value CNT by "1" when the similarity A is greater than the threshold value THB (e.g., "0.75") and smaller than the threshold value THA (e.g., "0.84"), and can increase the count value CNT by "2" when the similarity A is smaller than the threshold value THB (e.g., "0.75").

[0043] Then, the traveling control unit 25 checks whether the count value CNT has increased in the above processing (step S110). That is, the traveling control unit 25 checks whether the count value CNT has increased by at least one of the processing of step S103 and the processing of step S109. If the count value CNT has not increased ("N" in step S110), the traveling control unit 25 decrements the count value CNT (step S111). That is, the traveling control unit 25 decrements the count value CNT, which is an evaluation value of a smoke-like object, by "1".

[0044] Next, the driving control unit 25 checks whether the count value CNT is greater than a predetermined count value CNT0 (step S112). If the count value CNT is greater than the predetermined count value CNT0 ("Y" in step S112), the driving control unit 25 suppresses the brake control (step S113).

[0045] That is, the traveling control unit 25 normally performs brake control so that the vehicle 10 decelerates when a detection target in front of the vehicle 10 decelerates or stops. Therefore, for example, as shown in FIG. 5, when the vehicle 10 is traveling following a preceding vehicle 90, if the preceding vehicle 90, which is a detection target, decelerates, the traveling control unit 25 performs brake control so that the vehicle 10 decelerates. Similarly, when a smoke-like object 91 occurs as shown in FIG. 6, the detection target is changed from the preceding vehicle 90 to the smoke-like object 91, so the traveling control unit 25 tries to perform brake control so that the vehicle 10 suddenly stops. However, since the smoke-like object 91 occurs temporarily and the vehicle 10 cannot collide with the smoke-like object 91 itself, brake control that suddenly stops the vehicle 10 is not necessarily appropriate. Therefore, the traveling control unit 25 suppresses such brake control. Specifically, the traveling control unit 25 performs brake control so that the vehicle 10 is slightly decelerated, for example. Note that the present invention is not limited to this, and brake control may be suppressed so as not to decelerate the vehicle 10.

[0046] This is the end of this flow. Here, the threshold value TH corresponds to a specific example of a "first threshold value" in the present disclosure. The threshold values ​​THA and THB correspond to a specific example of a "multiple threshold values" in the present disclosure. The threshold value TH0 corresponds to a specific example of a "second threshold value" in the present disclosure.

[0047] In this way, the cruise control device 1 generates gradient data DG by detecting the gradient of pixel values ​​in the calculation region Rg set according to the vehicle region Rv, compares two gradient data DG generated based on different images P, and suppresses brake control of the vehicle 10 based on the comparison result. As a result, for example, as shown in Fig. 6, when a smoke-like object 91 occurs, the cruise control unit 25 can perform brake control to slightly decelerate the vehicle 10 instead of performing brake control to bring the vehicle to an abrupt stop. Therefore, the cruise control device 1 can appropriately perform brake control of the vehicle.

[0048] In addition, the driving control device 1 calculates the similarity A by comparing the two gradient data DG0, DG1, and when the similarity A is lower than the threshold value TH, the brake control can be suppressed. In particular, the driving control device 1 is configured to provide a plurality of threshold values ​​TH. As a result, for example, even when a thin smoky object is detected, the count value CNT is lower than the high threshold value TH, so that the count value CNT can be increased and the brake control can be suppressed. Also, when a thick smoky object is detected, the count value CNT is lower than the low threshold value TH, so that the response time until the brake control is suppressed can be shortened. In this way, the driving control device 1 can flexibly detect a smoky object based on the similarity A, so that the brake control of the vehicle can be appropriately performed.

[0049] In addition, the driving control device 1 is configured to calculate a count value CNT by accumulating a value according to a comparison result between the similarity A and a threshold value TH, and suppress brake control based on a comparison result between this count value CNT and a predetermined count value CNT0. This makes it possible to reduce the risk of suppressing brake control due to erroneous detection, for example, and therefore to appropriately control the vehicle brakes.

[0050] Furthermore, the driving control device 1 calculates the count value CNT by accumulating values ​​according to the magnitude relationship between the similarity A and each of a plurality of threshold values ​​TH (threshold values ​​THA, THB). This makes it possible to increase the count value CNT by two when a thick smoky object is detected, and to increase the count value CNT by one when a thin smoky object is detected. As a result, when the driving control device 1 detects a thick smoky object, for example, it is possible to shorten the response time until braking control is suppressed, and therefore to appropriately control the vehicle's brakes.

[0051] In addition, the driving control device 1 checks whether the pixel values ​​in the image P satisfy a predetermined condition, and calculates the count value CNT by accumulating values ​​according to whether the pixel values ​​satisfy the predetermined condition. As a result, the driving control device 1 suppresses brake control using a process based on pixel values ​​performed by the pixel value analysis unit 31 and a process based on the similarity A of the gradient data DG performed by the similarity calculation unit 33, which are based on different principles, so that it is possible to appropriately determine whether to suppress brake control. In addition, the driving control device 1 calculates the count value CNT using these two processes, so that it is possible to shorten the response time until the brake control is suppressed. As a result, the driving control device 1 can appropriately perform brake control of the vehicle.

[0052] [effect] As described above, in this embodiment, gradient data is generated by detecting the gradient of pixel values ​​in the calculation area set according to both areas, two gradient data generated based on different images are compared, and vehicle brake control can be suppressed based on the comparison result, so that vehicle brake control can be performed appropriately.

[0053] In this embodiment, the similarity is calculated by comparing two gradient data, and when the similarity is lower than a threshold value, the brake control can be suppressed, so that the vehicle brake control can be performed appropriately.

[0054] In this embodiment, a count value is calculated by accumulating values ​​corresponding to the results of comparing the similarity with a threshold value, and brake control can be suppressed based on the results of comparing this count value with a predetermined count value, thereby enabling appropriate vehicle brake control.

[0055] In this embodiment, the count value is calculated by accumulating values ​​according to the magnitude relationship between the similarity and each of a number of threshold values, thereby shortening the response time until brake control is suppressed, thereby enabling appropriate vehicle brake control.

[0056] In this embodiment, it is confirmed whether or not pixel values ​​in an image satisfy a predetermined condition, and further, a count value is calculated by accumulating values ​​according to whether or not this predetermined condition is satisfied, so that appropriate vehicle brake control can be performed.

[0057] [Variation 1] In the above embodiment, as shown in FIG. 7A, when the conditions of steps S104 to S106 are satisfied, the process based on the similarity A from step S107 onwards is performed, but the present invention is not limited to this. In addition, when other conditions are satisfied, the process based on the similarity A may be performed. The driving control device 1A according to this modification will be described in detail below. The driving control device 1A includes a processing unit 20A similar to the driving control device 1 (FIG. 1) according to the above embodiment. The processing unit 20A includes a driving control unit 25A.

[0058] 8A and 8B show an example of the operation of the driving control device 1A according to this modification. Steps S101 to S103 and steps S104 to S113 are similar to those of the driving control device 1 (FIGS. 7A and 7B) according to the above embodiment.

[0059] In steps S102 and S103, after the driving control unit 25A finishes the process according to the determination result of the pixel value analysis unit 31, the driving control unit 25A checks whether the detection target has suddenly stopped (step S121). Specifically, the driving control unit 25A acquires the distance from the vehicle 10 to the detected three-dimensional object (detection target) based on the detection result of the three-dimensional object detection unit 22, for example, and can obtain the relative speed of the detection target as seen from the vehicle 10 based on the change in the distance over time. In addition, the driving control unit 25A can obtain the running speed of the vehicle 10 based on the vehicle information INF. The driving control unit 25A can obtain the running speed of the detection target based on these pieces of information. The driving control unit 25A can check whether the detection target has suddenly stopped based on the running speed of the detection target obtained in this way. Specifically, the driving control unit 25A can determine that the detection target has suddenly stopped when, for example, the result of subtracting the average value of the running speed of the detection target (average running speed Vave) by the instantaneous value of the running speed of the detection target (running speed V1) satisfies the following formula. Vave / V1 > 1.7 The average traveling speed Vave is calculated, for example, by filtering the traveling speed V1 to remove noise, accumulating the removed value for a certain period of time, and performing a weighted average based on the accumulated value. The "1.7" on the right side is an example, and other values ​​may be used. This determination may be performed, for example, when the traveling speed V1 is slower than a predetermined speed (for example, 20 km / h). If the detection target is not suddenly stopped ("N" in step S121), the process proceeds to step S110. That is, in this case, the traveling control unit 25A does not attempt to perform brake control to suddenly stop the vehicle 10, and therefore there is no need to suppress the brake control of the vehicle 10, and the process proceeds to step S110.

[0060] In step S121, if the detection target has suddenly stopped ("Y" in step S121), the cruise control unit 25A checks whether the vehicle 10 is decelerating (step S122). Specifically, the cruise control unit 25A can obtain the traveling speed of the vehicle 10 based on the vehicle information INF, and can therefore check whether the vehicle 10 is decelerating. If the vehicle 10 is decelerating ("Y" in step S122), the process proceeds to step S110. That is, for example, if the automatic brake of the vehicle 10 has already been activated, there is no need to suppress the brake control, and so the process proceeds to step S110.

[0061] In step S122, if the vehicle 10 is not decelerating ("N" in step S122), the driving control unit 25A checks whether the detection target has moved out of the angle of view of the stereo camera 11 (step S123). Specifically, the driving control unit 25A can check whether the detection target has moved out of the angle of view, for example, based on the detection result of the vehicle search unit 23. If the detection target has moved out of the angle of view ("Y" in step S123), the process proceeds to step S110. That is, in this case, the calculation region Rg (FIG. 5) cannot be set appropriately, and as a result, the gradient data DG cannot be generated appropriately, so the process proceeds to step S110.

[0062] In step S123, if the detection target is not outside the angle of view ("N" in step S123), the process proceeds to step S104.

[0063] In this example, the three conditions of steps S121 to S123 are used, but the present invention is not limited to this, and only one or two of these three conditions may be used.

[0064] [Variation 2] In the above embodiment, the multiple thresholds TH include two thresholds THA and THB, but are not limited to this and may include, for example, three or more thresholds TH. Also, are not limited to multiple thresholds TH and may be one threshold TH.

[0065] [Other variations] Also, two or more of these variations may be combined.

[0066] Although the present technology has been described above by giving embodiments and modifications, the present technology is not limited to these embodiments and the like, and various modifications are possible.

[0067] For example, in the above embodiment, the stereo camera 11 is configured to capture an image in front of the vehicle 10, but this is not limited thereto. For example, the stereo camera 11 may capture an image of the side or rear of the vehicle 10.

[0068] It should be noted that the effects described in this specification are merely examples and are not limiting, and other effects may also be obtained. [Explanation of symbols]

[0069] 1...driving control device, 11...stereo camera, 11L...left camera, 11R...right camera, 20...processing unit, 21...distance image generation unit, 22...three-dimensional object detection unit, 23...vehicle search unit, 25...driving control unit, 30...smoke-like object detection unit, 31...pixel value analysis unit, 32...gradient data generation unit, 33...similarity calculation unit, A...similarity, CNT, CNT0...count value, CTL...driving control information, DG...gradient data, INF...vehicle information, PIC...stereo image, P...image, PL...left image, PR...right image, PZ...distance image, Rg...calculation area, Robj...three-dimensional object area, Rp...processing target area, Rv...vehicle area, TH0, THA, THB...threshold value.

Claims

1. a generation unit that generates gradient data by detecting a gradient of pixel values ​​in a second image area that is set in accordance with a first image area including an image of a leading vehicle in a captured image; a comparison unit that compares first gradient data and second gradient data generated as the gradient data by the generation unit based on the captured images captured at different times; a control unit capable of suppressing brake control of the host vehicle based on a comparison result of the comparison unit so as to prevent the host vehicle from suddenly stopping; A driving control device comprising:

2. the comparison unit calculates a similarity between the first gradient data and the second gradient data by comparing the first gradient data and the second gradient data; The control unit is capable of suppressing the brake control when the degree of similarity is lower than a first threshold value. The cruise control device according to claim 1.

3. The control unit is capable of calculating an evaluation value by accumulating a value according to a result of comparing the similarity with the first threshold value, and suppressing the brake control based on a result of comparing the evaluation value with a predetermined value. The cruise control device according to claim 2.

4. the first threshold value includes a plurality of threshold values; The control unit calculates the evaluation value by accumulating values ​​according to a magnitude relationship between the similarity and each of the plurality of threshold values. The cruise control device according to claim 3.

5. An analysis unit that checks whether a pixel value in the captured image satisfies a predetermined condition, The control unit further calculates the evaluation value by accumulating values ​​according to whether the predetermined condition is satisfied. The driving control device according to claim 3 or 4.

6. The control unit suppresses the brake control when a plurality of similarities obtained in a past period having a predetermined time length are higher than a second threshold value and the similarity is lower than the first threshold value. The cruise control device according to any one of claims 2 to 5.

7. The control unit is capable of suppressing the brake control based on a comparison result of the comparison unit when the captured image is normal, the surroundings of the host vehicle are bright, and the wipers of the host vehicle are in an off state. The cruise control device according to any one of claims 1 to 6.

8. The comparison unit compares the first gradient data with the second gradient data to determine whether the image of the first image region is likely to include an image of a smoke-like object. The cruise control device according to any one of claims 1 to 7.

9. generating gradient data by detecting a gradient of pixel values ​​in a second image area in the captured image, the second image area being set in accordance with a first image area including an image of the leading vehicle; comparing first gradient data and second gradient data generated as the gradient data based on the captured images captured at different times; suppressing brake control of the host vehicle so as not to suddenly stop the host vehicle based on a comparison result between the first gradient data and the second gradient data; A driving control method comprising:

Images