JP5381665B2 - Vehicle control device - Google Patents

Description

  The present invention relates to a vehicle control device that controls a vehicle.

  As a conventional vehicle control device, an object detection unit that detects a position of a target object, and a target object according to a correspondence situation between the position of the target object detected by the object detection unit and the position of the target object detected in the past A certainty factor calculating means for calculating a certainty factor representing the certainty of the existence degree, an environment confirming means for recognizing the object detection environment of the object detecting means, and correcting the certainty factor or the threshold value of the certainty factor according to the object detection environment What can be known is known (for example, refer to Patent Document 1). This vehicle control device can make the existence probability of the target object appropriate by considering the position of the target object detected in the past.

JP-A-7-209410

  However, in the above-described vehicle control device, although it is possible to perform an appropriate process for an object detected in the past, it is appropriate for a newly detected object or an object that suddenly appears from a blind spot. It may not be possible to calculate the risk potential and support driving. In other words, depending on the situation around the host vehicle, the newly detected object is likely not to be falsely detected (for example, when an object jumps out of the blind spot), but the surrounding situation is not taken into account. It takes time to recognize the travel time and it takes time to perform the travel support, and thus it may not be possible to provide appropriate travel support.

  The present invention has been made to solve such a problem, and an object of the present invention is to provide a vehicle control device capable of performing appropriate driving support even when an object appears from a blind spot. And

  A vehicle control device according to the present invention includes an object detection unit that detects an object around the host vehicle, an object recognition unit that recognizes an object detected by the object detection unit, and a blind spot information acquisition unit that acquires information related to a blind spot with respect to the host vehicle. The object recognition means recognizes the object by calculating a certainty factor indicating the certainty of the existence of the object, and the object recognition means is an object in which the detected object appears from a blind spot. If it is determined that the detected object is not an object that has emerged from the blind spot, the certainty factor is calculated so as to increase the certainty factor.

  According to the vehicle control device of the present invention, the object recognizing unit can recognize the object detected by the object detecting unit by calculating the certainty factor indicating the certainty of the existence of the object. In addition, when the object recognition unit determines that the detected object is an object that has appeared from a blind spot, the object recognition unit increases the certainty factor compared to a case where it is determined that the detected object is not an object that has appeared from the blind spot. Can be calculated as follows. Since the object recognition means can calculate to increase the certainty factor, for an object that suddenly appears from the blind spot, shorten the time required from detection to recognition by the object detection means. Can do. By shortening the time until recognition, it is possible to perform the subsequent processing at an early stage, so that adequate travel support is provided by securing sufficient time even for objects that suddenly appear from the blind spot. Can do.

  Further, in the vehicle control device according to the present invention, the blind spot information acquisition unit sets a start point of the blind spot area, and the object recognition unit includes an object that has been detected within a predetermined range from the blind spot start point and an object that has appeared from the blind spot. It is preferable to determine. In this way, the blind spot start point is set, and it is uniformly determined whether a newly detected object has appeared from the blind spot depending on whether or not it exists within a predetermined range from the blind spot start point. Thus, the calculation load can be reduced.

  In the vehicle control device according to the present invention, the object recognition unit recognizes an object based on the detection of the L-shaped edge by the object detection unit, and the blind spot information acquisition unit recognizes based on the detection of the L-shaped edge. The formed object is preferably set as a blind spot generation obstacle that generates a blind spot area. An L-shaped edge can be detected for a wall or a building that can be an obstacle that generates a blind spot area. By setting an object recognized based on detection of the L-shaped edge as a blind spot generation obstacle that generates a blind spot area, the blind spot can be grasped early.

  Further, the vehicle control apparatus according to the present invention further includes a travel support unit that supports the travel of the host vehicle with respect to the object recognized by the object recognition unit, and the object detected by the object recognition unit appears from a blind spot. When it is determined that the object is an object, it is preferable that the driving support means advance the start time of driving support compared to a case where it is determined that the detected object is not an object that has appeared from the blind spot. When an object suddenly appears from a blind spot, the driving support means can advance the start time of driving support compared to the normal time. Therefore, for an object that suddenly appears from the blind spot, the time required from recognition to the start of driving support can be shortened. By shortening the time until the driving support, it is possible to secure a sufficient time even for an object that suddenly appears from the blind spot and perform appropriate driving support.

  Further, the vehicle control apparatus according to the present invention further includes a travel support unit that supports the travel of the host vehicle with respect to the object recognized by the object recognition unit, and the object detected by the object recognition unit appears from a blind spot. When it is determined that the object is an object, the driving support means preferably increases the intervention amount for driving support as compared to the case where it is determined that the detected object is not an object that has emerged from the blind spot. When an object suddenly appears from a blind spot, the driving support means can increase the amount of driving support intervention. Accordingly, it is possible to perform appropriate driving support that can reliably avoid an object that suddenly appears from the blind spot.

  According to the present invention, it is possible to perform appropriate driving support even when an object appears from a blind spot.

It is the figure which showed the block configuration of the vehicle control apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the surrounding situation of the own vehicle. It is a figure which shows an example of the surrounding situation of the own vehicle. It is a flowchart which shows the processing content of the vehicle control apparatus which concerns on embodiment of this invention.

  Hereinafter, a preferred embodiment of a vehicle control device according to the present invention will be described in detail with reference to the drawings.

  First, the structure of the vehicle control apparatus 1 which concerns on embodiment of this invention is demonstrated. FIG. 1 is a diagram showing a block configuration of a vehicle control device 1 according to an embodiment of the present invention. 2 and 3 are diagrams illustrating an example of a situation around the host vehicle M1. For example, as shown in FIG. 2, the vehicle control device 1 recognizes an obstacle B2 that suddenly appears from the blind spot of the obstacle B1, calculates a risk level for the object, and performs a driving support function based on the risk level. Have. As shown in FIG. 1, the vehicle control device 1 includes an ECU (Electronic Control Unit) 2, an object detection unit (object detection means) 3, a road shape detection unit 6, a traveling state detection unit 7, a display device 12, and a sound device 13. The support device 14 is provided. In the present invention, “detection” of an object means a state in which a signal from a radar is input but the presence of the object is not recognized as an apparatus (for example, erroneous detection of a radar or noise may be included). The “recognition” of an object is a state in which the accuracy of detection is confirmed by performing an operation based on the detection result of the radar, and the presence of the object is recognized as a device.

  The object detection unit 3 has a function of detecting an object existing around the host vehicle M1. Objects that can be detected by the object detection unit 3 include obstacles that generate blind spots such as walls, structures, and stopped vehicles, and pedestrians, bicycles, two-wheeled vehicles, and other vehicles that exist around the host vehicle M1, for example. Etc. In the example shown in FIGS. 2 and 3, an obstacle B1 that generates a blind spot area and an obstacle B2 that suddenly appears from the blind spot area are described as objects. The object detection unit 3 can acquire information such as position information and size of the detected object. The object detection unit 3 is configured by, for example, a millimeter wave radar or a laser radar. The object detection unit 3 has a function of outputting information related to the detected object to the ECU 2.

  The road shape detection unit 6 has a function of detecting the road shape on which the host vehicle M1 is traveling. The road shape detection unit 6 is configured by, for example, a laser radar. The road shape detection unit 6 has a function of outputting the detected information to the ECU 2. The traveling state detection unit 7 has a function of detecting the traveling state of the host vehicle M1. The traveling state detection unit 7 is configured by, for example, a wheel speed sensor and detects a vehicle speed, and is configured by a yaw rate sensor and has a function of detecting the yaw rate. The traveling state detection unit 7 has a function of outputting the detected traveling state to the ECU 2.

  The display device 12 has a function of displaying a message, a warning or the like to the driver. The display device 12 includes a liquid crystal display provided in the vehicle, a meter display unit, and a HUD. The display device 12 performs display based on a control signal from the ECU 2. The voice device 13 has a function of notifying a driver of a message or warning by voice. The audio device 13 includes a speaker and an alarm provided in the vehicle. The sound device 13 emits sound based on a control signal from the ECU 2. The support device 14 has a function of supporting driving of the host vehicle M1. The support device 14 is configured by a brake or an EPS actuator, and has a function of performing driving support based on a control signal from the ECU 2.

  The ECU 2 is an electronic control unit that performs overall control of the vehicle control device 1. The ECU 2 is mainly configured by a CPU, for example, and includes a ROM, a RAM, an input signal circuit, an output signal circuit, a power supply circuit, and the like. The ECU 2 includes an object recognition unit (object recognition unit) 16, a blind spot information acquisition unit (dead zone information acquisition unit) 17, a risk level calculation unit 18, a travel route setting unit 19, and a travel support unit (travel support unit) 21. Has been.

  The object recognition unit 16 has a function of recognizing an object detected by the object detection unit 3. The object recognition unit 16 can recognize the object by calculating a certainty factor indicating the certainty of the presence of the detected object. The object recognition unit 16 can recognize an object when the calculated certainty factor is equal to or greater than a predetermined threshold. The certainty factor can be calculated by tracking the object a plurality of times at intervals of several tens of milliseconds by the object detection unit 3. For example, the certainty factor can be obtained by calculating the existence probability of an object by one tracking and adding a value obtained based on the existence probability for each tracking. Alternatively, the existence probability can be increased for each tracking, and the value obtained based on the existence probability can be used as the certainty factor. The existence probability can be set to a different value depending on the lateral position and distance of the object in consideration of the accuracy of the radar. In addition, the certainty factor may be obtained from the number of times of tracking on the object, or the certainty factor may be obtained from the time during which the object is continuously detected.

  The object recognition unit 16 has a function of determining whether or not a newly detected object has suddenly appeared from the blind spot. Furthermore, when the object recognition unit 16 determines that the detected object is an object that has appeared from the blind spot, the object recognition unit 16 can increase the degree of confidence compared to the case where it is determined that the detected object is not the object that has appeared from the blind spot. It has a function of calculating so as to increase. That is, the degree of certainty increases more easily when it is determined that the detected object is an object that has appeared from the blind spot than when it is determined that the detected object is not an object that has appeared from the blind spot. Become. In the following description, the processing when it is determined that the detected object is not an object that has appeared from the blind spot will be described as “normal time” processing. By performing the process so as to increase the certainty factor, the time required from the detection of the object to the recognition can be shortened. As a method of increasing the certainty level, a method for increasing the existence probability for an object, a method for increasing the recognition reliability for an object, a method for reducing the number of tracking times from when a new object is detected until it is recognized, or a new object is detected. There is a method of lowering the threshold of recognition determination time from recognition to recognition. For example, the initial value of the existence probability can be increased. Further, when the existence probability increases for each tracking, the increase rate can be increased. Further, the certainty factor can be easily increased by increasing the value of the predetermined coefficient used for the certainty factor calculation. When the object is recognized when the number of times of tracking for the object exceeds a predetermined threshold, the threshold can be lowered. Alternatively, the threshold can be lowered when the object is recognized when the time for which the object has been detected exceeds the threshold. Furthermore, when the object is recognized when the certainty factor exceeds the threshold value, the threshold value can be lowered as compared with the normal time. In this case, the certainty factor is relatively increased as compared with the normal time by lowering the threshold value.

  The blind spot information acquisition unit 17 has a function of acquiring information related to the blind spot for the host vehicle M1. Specifically, the blind spot information acquisition unit 17 determines whether or not the object recognized by the object recognition unit 16 generates a blind spot, and calculates the position of a blind spot area generated by the object. Can do. In this embodiment, the blind spot information acquisition unit 17 determines whether or not the object generates a blind spot by using a predetermined determination condition for the recognized object, and based on the position and shape of the object. The starting point of the blind spot area can be set. Detailed processing contents will be described later.

  The risk level calculation unit 18 has a function of calculating the risk level for the recognized object. The degree of danger indicates the degree of proximity between the host vehicle M1 and the object. The higher the degree of danger, the higher the possibility of contact. The travel route setting unit 19 has a function of setting a travel route on which the host vehicle M <b> 1 travels based on the calculation result of the risk calculation unit 18. The travel support unit 21 has a function of performing travel support of the vehicle based on the risk calculated by the risk calculation unit 18 and the travel route set by the travel route setting unit 19. Specifically, the travel support unit 21 can output a control signal to the support device 14 to control the brake and the accelerator so as to avoid an object or travel along a set travel route. The driving support unit 21 can output a control signal to the display device 12 or the audio device 13 to issue a message or a warning to the driver so as to avoid the object or follow the set driving route. .

  Next, with reference to FIG.3 and FIG.4, operation | movement of the vehicle control apparatus 1 which concerns on this embodiment is demonstrated. FIG. 4 is a flowchart showing the processing contents for obstacle recognition of the vehicle control apparatus 1 according to the embodiment of the present invention. In the present embodiment, processing contents in the situation shown in FIG. 3 will be described. That is, processing contents in a situation where a blind spot area DA exists due to an obstacle B1 such as a wall or a building in front of the host vehicle M1 and a new obstacle B2 appears in front of the host vehicle M1 will be described. This process is repeatedly executed in the ECU 2 at a predetermined timing during driving of the vehicle. The process of FIG. 4 is an example in the case where the object detection unit 3 is configured by a laser radar.

  As shown in FIG. 4, the object recognition unit 16 of the ECU 2 detects an L-shaped edge E existing in front of the host vehicle M1 based on the detection result of the object detection unit 3 (step S10). Next, the object recognition unit 16 performs tracking on the L-shaped edge E detected in S10 using the detection result of the object detection unit 3 (step S12). The object recognition unit 16 recognizes the obstacle B1 based on the result of S12 and acquires its position and size (step S14). At this time, the blind spot information acquisition unit 17 also acquires information on the position and size of the obstacle B1 for setting the blind spot area.

  The blind spot information acquisition unit 17 determines whether or not the obstacle B1 recognized in S14 generates a blind spot area, so that the obstacle B1 satisfies a predetermined determination condition. (Step S16). In S16, it is determined that the determination condition is satisfied when all of Condition 1, Condition 2, Condition 3, and Condition 4 are satisfied. Condition 1 is that the distance D between the obstacle B1 and the host vehicle M1 is equal to or less than a predetermined threshold value D0 (D ≦ D0). Condition 2 is that the speed V of the obstacle B1 is equal to or lower than a predetermined threshold value V0 (V ≦ V0). Condition 3 is that the distance L between the host vehicle M1 and the obstacle B1 is not less than a predetermined threshold L0 and not more than the threshold L1 (L0 ≦ L ≦ L1). Condition 4 is that the total number C of laser reflection points in a region behind the obstacle B1 is equal to or less than a predetermined threshold value C0 (C ≦ C0). Since the laser does not reflect behind the obstacle B1 that generates the blind spot, the number of reflection points is 0. However, the threshold C0 can be set to a value of 1 or more according to the measurement accuracy of the laser radar. If it is determined in S16 that the determination condition is not satisfied, it is determined that the obstacle B1 does not generate a blind spot area, and the process shown in FIG. 4 ends.

  On the other hand, if it is determined in S16 that the determination condition is satisfied, the blind spot information acquisition unit 17 sets the obstacle B1 as a blind spot generation obstacle capable of generating a blind spot area (step S18). In FIG. 3, a blind spot area DA is generated behind the obstacle B1. Next, the blind spot information acquisition unit 17 sets the blind spot start point SP of the blind spot area DA due to the obstacle B1 (step S20). The blind spot start point SP is set at the end of the L-shaped edge E. Further, the blind spot information acquisition unit 17 acquires the coordinates (x0, y0) of the blind spot start point SP.

  Next, the object recognition unit 16 determines whether or not a new obstacle B2 has been detected by the object detection unit 3 (step S22). If it is determined in S22 that a new obstacle is not detected, the process shown in FIG. 4 ends. On the other hand, when it is determined in S22 that a new obstacle B2 has been detected, the object recognition unit 16 causes the obstacle B2 to suddenly appear to determine whether or not the obstacle B2 has suddenly appeared from the blind spot area. It is determined whether or not the area CA exists (step S24). The sudden appearance area CA is an area within the radius R1 from the blind spot start point SP and is an area on the own vehicle M1 side (a semicircle area hatched in FIG. 3).

  When it is determined in S24 that the obstacle B2 does not exist in the sudden appearance area CA, the object recognition unit 16 determines that the obstacle B2 does not suddenly appear from the blind spot area DA, and the existence probability of the obstacle B2 Is set to P_low, which is a value used for normal obstacle recognition (step S28). Thereby, the object recognition unit 16 performs recognition in the same process as when a normal obstacle existing in front of the host vehicle M1 is recognized. The certainty of detection with respect to the obstacle B1 is the same as that in the normal time, and the time required from the detection of the obstacle B1 until it is recognized is the same as in the normal time.

  On the other hand, when it is determined in S24 that the obstacle B2 exists in the sudden appearance area CA, the object recognition unit 16 determines that the obstacle B2 has suddenly appeared from the blind spot area DA, and the presence of the obstacle B2 is present. The initial value P0 of the probability is set to P_high higher than P_low (step S26). As a result, the probability of existence of the obstacle B2 when calculating the certainty of detection for the obstacle B2 is increased, and the certainty can be increased as compared with the normal time. The object recognition unit 16 can shorten the time required from the detection of the obstacle B2 to the recognition by the fact that the certainty factor is likely to increase. When the process of S26 ends, the process shown in FIG. 4 ends.

  After the process shown in FIG. 4, the object recognition unit 16 calculates the certainty factor for the obstacle B2, and recognizes the presence of the obstacle B2 when the certainty factor exceeds a predetermined threshold. Thereafter, the risk level calculation unit 18 calculates the risk level for the recognized obstacle B2, and the travel route setting unit 19 sets an optimal travel route based on the risk level. Furthermore, the driving support unit 21 outputs a control signal to the support device 14 so as to drive on the driving route set by the driving route setting unit 19 and performs driving support of the host vehicle M1. Alternatively, when the degree of risk is equal to or greater than a certain level, the driving support unit 21 controls the support device 14 to turn, decelerate, or stop so as to avoid contact with the obstacle B2, and the display device 12 or the audio device Warning is performed at 13.

  As described above, in the vehicle control device 1 according to the present embodiment, the object recognition unit 16 recognizes the object detected by the object detection unit 3 by calculating the certainty factor indicating the certainty of the existence of the object. Can do. In addition, when the object recognition unit 16 determines that the detected object is an object that has appeared from the blind spot, the object recognition unit 16 can perform calculation so as to increase the certainty factor as compared with the normal time. The object recognizing unit 16 can calculate the certainty factor so as to increase as compared with the normal time. Therefore, for an object that suddenly appears from the blind spot, the object detecting unit 3 does not detect until the object is recognized. Can be shortened. By shortening the time until recognition, it is possible to perform the subsequent processing at an early stage, so that adequate travel support is provided by securing sufficient time even for objects that suddenly appear from the blind spot. Can do.

  In the vehicle control device 1 according to the present embodiment, the blind spot information acquisition unit 17 sets the start point SP of the blind spot area DA, and the object recognition unit 16 detects the object detected within a predetermined range from the blind spot start point SP. Can be determined as an object that has appeared from the blind spot. In this way, the blind spot start point SP is set, and it is uniformly determined whether the newly detected object has appeared from the blind spot depending on whether or not it exists within a predetermined range from the blind spot start point SP. By doing so, the calculation load can be reduced.

  In the vehicle control device 1 according to the embodiment, the object recognition unit 16 recognizes an object based on the detection of the L-shaped edge E by the object detection unit 3, and the blind spot information acquisition unit 17 performs the L-shaped edge E. The object recognized on the basis of the detection can be set as a blind spot generation obstacle that generates a blind spot area. An L-shaped edge E can be detected for a wall or building that may be an obstacle that generates the blind spot area DA. By setting an object recognized based on detection of the L-shaped edge E as a blind spot generation obstacle that generates the blind spot area DA, the blind spot can be grasped at an early stage.

  The vehicle control device according to the present invention is not limited to the above-described embodiment.

  For example, when an object suddenly appears from a blind spot, the start time of driving support may be advanced. Specifically, when it is determined by the object recognition unit 16 that the object detected by the object detection unit 3 is an object that has appeared from the blind spot, the travel support unit 21 starts the travel support start time compared to the normal time. Can be expedited. For example, when the risk calculating unit 18 calculates the risk for the object recognized by the object recognizing unit 16, and the driving support unit 21 starts driving support when the risk exceeds a predetermined threshold. When the object suddenly appears from the blind spot, the driving support start time can be advanced by lowering the threshold value. As a result, for an object that suddenly appears from the blind spot, the time required from recognition to the start of driving support can be shortened. By shortening the time until the driving support, it is possible to secure a sufficient time even for an object that suddenly appears from the blind spot and perform appropriate driving support.

  Further, when an object suddenly appears from the blind spot, when the object recognition unit 16 determines that the object detected by the object detection unit 3 is an object that has appeared from the blind spot, the driving support unit 21 is compared with the normal time. As a result, the amount of intervention for driving support can be increased. Specifically, when an obstacle suddenly appears from a blind spot, it is possible to increase the strength of steering and brakes related to driving support or to increase the warning to the driver. Accordingly, it is possible to perform appropriate driving support that can reliably avoid an object that suddenly appears from the blind spot.

  When a radar is used as the object detection unit, the depth of the obstacle B1 can be set from the edge attribute after recognizing the obstacle B1 that generates a blind spot.

  DESCRIPTION OF SYMBOLS 1 ... Vehicle control apparatus, 3 ... Object detection part (object detection means), 16 ... Object recognition part (object recognition means), 17 ... Blind spot information acquisition part (dead angle information acquisition means), 21 ... Travel support part (travel support means) ).

Claims (9)

Object detection means for detecting objects around the host vehicle;
Object recognition means for recognizing an object detected by the object detection means;
A blind spot information acquisition means for acquiring information relating to the blind spot for the host vehicle,
The object recognition means recognizes the object by calculating a certainty factor indicating the certainty of the existence of the object,
The blind spot information acquisition means determines whether or not the object recognized by the object recognition means is an object that generates a blind spot for the host vehicle, and a blind spot generated by an object that generates a blind spot for the host vehicle. Calculate the area,
The object recognition means, the detected object by the object detecting means, it is determined whether the object emerging from the blind spot region, it is determined that the detected object is an object that appeared from the blind spot region In the case where the detected object is not an object that has emerged from the blind spot area , the vehicle control device calculates so as to increase the certainty factor as compared to a case where it is determined that the detected object is not an object that has appeared from the blind spot area . The blind spot information acquisition means sets the starting point of the blind area,
The object recognition unit, the vehicle control apparatus according to claim 1, wherein determining that the detected object within a predetermined range from the start point emerged object from the blind spot region. The object recognition means recognizes an object based on detection of an L-shaped edge by the object detection means;
The blind spot information obtaining means, L-shaped the recognized object based on the detection of edges, the vehicle control apparatus according to claim 1 or 2, wherein the set as a blind spot product obstacle for generating the blind spot region . The vehicle further comprises travel support means for performing travel support of the host vehicle on the object recognized by the object recognition means,
By the object recognition unit, when a detected object is determined to be the object emerging from the blind spot region, the driving support unit has been determined that the object detected is not the object emerging from the blind spot region The vehicle control device according to any one of claims 1 to 3, wherein the start time of the driving support is advanced as compared with the case. The vehicle further comprises travel support means for performing travel support of the host vehicle on the object recognized by the object recognition means,
By the object recognition unit, when a detected object is determined to be the object emerging from the blind spot region, the driving support unit has been determined that the object detected is not the object emerging from the blind spot region The vehicle control device according to any one of claims 1 to 4, wherein an intervention amount for driving support is increased as compared with a case. The said object recognition means increases the said reliability by increasing the initial presence probability of the object detected by the said object detection means, It is any one of Claims 1-5 characterized by the above-mentioned. Vehicle control device. The vehicle control according to any one of claims 1 to 5, wherein the object recognition unit increases the certainty factor by increasing a recognition reliability for the object detected by the object detection unit. apparatus. The said object recognition means increases the said reliability by reducing the frequency | count of tracking after detecting an object by the said object detection means until it recognizes, The any one of Claims 1-5 characterized by the above-mentioned. The vehicle control device according to the item. 6. The object recognition unit according to claim 1, wherein the object recognition unit increases the certainty factor by lowering a threshold value of a recognition determination time from detection of an object by the object detection unit to recognition. The vehicle control device according to any one of claims.

Images