Driving assistance device and driving assistance method

Abstract

To provide a technique for appropriately controlling a plurality of blind spots in a technique for supporting drive of a vehicle by controlling the vehicle corresponding to the blind spot.SOLUTION: A drive support device supports drive of a vehicle by controlling the vehicle corresponding to a blind spot. The drive support device stores risk estimation information for calculating a risk estimation value indicating a level of risk of a blind spot, and when a plurality of blind spots are detected in front of the vehicle and the plurality of blind spots satisfy predetermined conditions, determines the plurality of blind spots as one composite risk. When the composite risk is present, the drive support device calculates the risk estimation value indicating the level of the risk of each of the plurality of blind spots based on the risk estimation information, and determines a control end position at which control of the vehicle corresponding to the blind spot is ended and target passing speed of the control end position of the vehicle based on the risk estimation value.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present disclosure relates to a driving assistance device and a driving assistance method. [Background technology]

[0002] Patent Document 1 discloses a driving assistance device that assists a vehicle driver. The driving assistance device detects an obstacle ahead of the vehicle and sets a vehicle trajectory assuming a virtual moving object as a potential risk in the blind spot of the obstacle. In this way, the driving assistance device provides assistance that takes into account unobserved risks such as pedestrians appearing from the blind spot of the obstacle. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-206117 A Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure relates to a driving assistance device that assists driving of a vehicle by controlling the vehicle while taking into account potential risks present in blind spots. The present disclosure provides a technology that can perform appropriate control for multiple blind spots. [Means for solving the problem]

[0005] The first aspect relates to a driving assistance device that assists in driving a vehicle by controlling the vehicle corresponding to a blind spot when a blind spot is detected in front of the vehicle. The driving assistance device stores risk estimation information for calculating a risk estimation value indicating the level of risk of the blind spot, and when multiple blind spots are detected in front of the vehicle and the multiple blind spots satisfy a predetermined condition, determines the multiple blind spots as one compound risk. When a compound risk exists, the driving assistance device calculates an individual risk estimation value indicating the level of risk of each of the multiple blind spots and a compound risk estimation value indicating the level of risk of the entire compound risk based on the risk estimation information, determines a control end position for ending the control of the vehicle corresponding to the blind spot to a position passing through the blind spot with the highest individual risk estimation value, and determines a target passing speed of the control end position of the vehicle according to the compound risk estimation value. The compound risk estimation value is set to the total value of the individual risk estimation values ​​of the blind spots included in the compound risk.

[0006] The second aspect relates to a driving assistance method that assists driving of a vehicle by controlling the vehicle in response to a blind spot detected ahead of the vehicle. When multiple blind spots are detected ahead of the vehicle and the multiple blind spots satisfy a predetermined condition, the driving assistance method determines the multiple blind spots as one compound risk, acquires risk estimation information for calculating a risk estimation value indicating the level of risk of the blind spots, and calculates a risk estimation value for each of the multiple blind spots based on the risk estimation information. For the compound risk, a control end position for terminating the control of the vehicle corresponding to the blind spots is set to a position passing through the blind spot with the highest individual risk estimation value, and a target passing speed of the vehicle at the control end position is determined according to the total value of the calculated risk estimation values. Effect of the Invention

[0007] According to the technology disclosed herein, in a technology for assisting vehicle driving by controlling the vehicle in response to blind spots, appropriate control can be performed for multiple blind spots. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a conceptual diagram for explaining a blind spot. [Diagram 2] FIG. 1 is a conceptual diagram for explaining a plurality of blind spots. [Diagram 3] FIG. 13 is a diagram illustrating a situation in which multiple blind spots are not treated as a compound risk. [Figure 4] FIG. 13 is a diagram illustrating a situation in which multiple blind spots are treated as a compound risk. [Diagram 5] 11 is a table for explaining the relationship between a single unit risk estimate value and a controlled variable. [Figure 6] FIG. 11 is a graph for explaining the relationship between a composite risk estimated value and a deceleration support amount. [Figure 7] 4 is a time chart illustrating control performed by the driving assistance device according to the present embodiment. [Figure 8] 13 is a table for explaining a method for calculating a single risk estimate value. [Figure 9] FIG. 13 is a diagram for explaining a pre-evaluation section. [Figure 10] FIG. 13 is a diagram showing an example of a situation in which the driving assistance device does not treat a compound risk as a compound risk in a modified example of the compound risk judgment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Embodiments of the present disclosure will be described with reference to the accompanying drawings.

[0010] 1. Complex risks The shape of the road and vehicles parked on the road may cause blind spots for vehicles traveling on the road. When a blind spot exists, an unexpected moving object may jump out, and there is a risk that the vehicle will not be able to decelerate in time and will collide with the moving object that jumps out from the blind spot. Therefore, when a blind spot is detected in front of the vehicle, the driving assistance device according to the present embodiment controls the vehicle in response to the blind spot to assist the driver or the automatic driving device in driving the vehicle. FIG. 1 shows a blind spot created by an intersection with poor visibility as an example of a blind spot. In preparation for the possibility that a moving object such as a pedestrian may jump out, the driving assistance device performs deceleration assistance to reduce the vehicle speed of the vehicle 10 when passing through the blind spot by gentle braking, and secures a lateral margin by controlling the driving trajectory of the vehicle 10 so as to keep a sufficient distance from the blind spot. These controls performed by the driving assistance device can increase the possibility that the vehicle 10 can avoid a collision even if a moving object suddenly jumps out from the blind spot.

[0011] There may be a plurality of consecutive blind spots on a road. In the example of FIG. 2, the first blind spot is a blind spot created by a parked vehicle, and the second blind spot is a blind spot created by an intersection. Even when there are a plurality of blind spots, the driving assistance device controls the vehicle 10 to secure a lateral margin and assist in deceleration. However, when there are a plurality of blind spots, if control is performed for each blind spot alone, the running trajectory and vehicle speed of the vehicle 10 may fluctuate irregularly, which may cause discomfort to the driver of the vehicle 10. In addition, when there are a plurality of blind spots, the risk of occurrence may be greater than when there is only one blind spot. Therefore, when a plurality of blind spots detected ahead of the vehicle 10 satisfy a predetermined condition, the driving assistance device treats the plurality of blind spots as a single risk, called a "composite risk." Then, by performing control collectively for the plurality of blind spots, it is possible to reduce the inconvenience to the driver while adequately preparing for risks.

[0012] Specifically, when the vehicle 10 deviates from the driving lane to secure a lateral margin for one blind spot and does not return to the original driving lane before passing through the next blind spot, the driving assistance device treats the blind spots as a compound risk. The driving trajectory of the vehicle 10 is calculated by the driving assistance device for control of the vehicle 10. At this time, a condition may be set in advance for the driving trajectory calculated by the driving assistance device. For example, a condition may be set in advance that the lateral acceleration is 0.2 G or less so that the movement of the vehicle 10 is not abrupt. The driving assistance device calculates the driving trajectory of the vehicle 10 according to such a condition, and then determines whether the vehicle 10 can return to the original driving lane.

[0013] In the example of FIG. 3, the distance D between the blind spot 1 and the blind spot 2 is sufficient, so the vehicle 10 that has left the driving lane to secure a lateral margin for the blind spot 1 can return to the original driving lane before passing the blind spot 2. In such a case, the driving assistance device does not treat the blind spot 1 and the blind spot 2 as a composite risk, but controls each of them independently. On the other hand, in the example of FIG. 4, the distance D between the blind spot 1 and the blind spot 2 is relatively short, so the vehicle 10 that has left the driving lane to secure a lateral margin for the blind spot 1 cannot return to the original driving lane before passing the blind spot 2. In such a case, the driving assistance device treats the blind spot 1 and the blind spot 2 as a composite risk, and integrally controls the vehicle 10 for driving assistance. The same applies when there are three or more blind spots. For example, in the example of FIG. 4, if the driving trajectory of the vehicle 10 cannot return to the original driving lane before passing the blind spot 3 next to the blind spot 2, the blind spot 1, the blind spot 2, and the blind spot 3 are treated as composite risks.

[0014] Whether or not the vehicle 10 can return to the original driving lane is affected by the distance D between the first and second blind spots, the vehicle speed, the road width, etc. For example, when the distance D is sufficiently long or the vehicle speed of the vehicle 10 is low, the vehicle 10 is more likely to be able to return to the original driving lane. Also, if the road width is wide, the amount by which the vehicle 10 deviates from the original driving lane in order to secure a lateral margin is reduced, and the vehicle 10 is more likely to be able to return to the original driving lane.

[0015] 2.Determining the amount of control When blind spots exist consecutively, the risk created by the blind spots may be greater than when a single blind spot exists. Therefore, when the driving assistance device treats multiple blind spots as a compound risk, it is important to make the control amount, particularly the amount and timing of deceleration assistance, appropriate. If the speed when passing through the blind spot is not sufficiently reduced, the risk cannot be adequately prepared, and if the speed is reduced too much or the distance traveled while decelerating is too long, the driver will feel annoyed. Therefore, the driving assistance device calculates an index indicating the magnitude of the risk created by the blind spots as a "risk estimate value" and determines the control amount based on the risk estimate value. Information for calculating the risk estimate value is stored in advance by the driving assistance device as risk estimate information. Among the risk estimate values, a value indicating the magnitude of the risk created by each blind spot included in the compound risk is set as a single risk estimate value. The control amount is determined based on the single risk estimate value.

[0016] The table in Fig. 5 shows an example of the relationship between the risk estimate value and the control amount when blind spot 1 and blind spot 2 are treated as a compound risk. As shown in Fig. 5, the control end position at which the driving assistance device ends the control of the vehicle 10 corresponding to the blind spot is determined as the position of the blind spot with the highest individual risk estimate value among the blind spots included in the compound risk. Generally, drivers are aware of the position of the blind spot with the highest risk when considering the position where they should decelerate the most to pass through, so by determining the control end position in this way, the control by the driving assistance device can be made to match the driver's sense.

[0017] The driving assistance device also determines the target passing speed when the vehicle 10 passes the control end position, that is, the position of the blind spot with the highest individual risk estimate value, based on the individual risk estimate value. The driving assistance device first calculates a composite risk estimate value indicating the magnitude of the overall risk of the composite risk based on the individual risk estimate value. Then, the driving assistance device determines the target passing speed so that the target passing speed is lower as the composite risk estimate value is larger and the target passing speed is higher as the composite risk estimate value is smaller. Here, the composite risk estimate value is the sum of the individual risk estimate values ​​of the blind spots included in the composite risk. When considering the risk caused by the blind spots, the driver generally considers the magnitude of the risk as a single mass of multiple blind spots, so by setting the composite risk estimate value in this way, the amount of deceleration assistance performed by the driving assistance device can be made to match the driver's sense.

[0018] The graph in FIG. 6 shows the relationship between the composite risk estimate value and the target passing speed. The larger the composite risk estimate value, the greater the risk of multiple blind spots occurring, and the target passing speed is set lower to increase the risk prevention. If the vehicle speed when the blind spot is detected is the same, the amount of deceleration assistance performed by the driving assistance device to reduce the vehicle speed to the target passing speed increases as the composite risk estimate value increases. In this way, the control amount is determined based on the risk estimate value, making it possible to make the control amount appropriate.

[0019] 3. Time chart The control of the vehicle 10 performed by the driving support device will be specifically described using a time chart. Figure 7 shows the transition of the vehicle speed when there are blind spots 1 and 2, which are treated as a compound risk. The upper time chart corresponds to the case of (1) in the table of Figure 5, and the lower time chart corresponds to the case of (2) in the table of Figure 5.

[0020] At point P1, the driving assistance device detects a blind spot. Having detected the blind spot, the driving assistance device starts estimating the driving trajectory of the vehicle 10 and the magnitude of the risk of the blind spot. The driving assistance device determines whether or not to treat the blind spot 1 and blind spot 2 detected by estimating the driving trajectory as a compound risk, and at point P2, it is determined that the blind spot 1 and blind spot 2 are to be treated as a compound risk. Therefore, the driving assistance device calculates individual risk estimates of the blind spot 1 and blind spot 2 at the latest before starting control at point P3, and determines the compound risk estimate, the target passing speed, and the control end position based on the individual risk estimates. Point P3 at which control starts is, for example, a point at which the arrival time to the first blind spot 1 among the blind spots included in the compound risk, that is, the distance to the blind spot 1 divided by the current vehicle speed of the vehicle 10, falls below a predetermined threshold.

[0021] The driving assistance device gently decelerates the vehicle 10 so that the target passing speed is reached at the control end position. At this time, an upper limit of the deceleration (e.g., 0.2 G) may be determined in advance so that deceleration assistance is performed by gentle braking. As described above, the control end position and the target passing speed are determined based on the individual risk estimates of blind spots 1 and 2. The target passing speed is determined according to the composite risk estimate, which is the sum of the individual risk estimates, and is therefore equal in the cases (1) and (2) of FIG. 7. Meanwhile, the control end position is set to point P5, which is the position of blind spot 2, in (1), and to point P4, which is the position of blind spot 1, in (2).

[0022] As a result, in (1), the vehicle speed when passing through blind spot 1 is lower than the vehicle speed when the risk is detected, but is higher than the vehicle speed when passing through blind spot 2. Here, the individual risk estimate is greater for blind spot 2 than for blind spot 1, so the vehicle 10 needs to prepare for blind spot 2 by slowing down. Therefore, it can be said that it is appropriate control to set the target passing speed as the vehicle speed when passing through blind spot 2 and perform gradual deceleration.

[0023] In the case of (2), control by the driving assistance device has already ended by the time the vehicle passes blind spot 2. However, because the vehicle has already decelerated by the time it passes blind spot 1, the vehicle speed when passing blind spot 2 is reduced compared to the case where there is no assistance from the driving assistance device. This improves the possibility of avoiding a collision even if a moving object suddenly appears from blind spot 2. In this way, assistance by the driving assistance device can be made appropriate even in a case such as (2).

[0024] In both (1) and (2), when control by the driving assistance device ends, the vehicle 10 returns to a state in which it is controlled by the driver or the automatic driving device. If a moving object suddenly appears from the blind spot, the driver or the automatic driving device will decelerate the vehicle by applying emergency brakes. If no moving object suddenly appears, the driver or the automatic driving device will normally accelerate the vehicle.

[0025] 4. Method for calculating individual risk estimates The individual risk estimate is calculated, for example, by the following formula:

number

[0026] In this way, the individual risk estimates (R1, R2) are calculated based on the environmental factors (F1, F2, F3...) related to the target blind spot and the factor coefficients (α1, α2, α3... or β1, β2, β3...). The environmental factors are determined by the blind spot and the surrounding environment of the blind spot, for example, as shown in the table of FIG. 8. The driving assistance device can acquire information on the blind spot and the surrounding environment of the blind spot for determining the environmental factors from information detected by an autonomous sensor mounted on the vehicle 10, map information stored in advance by the vehicle 10, or acquired from a server via a network. The factor coefficients are determined in advance from statistically processed data. In addition, different environmental factors and factor coefficients are selected depending on the type of blind spot. For example, when the detected blind spot is a blind spot created by an intersection, β is selected as the factor coefficient, and the environmental factor is selected according to the right column of the table of FIG. 8. These pieces of information for calculating the individual risk estimates are included in the risk estimation information stored in the driving assistance device.

[0027] 5. Modifications to the calculation of composite risk estimates A modified example of a method for calculating a composite risk estimate value is shown. In this example, as shown in FIG. 9, a pre-evaluation section H is provided up to a point (point P1 in FIG. 7) where the driving assistance device detects a blind spot, and when a predetermined condition for the pre-evaluation section H is satisfied, the composite risk estimate value is changed to a maximum value instead of the sum of the individual risk estimate values ​​of the blind spots included in the composite risk. The predetermined condition for the pre-evaluation section H is that the sum of the number of pedestrians and the number of oncoming vehicles detected while the vehicle 10 is traveling in the pre-evaluation section H is equal to or less than a certain threshold value A. The satisfaction of this condition means that the risk of the entire road is small, and it becomes relatively easy for the vehicle 10 to avoid a moving object that jumps out from a blind spot by emergency braking or the like, so that it is possible to make preparations for the composite risk relatively small. Therefore, by reducing the composite risk estimate value to the same extent as when one blind spot of the composite risks exists, excessive deceleration assistance can be suppressed, and driving can be made comfortable for the driver.

[0028] The number of pedestrians and the number of oncoming vehicles can be detected by a camera or LiDAR mounted on the vehicle 10. The threshold A may be a value that is changed depending on the environment of the road on which the vehicle 10 travels. For example, in a rainy situation where visibility is poor or in a situation where moving objects are difficult to detect due to snowfall, the threshold A may be set small in consideration of the possibility that the risk due to blind spots may be increased. The point where the pre-evaluation section H starts may be stored in advance by the driving assistance device as map information, or a new pre-evaluation section H may be started each time an intersection is detected.

[0029] 6. Variations in the assessment of complex risks In the above description, a compound risk is determined when the lateral position of the vehicle 10 cannot return to the original driving lane after passing one blind spot before passing the next blind spot. However, there may be cases where the vehicle 10 is not originally traveling in the leftmost lane and does not need to move laterally to secure a margin, or where the vehicle is originally traveling outside the lane to avoid an object different from the object that creates the blind spot. In this modified example, if the vehicle 10 is not traveling within the leftmost lane at the time of blind spot detection, multiple blind spots are not treated as a compound risk regardless of the distance between the blind spots.

[0030] FIG. 10 is an example of a scene that is not determined to be a compound risk in the modified example. In the upper diagram, the vehicle 10 is already traveling outside the lane when the driving support device detects the blind spot in order to secure a lateral distance from the pedestrian. In the lower diagram, the vehicle 10 moves to the next lane because of a construction site, and is no longer traveling in the leftmost lane when the driving support device detects the blind spot. In such a case, the risk of the vehicle 10 traveling on the road is more affected by factors other than the blind spot. In addition, since the lateral distance from the blind spot is already secured, it is considered that there is no need to increase the control amount. Therefore, in this modified example, the driving support device does not treat such a scene as a compound risk even if multiple blind spots are detected. This allows control to be more in line with the driver's senses. [Explanation of symbols]

[0031] 10 Vehicles

Claims

1. A driving assistance device that assists driving of a vehicle by controlling the vehicle in response to a blind spot detected in front of the vehicle, When a plurality of blind spots are detected in front of the vehicle and the plurality of blind spots satisfy a predetermined condition, the plurality of blind spots are determined to be a single compound risk; When the above-mentioned combined risks exist, Calculating an individual risk estimate value indicating the level of risk of each of the plurality of blind spots and a composite risk estimate value indicating the level of risk of the entire composite risk; determining a control end position for ending the control of the vehicle corresponding to the blind spot as a position passing through the blind spot having the highest individual risk estimate value; determining a target passing speed of the vehicle at the control end position in accordance with the composite risk estimate value; The composite risk estimate is the sum of the individual risk estimates of the blind spots included in the composite risk. A driving assistance device comprising:

2. The driving assistance device according to claim 1, The predetermined condition is: When only a first blind spot included in the plurality of blind spots is present, the travel trajectory of the vehicle calculated by the driving assistance device is one in which the vehicle deviates from an original travel lane in which the vehicle was traveling before the plurality of blind spots were detected, and the vehicle does not return to the original travel lane before passing a second blind spot that is located next to the first blind spot among the plurality of blind spots. A driving assistance device comprising:

3. The driving assistance device according to claim 2, The predetermined condition further includes: The original driving lane is the leftmost lane, and the position where the vehicle was traveling before the plurality of blind spots were detected is within the leftmost lane. A driving assistance device comprising:

4. A driving assistance device according to any one of claims 1 to 3, When the sum of the number of pedestrians and the number of oncoming vehicles detected in a pre-evaluation section provided until the driving assistance device detects a blind spot is equal to or less than a predetermined threshold, the composite risk estimate value is changed to a maximum value of the individual risk estimate values ​​of the blind spot included in the composite risk. A driving assistance device comprising:

5. A driving assistance method for assisting driving of a vehicle by controlling the vehicle in response to a blind spot detected ahead of the vehicle, the method comprising: When a plurality of blind spots are detected in front of the vehicle and the plurality of blind spots satisfy a predetermined condition, the plurality of blind spots are determined to be a single compound risk; Obtaining risk estimation information for calculating a risk estimation value indicating a level of risk of the blind spot; calculating a risk estimate for each of the blind spots based on the risk estimation information; For the composite risk, a control end position for ending the control of the vehicle corresponding to the blind spot is determined to be a position passing through the blind spot with the highest calculated risk estimate value, and a target passing speed of the vehicle at the control end position is determined according to the total value of the calculated risk estimate values. A driving assistance method comprising:

Images