JP2013137710A - Obstacle detector - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an obstacle detector capable of providing driving assistance at appropriate timing when entering a curve.SOLUTION: An obstacle detector includes: object detection means 15, 16 which detect positions of stationary objects, preceding vehicles, and oncoming vehicles; curve obstacle detection means 13 which computes detection certainty of a curve obstacle located along a curved road on the basis of positional information of a stationary object; detection certainty correction means 13 which increases or decreases the detection certainty on the basis of either a distance between a preceding vehicle and the curve obstacle or a distance between an oncoming vehicle and the curve obstacle or both; flag setting means 13 which sets an assistance delay flag on when the detection certainty is greater than a threshold; and driving assistance system control means 13 which starts providing driving assistance when expected time to collision with a target object coincides with a first time period if the assistance delay flag is off, and starts providing driving assistance when expected time to collision with a target object coincides with a second time period, which is shorter than the first time period, if the assistance delay flag is on.

Description

  The present invention relates to an obstacle detection device that detects an obstacle around a vehicle.

  There is known a driving support device that supports a driver by mounting a radar device or the like on a vehicle and detecting an obstacle. However, in the detection of an obstacle by a radar, there is a case where an object that should not be detected as an obstacle may be erroneously detected, and a technique for suppressing the erroneous detection has been considered (for example, see Patent Document 1). In Patent Document 1, an obstacle is detected in the estimated lane more than the reference time in order to prevent erroneous recognition of a traveling vehicle in a different lane adjacent to the preceding vehicle without recognizing the preceding vehicle that entered the curve. A front obstacle collision damage reduction braking control device that performs braking control when present is disclosed.

  In addition to the preceding vehicles, there are laying objects such as guardrails that should not be detected as obstacles. Since the guardrail exists in front of the host vehicle at the entrance of the curve, the guardrail may be erroneously detected as a preceding vehicle, and an unnecessary alarm may be issued. For this reason, a technique for estimating a road shape and delaying or canceling an alarm or the like has been considered (for example, see Patent Document 2). Patent Document 2 discloses an obstacle detection device that obtains an approximate line from a detected stationary object and estimates an estimated R (road shape) from the approximate line.

JP 2009-101737 A JP 2007-230267 A

  However, if the road shape is estimated from the position information of the target (the object detected by the sensor, not necessarily a three-dimensional object) existing in the traveling direction of the host vehicle as in Patent Document 2, the road shape is erroneously determined. There is a problem of fear.

  FIG. 1 is an example of a diagram illustrating a road shape determination method in the prior art. A preceding vehicle that is determined to be likely to collide is traveling ahead of the host vehicle. However, if there is a target that easily reflects the radar in the irradiation range of the radar of the host vehicle, the host vehicle detects the target as an obstacle. In the figure, a roadside object (for example, a stopped vehicle or a signboard) is on the left front and right front of the host vehicle, and a road iron plate such as a manhole is on the front, and three targets are detected. In this case, if the shape connecting the positions of the three targets is similar to the curve shape and the distinction, the vehicle determines that it has detected a curve obstacle (for example, a guardrail) and delays an alarm or the like. There is a risk that the warning for the vehicle will be delayed.

  In view of the above problems, an object of the present invention is to provide an obstacle detection device that can perform driving support at an appropriate timing when entering a curve.

  The present invention relates to an object detection means for detecting the positions of a stationary object, a preceding vehicle, and an oncoming vehicle, and a curve obstacle for calculating a detection accuracy of a curved obstacle existing along a curved road shape based on the position information of the stationary object. An object detection unit, a detection accuracy correction unit that increases or decreases the detection accuracy based on at least one of a distance between the preceding vehicle and the curve obstacle, or a distance between the oncoming vehicle and the curve obstacle, and the detection accuracy is greater than a threshold value. If large, flag setting means for setting a support delay flag to ON, and if the support delay flag is OFF, driving support is started when the predicted collision time with the target is the first timing, and the support delay Driving support device control means for starting driving support when the predicted collision time with the target is a second timing that is later than the first timing when the flag is on. Providing an obstacle detection system according to claim.

  It is possible to provide an obstacle detection device that can perform driving support at an appropriate timing when entering a curve.

It is an example of the figure explaining the determination method of the road shape in a prior art. It is an example of the system block diagram of an obstacle detection apparatus. It is an example of the figure explaining the correction calculation of curve obstruction reliability. It is an example of the figure explaining the correction | amendment calculation of the curve obstruction reliability at the time of a preceding vehicle detection. It is an example of the flowchart figure which shows the procedure in which an obstacle detection apparatus calculates a curve obstacle reliability, and supports driving | operation.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.
FIG. 2 shows an example of a system configuration diagram of the obstacle detection apparatus. The obstacle detection device 100 includes a host vehicle information acquisition unit 12, an obstacle information acquisition unit 11, a system ECU 13, and an operation device 14. The present embodiment is characterized in that the system ECU (Electronic Control Unit) 13 corrects the reliability of the curve obstacle.

  The own vehicle information acquisition unit 12 is, for example, a yaw rate sensor 17 and a wheel speed sensor 18. The yaw rate sensor 17 detects an angular velocity at which an axis perpendicular to the axle rotates horizontally on the road surface, for example, by a change in electrostatic capacitance caused by movement of a vibration piece manufactured by MEMS. Further, the yaw angle, that is, the traveling direction can be detected by integrating the detection results. The wheel speed sensor 18 detects a wheel speed from a wheel pulse output when a magnetic body or the like arranged at a predetermined interval in the circumferential direction of the wheel is detected by the magnetic sensor. The vehicle speed can be obtained from the wheel speed and the wheel diameter.

  The obstacle information acquisition unit 11 includes a millimeter wave radar device 15 and a radar ECU 16. The millimeter wave radar device 15 is disposed at the front center of the vehicle such as the front grille of the vehicle. The millimeter wave radar device 15 emits a millimeter wave at a predetermined angle centered on the front of the vehicle, and receives the millimeter wave reflected by a target existing in this range.

  The radar ECU 16 is a microcomputer provided with a CPU, RAM, ROM, input / output I / F, A / D converter, and the like. The radar ECU 16 performs a stationary object / preceding vehicle determination and a preceding vehicle relative position / speed calculation by the CPU executing a program stored in the ROM. The radar ECU 16 calculates the distance from the millimeter wave radar device 15 to the target by measuring the time from emission to reception. The millimeter wave radar device 15 has a plurality of receiving antennas, and the receiving antennas that can be received by the radar ECU 16 are switched by a switch, so that the receiving antennas receive the signals in a time division manner. The direction of the target is calculated based on the receiving antenna that has received the strongest millimeter wave. Further, the radar ECU 16 detects that there is a difference in frequency between the emitted millimeter wave and the received millimeter wave by using the Doppler effect, and calculates the relative speed. In this way, the radar ECU 16 detects the distance, direction, and relative speed from the target.

  Further, the radar ECU 16 compares the detected relative speed of the target with the vehicle speed of the host vehicle detected by the wheel speed sensor 18, and determines whether the detected target is a stationary object or a preceding vehicle. For example, the traveling direction of the host vehicle with respect to the road surface is a positive direction of the vehicle speed. If the relative speed is negative and the absolute value is the degree of identification, it is determined to be a stationary object, otherwise it is determined to be a preceding vehicle or an oncoming vehicle. The preceding vehicle or the oncoming vehicle can be determined as an oncoming vehicle if the relative speed is negative and the absolute value is greater than the vehicle speed of the host vehicle, and a preceding vehicle otherwise.

  Further, the radar ECU 16 obtains a relative position with respect to the target determined as the preceding vehicle. The relative position can be determined by taking the own vehicle position as the origin and specifying the direction and distance of the preceding vehicle with respect to the own vehicle position.

  The operation device 14 is an alarm device, an automatic brake device, or the like. The alarm device warns of an abnormal approach to a target by an alarm sound or voice, lights a warning lamp of the meter ASSY, or displays a message or icon notifying the abnormal approach on the LCD. The automatic brake device includes a brake ECU, a brake actuator, a wheel cylinder, and the like, and supplies brake hydraulic pressure to the wheel cylinder by controlling the opening degree of the valve of the brake actuator. Thus, the host vehicle can be braked without the driver depressing the brake pedal. In addition, the actuation device 14 includes a pre-crash seat belt that winds up the seat belt due to abnormal approach to the target. Hereinafter, such operation of the operation device 14 is collectively referred to as driving assistance.

  The system ECU 13 is a microcomputer including a CPU, a RAM, a ROM, an input / output I / F, an A / D converter, and the like. The system ECU 13 performs a curve obstacle determination calculation, a curve obstacle reliability correction calculation, a device operation determination, and the like by the CPU executing a program stored in the ROM.

  As a device operation determination, the system ECU 13 compares a TTC (Time To Collision) with a target and a threshold T to perform a device operation determination. It means that the larger the TTC, the more room is left for the collision, and the smaller the TTC, the higher the possibility of collision. If the system ECU 13 determines that there is a high possibility of a collision with the target based on the TTC, the warning is issued by operating the alarm device and sounding the alarm, or operating the automatic brake device and giving the driver deceleration G. . Further, when the possibility of a collision becomes higher due to further approach to the target, the braking hydraulic pressure of the automatic brake device is increased to reduce the collision impact.

  In this embodiment, the criterion based on the TTC is delayed according to the detection result of the curve obstacle. That is, if a curve obstacle is not detected and driving assistance is started with TTC <threshold value T1, driving assistance is started with TTC <threshold value T2 when threshold obstacle is detected (threshold value T2 <threshold value). T1). By doing so, it is possible to suppress driving assistance by erroneously detecting a curve obstacle as a preceding vehicle at the entrance of the curve or the like.

-Curve obstacle reliability Curve obstacle determination calculation calculates the curve obstacle reliability and determines whether or not it is a curve obstacle. The curve obstacle reliability refers to the accuracy with which the arrangement of the target determined as a stationary object is similar to the arrangement of the obstacle installed on the curve. Here, obstacles installed on the curve include a guard rail, a reflector with a predetermined interval, a signboard, and the like. In the present embodiment, the curve includes a T-junction. Here, the obstacle installed on the curve has the following characteristics.
a) A plurality of targets are arranged. b) The intervals in the width direction of the plurality of targets are within a predetermined value.
c) The intervals in the traveling direction of the plurality of targets are within a predetermined value.
d) A plurality of targets are arranged in the width direction, and the larger the position in the width direction (for example, with reference to the left end), the farther the target is, or the positions of the plurality of targets in the width direction are substantially equal. e) The system ECU 13 in which the vehicle is traveling on a curve, for example, quantifies the determination result of each item as follows. The digitization is only an example.
a) Target number judgment value A
When N or more targets are detected A = 1
When less than N targets are detected A = number of detections / N
b) Width direction interval judgment value B
If the distance between all targets is less than or equal to the threshold value Dh, B = 1
Other than that, the number of targets whose interval between targets is not less than the threshold value Dh is subtracted from 1 by 0.1. Note that Dh is less than the installation interval of the reflectors.
When the interval between all targets is less than or equal to the threshold value Ds, C = 1
Otherwise, subtract 1 from the number of targets whose interval between targets is not equal to or less than the threshold value Ds × 0.1, where Ds is, for example, less than half of the installation interval of the reflectors d) Curve shape determination value D
When the target is farther away as the position in the width direction becomes larger D = 1
The positions of multiple targets in the width direction are almost equal D = 1
Otherwise D = 0.3
e) Curve running judgment value E
A radius R of the curve is calculated from the yaw rate detected by the yaw rate sensor 17 and the wheel speed detected by the wheel speed sensor 18. The curve running determination value E is set as follows.
Curve running judgment value E = 50 / radius R
* When radius R = 50m is the minimum and R = 50, E = 1.

The system ECU 13 calculates the curve obstacle reliability from the respective determination values calculated as described above. The curve obstacle reliability is calculated as follows, for example.
(I) Curve obstacle reliability = A x B x C x D x E
(Ii) Curve obstacle reliability = A + B + C + D + E
Therefore, in (i), the maximum value of the curve obstacle reliability is “1”, and in (ii), the maximum value of the curve obstacle reliability is “5”. Each determination value may be weighted. In this case, the maximum value changes according to the weighting, but can be normalized to a value such as 1 or 10 if necessary.

  First, the system ECU 13 sets the curve obstacle detection flag to ON if the curve obstacle reliability is greater than the threshold value α.

-Curve obstacle reliability correction calculation The system ECU 13 finally determines the presence or absence of a curve obstacle by correcting the curve obstacle reliability by a correction calculation.

  FIG. 3 is an example of a diagram illustrating a curve obstacle reliability correction calculation. The system ECU 13 creates a virtual line by connecting the positions of a plurality of targets detected as elements of a curved obstacle with adjacent targets. Therefore, the virtual line represents the schematic shape of the curve obstacle. In addition, the distance between the estimated line and the preceding vehicle, or the estimated line and the oncoming vehicle, in the direction parallel to the road direction before entering the curve from the preceding vehicle or the oncoming vehicle (that is, viewed from the own vehicle position) The distance is calculated.

  In FIG. 3A, the preceding vehicle is traveling in front of the curve obstacle, and the system ECU 13 may erroneously detect the curve obstacle as the preceding vehicle at the entrance of the curve. For this reason, it is effective to delay driving assistance as before. When a curve obstacle actually exists and the preceding vehicle is traveling along a curve, the vehicle travels along the curve, so the distance between the curve obstacle and the preceding vehicle is about the width. Therefore, if the inter-vehicle distance between the curve obstacle and the preceding vehicle is less than the width threshold, the curve obstacle is highly likely to be an actual curve obstacle.

  On the other hand, in FIG. 3B, the curve obstacle actually does not exist, but the system ECU 13 erroneously detects that the curve obstacle exists from the roadside object and the manhole. For this reason, the system ECU 13 delays the driving support, and there is a possibility that the driving support for the preceding vehicle is delayed. However, it is generally not possible for a preceding vehicle to travel over a curve obstacle. Therefore, if the inter-vehicle distance between the curve obstacle and the preceding vehicle is equal to or greater than the threshold value of the width, it is highly possible that the curve obstacle is not an actual curve obstacle. Therefore, the determination is as follows.

(i) When a preceding vehicle is detected
When all of the conditions (i) -1 and below are satisfied, the system ECU 13 corrects the curve obstacle reliability so as to increase.
a. Curve obstacle detection flag = ON
b. | Position of estimated line-position of preceding vehicle (distance between curve obstacle and preceding vehicle) | <threshold (for example, 5.0 m)
→ Increase curve obstacle reliability by, for example, 5%
When all the conditions of (i) -2 and below are satisfied, the system ECU 13 corrects the curve obstacle reliability to be small.
a. Curve obstacle detection flag = ON
b. | Position of estimated line-position of preceding vehicle (distance between curve obstacle and preceding vehicle) | ≧ threshold value (for example, 5.0 m)
→ Reduce curve obstacle reliability by 20%, for example
(i) -3 If neither condition is satisfied, the system ECU 13 maintains the curve obstacle reliability as it is.

  In FIG. 3C, there is actually no curve obstacle, but the system ECU 13 erroneously detects that there is a curve obstacle signal from the roadside object and the manhole. The oncoming vehicle is running slightly ahead of the curve obstacle. For this reason, the system ECU 13 delays the driving assistance, and there is a possibility that the driving assistance for the oncoming vehicle is delayed. However, it is unlikely that an oncoming vehicle will run in the immediate vicinity of a curve obstacle, so if the distance between the curve obstacle and the oncoming vehicle is less than the width threshold, the curve obstacle may not be an actual curve obstacle. High nature. Therefore, the determination is as follows.

(ii) When an oncoming vehicle is detected
When the condition of (ii) -1 or less is satisfied, the system ECU 13 corrects the curve obstacle reliability to be small.
a. Curve obstacle detection flag = ON
b. | Position of estimated line-position of oncoming vehicle (distance between curve obstacle and oncoming vehicle) | <threshold (for example, 5.0 m)
→ Reduce curve obstacle reliability by 20%, for example
(ii) -2 When the above condition is not satisfied, the system ECU 13 maintains the curve obstacle reliability as it is.

  As described above, by correcting the curve obstacle reliability, the accuracy of determining the curve obstacle can be improved as compared with the conventional technique.

  The system ECU 13 sets the device delay flag to ON if the corrected curve obstacle reliability is greater than the threshold value α. When the device delay flag is ON, it means that driving support is delayed, and when it is OFF, driving support is not delayed (driving support is started at a normal timing).

<Another example of correction of curve obstacle reliability when detecting preceding vehicle>
FIG. 4 is an example of a diagram illustrating a correction operation for curve obstacle reliability when a preceding vehicle is detected. In FIG. 4, a curve obstacle and a preceding vehicle are detected. In this case, the system ECU 13 does not simply detect the distance between the curved obstacle and the preceding vehicle as shown in FIG. 3A, but compares the shape of the curved obstacle with the traveling locus of the preceding vehicle. When the two shapes are similar, it is highly likely that the preceding vehicle is traveling along a curve, and the curve obstacle is highly likely to be an actual curve obstacle.

The comparison between the shape of the curve obstacle and the traveling locus of the preceding vehicle is performed as follows.
1. The position information of the preceding vehicle is acquired in time series.
2. The position information of the first preceding vehicle is matched with the position of the nearest curve obstacle, and the position information other than the first preceding vehicle is moved in the same direction by the same amount.
3. The position of the preceding vehicle in time series and the shortest distance to the estimated line are calculated, and the total value is recorded.
4.2 and 3 are performed on the position information of all preceding vehicles, and the total value is recorded.
5. The smallest total value is extracted from all the total values.
6). The smallest total value is compared with a threshold value, and if it is within the threshold value, it is determined that the shape of the curve obstacle is similar to the traveling locus of the preceding vehicle.

(i) When a preceding vehicle is detected
When all of the conditions (i) -1 and below are satisfied, the system ECU 13 corrects the curve obstacle reliability so as to increase.
a. Curve obstacle detection flag = ON
b. The shape of the curve obstacle is similar to the driving track of the preceding vehicle → Increase the curve obstacle reliability by, for example, 5%
(i) -2 If this condition is not satisfied, the system ECU 13 maintains the curve obstacle reliability as it is.

[Operation procedure]
FIG. 5 is an example of a flowchart illustrating a procedure in which the obstacle detection apparatus 100 calculates a curve obstacle reliability and supports driving. The procedure of FIG. 5 starts when the switch of the obstacle detection device 100 is turned on and the vehicle speed or the like satisfies the operation condition.

  The system ECU 13 determines whether or not a curve obstacle has been detected based on the distance, direction, and relative speed from the target acquired by the obstacle information acquisition unit 11 (S10). The detection of the curve obstacle here may be performed to the extent that a plurality of obstacles are detected almost at the same time. One guardrail may be detected as a plurality of obstacles, and an obstacle that may be a curve obstacle may be detected.

  When a curve obstacle is detected (Yes in S10), the system ECU 13 calculates a curve obstacle reliability (S20). The target used for calculating the curve obstacle reliability is only a stationary object. Based on the curve obstacle reliability, the system ECU 13 sets the curve obstacle detection flag to ON.

  Next, the system ECU 13 determines whether a preceding vehicle or an oncoming vehicle is detected (S30).

  When the preceding vehicle or the oncoming vehicle is not detected (No in S30), the system ECU 13 does not correct the curve obstacle reliability, so the process proceeds to step S40.

  When the preceding vehicle or the oncoming vehicle is detected (Yes in S30), the system ECU 13 corrects the curve obstacle reliability (S40). The correction method is as described above, and the curve obstacle reliability is increased, decreased, or maintained by this correction.

  Next, the system ECU 13 determines whether or not the curve obstacle reliability is greater than a threshold value α (S50). Therefore, if the preceding vehicle is within the threshold from the curve obstacle, the curve obstacle is easily determined as a curve obstacle. If the preceding vehicle is beyond the threshold and exceeds the threshold, the curve obstacle Is easily determined not to be a curve obstacle, and when the oncoming vehicle is within the threshold from the curve obstacle, it is easy to determine that the curve obstacle is not a curve obstacle.

  If the curve obstacle reliability is greater than the threshold value α (Yes in S50), the system ECU 13 sets the device delay flag to ON (S60). Thereby, when an actual curve obstacle is detected, driving assistance can be delayed and false alarms can be suppressed. If the curve obstacle reliability is not greater than the threshold value α (No in S50), the device delay flag remains OFF. When there is no curve obstacle, driving support can be started at a normal timing, and driving support can be prevented from being delayed with respect to the preceding vehicle and the oncoming vehicle.

  As described above, in this embodiment, in the obstacle detection device 100 that delays driving assistance when a curve obstacle is detected, driving assistance is provided to the preceding vehicle and the oncoming vehicle by correcting the curve obstacle reliability. Delays can be suppressed.

DESCRIPTION OF SYMBOLS 11 Obstacle detection part 12 Own vehicle information acquisition part 13 System ECU
14 Operating Device 15 Millimeter Wave Radar Device 16 Radar ECU
17 Yaw Rate Sensor 18 Wheel Speed Sensor 100 Obstacle Detection Device

Claims (1)

Object detection means for detecting positions of stationary objects, preceding vehicles and oncoming vehicles;
A curve obstacle detection means for calculating the detection accuracy of a curve obstacle existing along a curved road shape based on the position information of the stationary object;
Detection accuracy correction means for increasing or decreasing the detection accuracy based on at least one of the distance between the preceding vehicle and the curve obstacle, or the distance between the oncoming vehicle and the curve obstacle;
Flag setting means for setting the support delay flag to ON when the detection accuracy is larger than a threshold;
When the assistance delay flag is off, driving assistance is started when the collision prediction time with the target is the first timing, and when the assistance delay flag is on, the collision prediction time with the target is Driving assistance device control means for starting driving assistance when the second timing is later than one timing;
An obstacle detection device comprising:

Images