JP4026400B2 - Vehicle control device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a vehicle control apparatus that controls a vehicle device so as to protect an occupant during a collision, and more particularly to a vehicle control apparatus that effectively regulates these controls.
[0002]
[Prior art]
Conventionally, in a vehicle that performs vehicle follow-up control in which the speed of the host vehicle is controlled in accordance with the speed of the preceding vehicle to follow the preceding vehicle, the distance, speed, direction, etc. of the target (for example, the preceding vehicle) with respect to the host vehicle It is known to use a highly accurate millimeter wave radar sensor as a sensor for acquiring target information.
[0003]
This millimeter wave radar sensor scans a predetermined area in front of the vehicle with a radio wave at predetermined intervals (for example, 0.1 s), recognizes a target using a reflected wave from the predetermined area, and detects the target. The target information about is output. When this millimeter wave radar sensor cannot recognize an object temporarily even though it actually exists due to noise of reflected waves, etc., or conversely when it temporarily recognizes it as a target even though it does not exist And the recognition state of the target is not always stable.
[0004]
In the former case, the millimeter wave radar sensor estimates and outputs the target information in the current cycle based on the target information recognized in the previous cycle, thereby achieving continuity of the target information. Then, when the state in which the object is not recognized continues for a predetermined period, the millimeter wave radar sensor stops outputting the target information related to the object.
[0005]
On the other hand, in the latter case, when the noise of the reflected wave is generated in the vicinity of the peak position related to the actual target, two or more target information having the same distance information coexists. In the vehicle follow-up control, the preceding vehicle for the follow-up control should be one target, so it is necessary to specify the target information related to the preceding vehicle from these two or more pieces of target information. At this time, if the target information output due to noise is specified, the preceding vehicle will be lost.
[0006]
Therefore, in vehicle follow-up control, target information that has been recognized for the longest time, for example, from two or more pieces of target information is specified as target information related to the preceding vehicle. Or, for example, as described in Japanese Patent Application Laid-Open No. 2002-22832, when a target lane probability that is the probability that a target exists on the same lane as the host vehicle is introduced and two or more target information coexist The target information having a higher own lane probability in the previous cycle is specified as the target information related to the preceding vehicle, thereby preventing the preceding vehicle from being lost.
[0007]
[Problems to be solved by the invention]
In recent years, the risk of a vehicle colliding with an obstacle is predicted based on the detection result of a millimeter wave radar sensor, the vehicle is controlled so as to avoid the collision, and the impact at the time of a collision received by an occupant is alleviated. Collision predictive control that activates a predetermined occupant protection device before a collision has been proposed.
[0008]
In this collision prediction control, stepwise control is executed according to the relationship between the host vehicle and the obstacle detected by the millimeter wave radar sensor. For example, there is a risk that the host vehicle may collide with the obstacle. When it is judged that the situation is such that the collision can be avoided, an alarm is issued to alert the driver, and it is judged that the situation should be avoided because the situation is more dangerous. In such a case, the vehicle behavior is controlled so as to avoid the collision by controlling the brake device, and if it is determined that the collision cannot be avoided, the predetermined occupant protection device such as the seat belt device is Activated before.
[0009]
In such collision predictive control, it is necessary to identify some targets that may collide with the host vehicle from among many existing targets ahead of the vehicle. However, in most cases, since the target and the own vehicle are independently moving objects, it is difficult to realize such identification with high accuracy.
[0010]
In addition, since the collision prediction control is a safety system for protecting an occupant from a vehicle collision, the target information output from the millimeter wave radar sensor can be estimated information as described above. It is desirable to sufficiently consider the degradation of detection performance and detection accuracy. In other words, if the recognition state of the target by the millimeter wave radar sensor is unstable due to reflected wave noise or the like, or if the detection performance deteriorates due to contamination of the millimeter wave radar sensor, etc. It is necessary to regulate the execution of predictive control in some way.
[0011]
Therefore, the present invention can specify a target for prediction determination with high accuracy from among many targets recognized by the millimeter wave radar sensor when performing collision prediction control and irreversible occupant protection device control. An object of the present invention is to provide a highly reliable vehicle control device that can regulate the operation of various vehicle devices in accordance with the target recognition state by the millimeter wave radar sensor and the position of the target based on the traveling path of the host vehicle. And
[0012]
[Means for Solving the Problems]
  The object of the present invention is to detect an object in front of the host vehicle and detect a relationship between the object and the host vehicle.
  A collision prediction means for predicting a collision between the host vehicle and the object based on the relationship detected by the detection means;
  A plurality of vehicle devices operated based on the prediction result by the collision prediction means;
  Recognition state of the object by the detecting meansGood or badThe recognition index that representsThis is an index indicating how long the object is recognized in time series by the detection means.Continuity of recognition of the object by the detection means;An index indicating how stably the object is recognized by the detection meansRecognition index calculation means for calculating based on at least one of a variation in recognition of the object by the detection means and a reflection intensity of the radio wave, sound wave or light wave emitted from the detection means from the object. When,
  Out of multiple vehicle devicesFromDepending on the recognition index by the recognition index calculation means, the vehicle device subject to operation regulationSelectThis is achieved by a vehicle control device.
[0013]
According to the above invention, it is possible to realize collision prediction control with improved accuracy of collision prediction determination and improved reliability.
[0014]
  The collision prediction control performs a collision prediction determination for predicting a collision that can actually occur in advance, and activates a predetermined vehicle device based on the result of the prediction determination. In the collision prediction control, the millimeter wave radar A collision prediction determination is made based on the relationship between the host vehicle and the object detected by a detecting means such as a sensor. In such a collision prediction determination, when the recognition state of the object by the detection means is taken into consideration, the collision prediction determination according to the stability of the recognition state can be realized.. Further, the recognition state of the object is determined based on the continuity of recognition of the object by the detection unit, variation in recognition of the object by the detection unit, and the object of radio waves, sound waves, or light waves emitted by the detection unit. If the reflection intensity is at least one of the reflection intensities, control of the vehicle device with further improved reliability can be realized. That is, an object based on estimation based on the discontinuity of the object recognition, such as when the detection means cannot recognize the object previously recognized, or the position information of the object recognized in the immediately preceding cycle. Variations in object recognition, such as when the error between the position of the object and the actual recognition position is large, when the error occurs continuously in each period, or when the error varies greatly between periods Control of vehicle devices by taking into account the recognition status of the object indicated by the reflection intensity from the object, such as the magnitude of the reflection intensity of the radio wave from the object and the fluctuation of the reflection intensity between cycles It is possible to improve the accuracy.
[0015]
  Further, the object is as described in claim 2.The plurality of vehicle devices include a first vehicle device that alerts the occupant or the outside of the vehicle, and a reversible second vehicle device that protects the occupant,
When the recognition index by the recognition index calculation means exceeds the first threshold value and the detection means determines that the recognition state of the object is bad, the operation of the first and second vehicle devices is restricted, and When the recognition index calculated by the recognition index calculation means is smaller than the first threshold value but exceeds the second threshold value, and when it is determined that the recognition state of the object by the detection means is slightly poor, the first and second vehicle devices The vehicle control device according to claim 1, wherein the operation of the second vehicle device is restricted.Achieved by:
[0016]
  According to the above invention,Of the plurality of vehicle devices, the vehicle device subject to operation restriction can be changed stepwise in an appropriate manner according to the recognition index by the recognition index calculation means.
[0018]
  The above object is also claimed.3As described in the above, the position of the object based on the traveling path of the host vehicle grasped using the relationship detected by the detecting meansAnd the recognition index by the recognition index calculation meansAnd a collision object probability calculating means for calculating a probability that the object is a collision object with respect to the host vehicle based on
  The plurality of vehicle devices include: a first vehicle device that alerts the occupant or the outside of the vehicle; a reversible second vehicle device that protects the occupant; and an irreversible third vehicle device that protects the occupant. Including
  When the collision object probability calculated by the collision object probability calculation means is low, the activation determination threshold of the third vehicle device is set to a high level at which activation determination is difficult to be performed, and the first and second When the operation of the vehicle device is prohibited and the collision object probability calculated by the collision object probability calculation means is a medium level, the activation determination threshold value of the third vehicle device is set to a high level that is difficult to determine the activation. In addition, when the operation of the second vehicle device is prohibited, the operation of the first vehicle device is permitted, and the collision object probability calculated by the collision object probability calculation means is high, the third vehicle device The vehicle control device according to claim 1 or 2, wherein the vehicle device activation determination threshold is set to a low level and the operation of the first and second vehicle devices is permitted.
[0021]
  Claims2Or3In the vehicle control device described above, preferably, the second vehicle device includes at least one of a brake device and a seat belt device.3The vehicle device includes an airbag device.
[0022]
  Claims 1 to4The vehicle control device according to claim 1,5If the relationship between the object and the host vehicle includes at least the relative distance, relative speed, and direction of the object with respect to the host vehicle, the collision prediction determination is performed accurately. Can do.
[0023]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a schematic configuration diagram of a control system according to the present invention. A vehicle control device 80 according to the present invention is configured around an electronic control unit 10 for vehicle control (hereinafter referred to as “vehicle control ECU 10”), and includes a millimeter wave radar sensor 14, collision prediction control, and an irreversible occupant protection device. Various ECUs 16A, 16B, 16C,... For controlling various devices (not shown) for realizing the control. These various ECUs are connected to the vehicle control ECU 10 via, for example, a multiplex communication network CAN18 so as to be capable of bidirectional communication.
[0024]
Although not shown, the millimeter wave radar sensor 14 is disposed, for example, in the vicinity of the front grill of the vehicle or inside the front bumper so as to monitor the front of the vehicle, and emits a millimeter wave band toward a predetermined area in front of the vehicle. A transmitting unit that radiates; and a receiving unit that receives a reflected wave from a target such as a vehicle or an obstacle in the region. The front surfaces of the transmission unit and the reception unit may be covered with a protective member made of a resin that efficiently transmits radio waves.
[0025]
The millimeter wave radar sensor 14 acquires target information indicating the relationship between the target and the host vehicle, for example, a relative speed Vr, a relative distance Lr, and an azimuth of the target with reference to the host vehicle in a predetermined cycle. The millimeter wave radar sensor 14 measures the relative velocity Vr of the target using the Doppler frequency of the radio wave, for example, by a two-frequency CW (Continuous Wave) method, and measures the relative distance Lr of the target from the phase information of the two frequencies. May be. Further, the millimeter wave radar sensor 14 may detect the azimuth of the target by scanning the radiation beam one-dimensionally or two-dimensionally.
[0026]
FIG. 2 is a power spectrum diagram used by the millimeter wave radar sensor 14 for target determination. This power spectrum is obtained by emitting a signal obtained by modulating two frequencies in a time division manner from the transmission unit of the millimeter wave radar sensor 14, mixing the transmission signal with the reception signal reflected from the target, and performing processing such as FFT. Can be obtained by:
[0027]
When the spectrum peak exceeds the target determination threshold as shown in FIG. 2, the millimeter wave radar sensor 14 determines that there is a target to be subjected to collision prediction determination described later, and relates to the spectrum peak. Target information is generated based on the frequency and phase difference.
[0028]
When the millimeter wave radar sensor 14 recognizes the target once in a certain cycle, the millimeter wave radar sensor 14 estimates the position of the target in the subsequent cycle based on the target information acquired in the cycle. Then, the millimeter wave radar sensor 14 checks whether or not a spectrum peak appears in the vicinity of the estimated position in the power spectrum diagram acquired in the subsequent period. When a spectrum peak appears, the millimeter wave radar sensor 14 outputs the target information generated based on the spectrum peak as target information related to the same target recognized in the previous cycle. In this way, when the same target can be continuously recognized in each subsequent cycle, the recognition state of the target by the millimeter wave radar sensor 14 is stable.
[0029]
On the other hand, if no spectral peak appears in the vicinity of the estimated position in the power spectrum diagram of the subsequent period due to reflected wave noise, the millimeter wave radar sensor is based on the target information acquired in the previous period. The target information related to the target estimated in this way is output as estimation information related to the same target recognized (or estimated) in the previous cycle. Thus, when the same target cannot be continuously recognized in each subsequent cycle, the recognition state of the target by the millimeter wave radar sensor 14 is not stable. Note that these target information (including estimation information) is transmitted to the vehicle control ECU 10 as detection data at the predetermined cycle.
[0030]
In this embodiment, the millimeter wave radar sensor 14 that can detect the target information with high accuracy without being relatively affected by weather conditions such as rain and fog is used as means for acquiring the target information ahead of the vehicle. However, the present invention is not particularly limited to this, and may be a sensor using laser light, ultrasonic waves, or the like.
[0031]
The vehicle control ECU 10 includes a microcomputer as a main component and is connected to various sensors such as a vehicle speed sensor, a steering angle sensor, and a throttle sensor. The microcomputer of the vehicle control ECU 10 is configured around a CPU, and includes a ROM that stores a predetermined control program and the like, a RAM that temporarily stores data, and an input / output interface (I / O). .
[0032]
Based on the detection data from the millimeter wave radar sensor 14, the vehicle control ECU 10 performs a collision prediction determination as to whether or not the host vehicle collides with a target in front of the vehicle. The collision prediction determination may be performed by comparing the relationship between the host vehicle and the target detected by the millimeter wave radar sensor 14 with a determination map, for example.
[0033]
Here, an example of the collision prediction determination using the determination map will be described in detail with reference to FIG.
[0034]
In the determination map shown in FIG. 3, the horizontal axis is the relative speed Vr (however, the direction in which the target approaches the host vehicle is positive), and the vertical axis is the relative distance Lr. In this determination map, the area α is an inevitable collision area in which it is determined that the vehicle collides with the target, and the area β needs to alert the occupant, but the vehicle can be avoided without colliding with the target. The determined collision avoidable area, area γ, is a safe area where it is determined that the vehicle can be avoided without colliding with the target.
[0035]
The collision area α and the collision avoidance area β are separated by a curve formed using, for example, a brake braking curve BL and a handle curve HA. The higher the relative speed with the target, the higher the probability that it will collide with the target even when braking with the brake. Such a boundary is the brake braking curve BL. If the detection data from the millimeter wave radar sensor 14 is below the brake braking curve BL, the collision probability is high. Similarly, as the relative speed with respect to the target increases, the probability that the driver will collide with the target increases even if the driver tries to avoid the collision by operating the steering wheel. Such a boundary is the handle curve HA. When the detection data from the millimeter wave radar sensor 14 is below the handle curve HA, the probability of collision is high.
[0036]
The collision area α is set as an area below the brake braking curve BL and the handle curve HA. Therefore, the collision area α is an area where it is determined that the vehicle collides with the target with a very high probability even when the vehicle is braked with the brake and the steering wheel is operated.
[0037]
Similarly, the safe area γ and the collision avoidance area β are delimited by a curve formed using the brake braking curve BL ′ and the handle curve HA ′. The collision avoidance region β is set to a region that belongs above the brake braking curve BL and the handle curve HA but belongs below the brake braking curve BL ′ and the handle curve HA ′.
[0038]
The vehicle control ECU 10 confirms the relative speed Vr and the relative distance Lr from the detection data from the millimeter wave radar sensor 14 at predetermined intervals, and determines a specific point on the determination map of FIG.
[0039]
When this specific point shifts from the region γ to the collision avoidance region β, the vehicle control ECU 10 executes the operation of the level 1 vehicle device. This level 1 vehicle device may be an alarm device that issues an alarm to alert the occupant or a hazard lamp that flashes to alert the following vehicle.
[0040]
When the specific point shifts from the collision avoidance region β to the collision region α, the vehicle control ECU 10 executes the operation of the level 2 vehicle device. In this level 2 vehicle device, a motor-type seat belt device to which a pretensioner function for removing seat belt slack by operating a motor or the like is added, and a bumper configured to be movable using a motor or the like are pushed forward. It may be a reversible occupant protection device such as a bumper moving device and a seat moving device that operates a motor or the like to return the seat position to the standard. These reversible occupant protection devices are devices that can protect the occupant in preparation for an actual collision. This reversible occupant protection device can be restored to its normal state by canceling the operation of the motor, etc., that is, by stopping or reversing it to return it to the standard position and releasing the previous operation. it can. The drive source is not limited to a motor, and hydraulic pressure or the like may be used.
[0041]
Furthermore, the operation of the level 2 vehicle device is an operation of a steering device that automatically steers the steering wheel in a direction to avoid a collision, or an operation of a brake device or a throttle drive device that decelerates the vehicle to avoid a collision. It may be.
[0042]
This determination map can be created based on data obtained by performing, for example, a vehicle collision test or simulation. The determination map is preferably stored in a ROM or the like accessed by the vehicle control ECU 10.
[0043]
As another collision prediction determination, a position vector representing the position of the target with reference to the host vehicle may be used. In this case, the vehicle control ECU 10 calculates the position vector of the target from the detection data from the millimeter wave radar sensor 14 every predetermined period, and uses the difference between the position vector of the period and the position vector of the previous period. Thus, the position vector of the target in the next cycle is estimated. On the other hand, the vehicle control ECU 10 predicts the movement vector of the host vehicle in the next cycle from the detected values from the steering angle sensor and the vehicle speed sensor output every predetermined cycle. And although detailed description is omitted, the vehicle control ECU 10 performs the collision prediction step by step from the relationship of these vectors, and operates the level 1 and level 2 vehicle devices stepwise as described above.
[0044]
When an actual collision occurs after execution of the above-described collision prediction control, the airbag ECU 16A activates the level 3 vehicle device. This level 3 vehicle device is an irreversible occupant protection device such as an air bag device that instantly deploys a bag using explosives, a seat belt device with a pretensioner that winds up a seat belt using explosives, a vehicle It may be a vehicle front airbag device for protecting pedestrians mounted on the front. Further, the level 3 vehicle device may be a pop-up hood or an emergency brake.
[0045]
The airbag ECU 16A determines whether or not to activate the level 3 vehicle device. Although not shown, this activation determination is realized using, for example, an acceleration signal from a touch sensor embedded in a bumper or an acceleration sensor arranged in front of a vehicle floor or a side member.
[0046]
Here, an example of activation determination using the activation determination map will be described with reference to FIG.
[0047]
In FIG. 4, three activation determination thresholds TH1, TH2, and TH3 are shown together with the acceleration signal. The airbag ECU 16A determines whether or not the acceleration signal exceeds the activation determination threshold, and activates the level 3 vehicle device with a predetermined operation output when the acceleration signal exceeds the activation determination threshold.
[0048]
The activation determination threshold used for actual determination may be appropriately selected from the three activation determination thresholds TH1, TH2, and TH3. For example, when the relative speed Vr of the target with respect to the host vehicle is very large, the low level activation determination threshold TH3 may be selected to ensure that the level 3 vehicle device is activated. Alternatively, while the above-described collision prediction control is not being executed, the high-level activation determination threshold value TH1 may be selected, so that unnecessary activation of the level 3 vehicle device is prevented.
[0049]
Collision predictive control performed by the vehicle control ECU 10 and the airbag ECU 16A and control of the irreversible occupant protection device as described above are known to those skilled in the art, and the contents of the collision predictive control performed by the vehicle control ECU 10 and the like, Since there are a wide variety of configurations, collision prediction determination methods, and irreversible occupant protection device activation determination methods that realize these, further description is omitted here. However, the present invention is applicable to all types of collision prediction control and irreversible occupant protection device control.
[0050]
By the way, the collision prediction control is executed based on the target information indicating the relationship between the target detected by the millimeter wave radar sensor 14 and the host vehicle, as is apparent from the above. Also in the control of the irreversible occupant protection device, the relationship between the target detected by the millimeter wave radar sensor 14 and the host vehicle can be used, for example, when selecting the activation determination threshold.
[0051]
For this reason, it is necessary to accurately specify a target whose relationship with the host vehicle is to be monitored from among a large number of targets recognized by the millimeter wave radar sensor 14. That is, it is necessary to exclude target information related to a target that cannot be a colliding object from the target in the above-described collision prediction determination or the like.
[0052]
Further, as described above, the recognition state of the target by the millimeter wave radar sensor 14 is not always stable due to noise or the like. Further, since the millimeter wave radar sensor 14 is mounted outside the vehicle room exposed to rain, snow, etc. as described above, moisture or dirt adheres to the protective member that protects the antenna portion while the vehicle is running, etc. Performance may be degraded.
[0053]
Further, the target information output to the vehicle control ECU 10 includes target information by estimation as described above, but in most cases, the host vehicle and the target are objects that move independently, The movement of the target relative to the host vehicle is not always stable.
[0054]
By taking these matters into consideration, the present invention further improves the reliability of the above-described collision prediction control and control of the irreversible occupant protection device.
[0055]
Specifically, as a first matter, accurate identification of a target whose relationship with the own vehicle is to be monitored is realized by considering the position of the target with reference to the traveling route of the own vehicle. This consideration may involve calculation of the probability that the target recognized by the millimeter wave radar sensor 14 will actually be a collision object in relation to the traveling path of the host vehicle, that is, the collision object probability Pro, as will be described next. . This collision object probability Pro is based on the idea that a target that already exists on the traveling path of the host vehicle or can move to the traveling path can be a collision object.
[0056]
Here, an example of a method of calculating the collision object probability Pro using the map will be described with reference to FIG. However, it should be understood that the collision object probability Pro does not necessarily have to be calculated using such a map.
[0057]
In this map, the area Α is formed in the vicinity of the host vehicle X and has the highest collision object probability Pro (for example, collision object probability Pro = 100%). The area Β is an area that is formed far away from the host vehicle X and has the second highest collision object probability Pro (for example, collision object probability Pro = 80%). The region Γ is formed on both sides of the region Α and the region Β and has the third highest collision object probability Pro (for example, the collision object probability Pro = 60%). Similarly, although illustration is omitted, region Δ, region Ε... May be set corresponding to collision object probability Pro = 30%, 10%.
[0058]
The position of the target on the map is determined at predetermined intervals based on the target information from the millimeter wave radar sensor 14. At this time, the position of the target determined on the map may be corrected in consideration of the curvature of the road on which the host vehicle X is traveling. For example, when the traveling route of the host vehicle X is a curve (that is, when traveling on a curved road), the position of the target determined on the map is estimated based on information obtained from a steering angle sensor or the like. It is corrected using the curvature of.
[0059]
The collision object probability Pro given to each target is calculated based on the position of each target determined on the map. The collision object probability Pro finally given to each target may be a value obtained by filtering the collision object probability Pro calculated for each period in this way.
[0060]
In the map of FIG. 5, the area Α becomes wider as it approaches the host vehicle X. This is because the closer the target is to the own vehicle X, the smaller the influence of the curvature and the traveling direction of the traveling path of the own vehicle X, and the wider the area where it can be determined that the target is reliably present on the traveling path of the own vehicle. Based.
[0061]
The area Β is set as an area in which the position of the target determined on the map can belong to the area で in a short time due to the movement of the target such as a course change or deceleration. In the map of FIG. 5, the area Β becomes wider as the distance from the host vehicle X increases. This is because, when the target is some distance away from the host vehicle X, the influence of the error on the curvature of the actual curve due to the above-described estimation becomes large, and the position of the target determined on the map becomes a region IV in a short time. This is because they may belong.
[0062]
It should be noted that each area described here is an example, and various changes can be made. The collision object probability Pro may be calculated using a captured image of a CCD camera that may recognize a target ahead in cooperation with the millimeter wave radar sensor 14. In such a case, the lane mark may be recognized from the captured image of the CCD camera, and the collision object probability Pro as described above may be calculated based on the position of the target relative to the lane on which the host vehicle is traveling.
[0063]
Next, as a second matter, a decrease in detection performance of the millimeter wave radar sensor 14 is determined based on a reflected wave received by the millimeter wave radar sensor 14. For example, the vehicle control ECU 10 may estimate the degree of decrease in the detection performance of the millimeter wave radar sensor 14 by monitoring the height (reflection intensity) of the spectrum peak of each target as shown in FIG. In this embodiment, the reflected wave is a radio wave. However, when a different type of sensor is used, it may be a light wave or an ultrasonic wave.
[0064]
Next, as a third matter, the recognition state of the target by the millimeter wave radar sensor 14 is whether the target information is estimated by the millimeter wave radar sensor 14, that is, the millimeter wave radar sensor 14 recognizes in the previous cycle. It is judged by whether or not the target has been recognized in the current cycle. This determination may involve consideration of how long the target is continuously recognized.
[0065]
In addition, this determination is made in consideration of an error between the estimated target information and the actual recognition target information when the millimeter wave radar sensor 14 recognizes the target recognized in the previous period. May be accompanied. For example, when the estimated target information (for example, the position of the target) matches the actual recognition target information, it is determined that the target recognition state by the millimeter wave radar sensor 14 is stable. Can do. On the contrary, if they are different, it can be determined that the recognition state of the target by the millimeter wave radar sensor 14 is not stable.
[0066]
Next, examples of the collision prediction control according to the present invention will be described.
[0067]
FIG. 6 is a flowchart shown as the first embodiment of the present invention. The process of this flowchart may be executed by the vehicle control ECU 10 or may be executed by another ECU connected to the vehicle control ECU 10. In the present embodiment, the execution of the collision prediction control is restricted in consideration of the above-described second and third matters.
[0068]
When this process is started by turning on the ignition switch IG of the vehicle, etc., in step 100, the second and third items described above, that is, the recognition state of the target by the millimeter wave radar sensor 14 and the millimeter wave radar sensor 14 A decrease in detection performance is considered. In this embodiment, in order to quantitatively consider these matters, an index that is a recognition index I is calculated by the following equation. The recognition index I shown in the following equation comprehensively considers the recognition state of the millimeter wave radar sensor 14.
[0069]
I = a × STAB1 + b × STAB2 + c × POWER... (Formula 1)
Here, STAB1 is an index that quantitatively indicates a recognition state related to the continuity of target recognition by the millimeter wave radar sensor 14, and may be a ratio of the number of target information by estimation in all target information. .
[0070]
STAB2 is an index that quantitatively indicates the recognition state related to the recognition variation of the target by the millimeter wave radar sensor 14,
STAB2 = 100 × {(X_est-X)²+ (Y_est-Y)²}^1/2/ (X²+ Y²)^1/2
May be defined as follows. (X, Y) is the position of the target of the current period N, and (X_est, Y_est) Is the estimated position of the target of the current cycle estimated based on the target information of the previous cycle N-1.
[0071]
POWER is an index related to the reflection intensity of the reflected wave from the target,
POWER = 100 × (P (N) −P (N−1)) / P (N−1)
May be defined as follows. P (N-1) is the reflection intensity of the reflected wave received at period N-1, and P (N) is the reflection intensity of the reflected wave received at period N.
[0072]
Note that there are various methods for calculating these indices, and the present invention is not particularly limited to the above-described calculation method. In addition, the indices STAB2 and POWER are
STAB2 (N) = ψ × STAB2 (N) + (1−ψ) × STAB2 (N−1)
POWER (N) = ψ × POWER (N) + (1−ψ) × POWER (N−1)
It may be filtered as follows. Further, as shown in Equation 1, the recognition index I may be calculated by weighting each index STAB1, STAB2, and POWER by appropriate coefficients a, b, and c.
[0073]
In the subsequent step 102, it is determined whether or not the recognition index I is greater than a predetermined first threshold value THRE1. If the determination is affirmative, the routine proceeds to step 104, where it is determined that the recognition state of the millimeter wave radar sensor 14 is bad, and the operation of the level 1 and level 2 vehicle devices described above is prohibited. On the other hand, if the determination in step 102 is negative, the process proceeds to step 106.
[0074]
In step 106, it is determined whether or not the recognition index I is greater than a predetermined second threshold value THRE2. If the determination is affirmative, the routine proceeds to step 108, where it is determined that the recognition state of the millimeter wave radar sensor 14 is slightly poor, and the operation of the level 2 vehicle device is prohibited. Accordingly, only level 1 vehicle devices such as alarm devices and hazard lamps are activated when the predetermined condition for collision prediction determination as described above is satisfied. On the other hand, if the determination in step 106 is negative, the process proceeds to step 110.
[0075]
In step 110, it is determined that the recognition state of the millimeter wave radar sensor 14 is good, and the operation of the level 1 and level 2 vehicle devices is permitted. The above steps 100 to 110 are repeatedly executed until it is confirmed in step 112 that, for example, the ignition switch IG is turned off.
[0076]
In this embodiment, the threshold values are determined so that THRE2 <THRE1, and the operation of the level 1 and level 2 vehicle devices is regulated in stages according to the recognition state of the millimeter wave radar sensor 14. Yes. However, when the recognition index I exceeds a predetermined value, the operation of the level 1 and level 2 vehicle devices can be uniformly regulated. Further, it is possible to further classify the level 2 vehicle devices to realize more multi-stage regulation. It is also possible to restrict the execution of the irreversible occupant protection device in place of restricting the execution of the collision prediction control, or instead of restricting the execution of the collision prediction control. However, the restriction on the execution of the control of the irreversible occupant protection device does not completely prohibit the execution of the control but changes the activation determination threshold.
[0077]
Further, in this embodiment, the recognition index I is defined so as to comprehensively consider the three factors of target recognition continuity, target recognition variation, and reflection intensity from the target. However, the recognition index I may be defined using any one or any two of these three elements.
[0078]
According to the first embodiment of the present invention described above, the reliability of the collision prediction control can be further improved. In addition, in consideration of the recognition state of the target by the millimeter wave radar sensor 14, the collision prediction control and the control of the irreversible occupant protection device can be regulated in stages.
[0079]
In the second embodiment to be described next, the first embodiment considers only the recognition state of the target by the millimeter wave radar sensor 14 (that is, the second and third matters). In consideration of the first item described above, that is, the position of the target with reference to the traveling path of the host vehicle, the execution of the collision prediction control and the control of the irreversible occupant protection device is restricted.
[0080]
FIG. 7 is a flowchart of processing that may be executed by the vehicle control ECU 10 as a second embodiment of the present invention.
[0081]
When this processing is started by turning on the ignition switch IG of the vehicle, in step 200, the vehicle control ECU 10 responds to the input of the target information from the millimeter wave radar sensor 14 and displays a map as shown in FIG. The collision object probability Pro is calculated for each target.
[0082]
In the following step 202, it is determined whether or not the collision object probability Pro is larger than a predetermined value (for example, 90%). If the determination is affirmative, the routine proceeds to step 204 where the activation determination threshold value of the level 3 vehicle device is set to a low level (for example, TH3 in FIG. 4), and the level 1 and level 2 vehicle devices described above are set. Allow operation. On the other hand, if the determination in step 202 is negative, the process proceeds to step 206.
[0083]
In step 206, it is determined whether or not the collision object probability Pro is larger than a predetermined value (for example, 70%). If the determination is affirmative, the routine proceeds to step 208, where the activation determination threshold value of the level 3 vehicle device is set to a medium level (for example, TH2 in FIG. 4) and the operation of the level 2 vehicle device is prohibited. On the other hand, if the determination in step 206 is negative, the process proceeds to step 210.
[0084]
In step 210, the activation determination threshold value of the level 3 vehicle device is set to a high level (for example, TH1 in FIG. 4), and the operation of the level 1 and level 2 vehicle devices is prohibited. The above steps 200 to 210 are repeatedly executed until it is confirmed in step 212 that, for example, the ignition switch IG is turned off.
[0085]
In this specification, the “prohibition” of the operation of the vehicle device not only completely prohibits the operation of the vehicle device but also partially restricts the operation content of the vehicle device and It also includes prohibiting the operation of only some of the vehicle devices.
[0086]
According to the second embodiment of the present invention described above, the reliability of the collision predictive control and the control of the irreversible occupant protection device can be further improved. In addition, it is possible to regulate the collision prediction control and the control of the irreversible occupant protection device in a stepwise manner according to the position of the target with reference to the traveling path of the host vehicle.
[0087]
In the second embodiment, the start determination threshold value of the level 3 vehicle device is changed stepwise, but the output level of the level 3 vehicle device can be changed stepwise. . For example, when an airbag apparatus that is a level 3 vehicle device includes two inflators that supply gas to the airbag, the stepwise change of the output level may be realized by changing the number of inflators that are ignited. .
[0088]
By the way, as described above, the millimeter wave radar sensor 14 is attached to the front portion of the vehicle such as a front grill of the vehicle, and scans the front of the vehicle one-dimensionally or two-dimensionally by radio waves. Therefore, as shown in FIG. 8, when the vehicle is traveling on an uphill road or when the vehicle is accelerating, the front portion of the vehicle faces upward. For example, a signboard, a sign, and a crossover near the top of the uphill road are recognized as targets. Similarly, when the vehicle is traveling on a downhill road or when traveling at a reduced speed, the front of the vehicle is directed downward, so that the millimeter wave radar sensor 14 is an object for which no collision can be considered. An object, for example, a road surface near the end point of a downhill road may be recognized as a target.
[0089]
In the third embodiment of the present invention to be described next, it is possible to exclude these objects that cannot become collision objects from the objects of determination in the above-described collision prediction control and control of the irreversible occupant protection device. .
[0090]
FIG. 9 shows whether or not a target recognized by the millimeter wave radar sensor 14 as a third embodiment of the present invention is an object to be excluded from determination targets (ie, a signboard, a sign, a crossover, etc.). It is a flowchart of the process to judge.
[0091]
In step 300, an acceleration determination value, a deceleration determination value, an uphill determination value, and a downhill determination value are calculated based on information from an acceleration sensor, a throttle sensor, and the like, a difference between actual torque and estimated torque, and the like.
[0092]
In the following step 302, it is determined whether the vehicle is accelerating, decelerating, traveling on an uphill road, or traveling on a downhill road. This determination is based on whether the acceleration determination value exceeds a predetermined value, whether the deceleration determination value exceeds a predetermined value, whether the uphill determination value exceeds a predetermined value, or whether the downhill determination value is a predetermined value. This is realized by determining whether or not the value is exceeded. If any of these determinations is affirmed, the process proceeds to step 304, and if all the determinations are negative, the process proceeds to step 306.
[0093]
In step 304, based on the target information from the millimeter wave radar sensor 14, the target is far away (for example, Lr> 100 m) and is substantially stopped (that is, Vr≈Vn (own vehicle speed)). And the target is excluded from the judgment target. For example, in the first embodiment, the exclusion is performed by setting the recognition index I for the target to a high level (for example, I> THRE1), and in the second embodiment, the collision object for the target. This may be realized by setting the probability Pro to a low level (for example, the collision object probability Pro <50%). In addition, the reason why the target located far away is excluded is that when the front of the vehicle is facing upward, the farther the recognized target is, the greater the height of the target relative to the road surface. is there. Also, the stopped targets are excluded because signs, signs, crossovers, and the like that are misidentified by the millimeter wave radar sensor 14 are stopped objects.
[0094]
The above steps 300 to 304 are repeatedly executed until it is confirmed in step 306 that the ignition switch IG is turned off, for example.
[0095]
According to the third embodiment of the present invention described above, since a target that cannot be a collision object is surely excluded, the reliability of the collision prediction control and the control of the irreversible occupant protection device can be further improved. .
[0096]
In the third embodiment described above, all of whether the vehicle is accelerating, decelerating, traveling uphill, and downhill. However, it is also possible not to perform any of these determinations.
[0097]
FIGS. 10 and 11 are flowcharts of the control program showing the fourth embodiment of the present invention in detail. This control program is executed every detection period (for example, every 0.1 sec) of the millimeter wave radar sensor 14. The control program may be executed by the vehicle control ECU 10 as described below, but may be executed by another ECU connected to the vehicle control ECU 10. This control program is stored in advance in the ROM of the vehicle control ECU 10.
[0098]
When execution of the control program is started by turning on the ignition switch IG of the vehicle or the like, target information about one or more targets recognized by the millimeter wave radar sensor 14 is obtained at step 400 in the vehicle control ECU 10. Is input.
[0099]
In the subsequent step 402, the vehicle control ECU 10 estimates the curvature of the traveling path based on information such as the steering angle value from the steering angle sensor, and uses the map as shown in FIG. 5 to estimate the target information and the target information. The collision object probability Pro for each target is calculated from the curvature. The collision object probability Pro may be filtered for each period, and the collision object probability Pro (N) of the period N is calculated by using the collision object probability Pro (N-1) of the previous period N-1 to Pro ( N) = ψ × Pro (N) + (1−ψ) × Pro (N−1).
[0100]
In subsequent step 404, in order to consider the continuity of target recognition by the millimeter wave radar sensor 14, a recognition continuity counter Cont is calculated for each target. The recognition continuity counter Cont is incremented by a predetermined value UP when the target information in the current cycle is not based on estimation, and is decremented by a predetermined value DN when the target information is based on estimation. .
[0101]
In the subsequent step 406, an index POWER indicating the intensity of the reflected wave from each target is taken into consideration in consideration of a decrease in detection performance of the millimeter wave radar sensor 14, the physical size of the target, and erroneous recognition of the millimeter wave radar sensor 14. Is calculated. The index POWER in this embodiment is a predetermined threshold value P._threshold, POWER = P_target-P_thresholdIt is expressed as Where P_targetIs the intensity of the reflected wave from the target, and the frequency f at which the reflection from each target appears._targetThis corresponds to the height of the spectrum peak according to (see FIG. 2). This exponent POWER may be filtered per period, and POWER (N) of period N is expressed as POWER (N) = ψ × POWER (N) + (1−ψ) × POWER (N−1). . However, when the target information in the current cycle is based on estimation, the value in the previous cycle may be held as POWER (N) = POWER (N−1).
[0102]
In the subsequent step 408, in order to consider the stability (degree of variation) of target recognition by the millimeter wave radar sensor 14, an index STAB2 relating to the stability of recognition is calculated for each target. The index STAB2 in this embodiment is the position of the target (X_est, Y_est) And the position (X, Y) of the target in the current cycle, STAB2 = (X_est-X)²+ (Y_est-Y)²It is expressed as This index STAB2 may be filtered for each period, and STAB2 (N) of period N is expressed as STAB2 (N) = ψ × STAB2 (N) + (1-ψ) × STAB2 (N−1). .
[0103]
In subsequent step 410, the collision object probability Pro obtained in step 402 is corrected using the counter value and each index obtained in steps 404 to 408, and a new collision object probability P is obtained._CRSHEstablish. This new collision object probability P_CRSHMay be expressed as the following equation using the above-mentioned collision object probability Pro or the like.
[0104]
P_CRSH= A * Pro + b * Cont + c * POWER-d * STAB2 where each coefficient a, b, c, d is the collision object probability P_CRSHIs set to 100% at the maximum.
[0105]
In the following step 412, in order to detect the vehicle state, an uphill determination value, an acceleration determination value, a downhill determination value, a deceleration determination value based on information from an acceleration sensor, a throttle sensor, or the like, a difference between an actual torque and an estimated torque, or the like. Are respectively calculated.
[0106]
Next, in step 414, it is determined whether the vehicle is traveling uphill or accelerating, and whether the vehicle is traveling downhill or decelerating. If any determination is satisfied, the process proceeds to step 416, and if any determination is not satisfied, the process proceeds to step 418.
[0107]
In step 416, in order to surely exclude a target that cannot be a colliding object (for example, a signboard near the top of the uphill road), based on the target information from the millimeter wave radar sensor 14, it is far away (for example, Lr> 100 m). ) Is specified, and the collision object probability P related to the target is determined._CRSHLow level (for example, P_CRSH= 49%).
[0108]
In this way, the collision object probability P_CRSHIs established, the program shifts to a control program for selectively restricting the collision prediction control (steps 418 to 422).
[0109]
In step 418, the collision object probability P_CRSHIs greater than a predetermined value (for example, 50%). If the determination is negative, the routine proceeds to step 420, where the activation determination threshold value of the level 3 vehicle device is set to a high level (for example, TH1 in FIG. 4), and the operation of the level 1 and level 2 vehicle devices is performed. Ban. On the other hand, if the determination in step 418 is affirmative, the process proceeds to step 422.
[0110]
In step 422, the collision object probability P_CRSHIs greater than a predetermined value (for example, 70%). If the determination is negative, the process proceeds to step 424, where the activation determination threshold value of the level 3 vehicle device is set to a high level (for example, TH1 in FIG. 4) and the operation of the level 1 vehicle device is permitted. On the other hand, if the determination in step 422 is affirmed, the process proceeds to step 426.
[0111]
In step 426, the collision object probability P_CRSHIs greater than a predetermined value (for example, 90%). If the determination is negative, the process proceeds to step 428, where the activation determination threshold value of the level 3 vehicle device is set to a high level (for example, TH1 in FIG. 4), and the operation of the level 1 and level 2 vehicle devices is performed. to approve. On the other hand, if the determination in step 426 is affirmative, the process proceeds to step 430.
[0112]
In step 430, the activation determination threshold of the level 3 vehicle device is changed to a low level (for example, TH3 in FIG. 4), and the operation of the level 1 and level 2 vehicle devices is permitted.
[0113]
After the above steps 420, 424, 428, and 430, it is determined in step 432 whether or not the ignition switch IG is turned off. If the determination is negative, the processing after step 400 is repeated. On the other hand, if the determination is affirmative, the process ends.
[0114]
The fourth embodiment of the present invention described above can achieve all the effects of the first, second, and third embodiments described above. That is, according to the fourth embodiment of the present invention, the reliability of the collision predictive control and the control of the irreversible occupant protection device can be improved. Also, the collision prediction control and the control of the irreversible occupant protection device are regulated in stages according to the recognition state of the target by the millimeter wave radar sensor 14 and the position of the target based on the traveling path of the host vehicle. Is possible.
[0115]
The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above-described embodiments, and various modifications and substitutions can be made to the above-described embodiments without departing from the scope of the present invention. Can be added.
[0116]
In the embodiment described above, the vehicle control ECU 10 or the like may be specified as the main body that executes the various processes described above, but all or some of the processes of the vehicle control ECU 10 are performed by other various ECUs. Alternatively, the vehicle control ECU 10 may perform some or all of the functions of the other various ECUs.
[0117]
  The “detecting means” described in the claims corresponds to the “millimeter wave radar sensor 14” described in the above embodiment. The “collision prediction means” described in the claims is realized by “collision prediction determination by the vehicle control ECU 10” described in the above embodiment.The
[0118]
【The invention's effect】
  Since the present invention is as described above, the following effects can be obtained. According to the first aspect of the present invention, it is possible to realize collision prediction control with improved accuracy of collision prediction determination and improved reliability.In addition, the operation of the vehicle device can be regulated in stages according to the recognition state of the detection means such as a millimeter wave radar sensor.
[0121]
  Claims4According to the described invention, a vehicle device such as a seat belt device can be accurately operated,5According to the described invention, the collision prediction determination can be accurately performed.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a control system according to the present invention.
FIG. 2 is a diagram showing a power spectrum of a reflected wave.
FIG. 3 is a diagram showing a determination map used when a collision prediction determination is performed.
FIG. 4 is a diagram showing a determination map used when determining the activation of the irreversible occupant protection device.
FIG. 5 is a diagram showing a map used when calculating the collision object probability.
FIG. 6 is a flowchart showing a first embodiment of the present invention.
FIG. 7 is a flowchart showing a second embodiment of the present invention.
8 is a diagram schematically showing a relationship between a vehicle state and a recognition range of the millimeter wave radar sensor 14. FIG.
FIG. 9 is a flowchart showing a third embodiment of the present invention.
FIG. 10 is a flowchart (No. 1) of a control program according to a fourth embodiment of the present invention.
FIG. 11 is a flowchart (No. 2) of the control program according to the fourth embodiment of the present invention.
[Explanation of symbols]
10 Vehicle control ECU
14 Millimeter wave radar sensor
16 Various ECUs
18 Bus
80 Vehicle control device

Claims (5)

Detecting means for recognizing an object ahead of the host vehicle and detecting a relationship between the target object and the host vehicle;
A collision prediction means for predicting a collision between the host vehicle and the object based on the relationship detected by the detection means;
A plurality of vehicle devices operated based on the prediction result by the collision prediction means;
The recognition index representing a good or bad recognition state of the object by the detection means, the target of the detection means is an index indicating whether the object is recognized by how continuous in time series by the detection means Continuity of object recognition, an index representing how stably the object is recognized by the detection means, variation in recognition of the object by the detection means, and radio waves emitted by the detection means or A recognition index calculating means for calculating based on at least one of reflection intensity of the sound wave or light wave from the object;
A vehicle control apparatus, wherein a vehicle device that is subject to operation restriction is selected from a plurality of vehicle devices according to a recognition index by the recognition index calculation means. The plurality of vehicle devices include a first vehicle device that alerts the occupant or the outside of the vehicle, and a reversible second vehicle device that protects the occupant,
When the recognition index by the recognition index calculation means exceeds the first threshold value and the detection means determines that the recognition state of the object is bad, the operation of the first and second vehicle devices is restricted, and If the recognition index calculated by the recognition index calculation means is smaller than the first threshold value but exceeds the second threshold value, and the detection state of the object by the detection means is determined to be slightly poor , the operation of the first device is permitted. and characterized by regulating the operation of the second vehicle device, the vehicle control device according to claim 1. Based on the position of the object based on the traveling path of the own vehicle grasped using the relationship detected by the detecting means, and the recognition index by the recognition index calculating means , the object is relative to the own vehicle. A collision object probability calculating means for calculating a probability of becoming an collision object;
The plurality of vehicle devices include: a first vehicle device that alerts the occupant or the outside of the vehicle; a reversible second vehicle device that protects the occupant; and an irreversible third vehicle device that protects the occupant. Including
When the collision object probability calculated by the collision object probability calculation means is low, the activation determination threshold of the third vehicle device is set to a high level at which activation determination is difficult to be performed, and the first and second When the operation of the vehicle device is prohibited and the collision object probability calculated by the collision object probability calculation means is a medium level, the activation determination threshold value of the third vehicle device is set to a high level that is difficult to determine the activation. In addition, when the operation of the second vehicle device is prohibited, the operation of the first vehicle device is permitted, and the collision object probability calculated by the collision object probability calculation means is high, the third vehicle device The vehicle control apparatus according to claim 1 or 2, wherein the vehicle device activation determination threshold is set to a low level and the operation of the first and second vehicle devices is permitted. The second vehicle device, the brake device comprises at least one of the seat belt device, the third vehicle device includes an air bag device, the vehicle control apparatus according to claim 2 or 3 wherein. The relationship between the object and the vehicle, the relative distance, the relative speed of the object and the vehicle as a reference, and including at least a heading, the vehicle control apparatus of any one of claims 1 to 4.

Images