JP2009157668A - External recognition device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that the function of a system is extremely deteriorated when any abnormality is generated in own vehicle information in determining collision. <P>SOLUTION: An external recognition device includes: an own vehicle information acquisition means for acquiring the information of an own vehicle; an object information acquisition means for acquiring the information of an object detected by an external recognition sensor; a predicted route setting means for setting the predicted route of the own vehicle on the basis of the own vehicle information acquired by the own vehicle information acquisition means; a first collision determination means for determining the possibility of the collision of the object detected by the external recognition sensor with the own vehicle on the basis of the object information acquired by the object information acquisition means and the predicted route set by the predicted route setting means; a second collision determination means for determining the possibility of collision of the object detected by the external recognition sensor with the own vehicle on the basis of the object information acquired by the object information acquisition means; an own vehicle information determination means for determining the abnormality of the own vehicle information acquired by the own vehicle information acquisition means; and a collision determination selection means for selecting the second collision determination means when it is determined that the own vehicle information acquired by the own vehicle information acquisition means is abnormal by the own vehicle information determination means. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an external recognition apparatus that determines the possibility of collision between an object detected by an external recognition sensor and a host vehicle.

  The number of fatalities due to traffic accidents has been declining due to the introduction of passenger protection after collision (airbags, collision safety bodies) called collision safety, but the number of accidents has been increasing linked to the number of cars owned. In order to reduce the number of accidents, it is important to develop a preventive safety system that prevents accidents. A preventive safety system is a system that operates before an accident.For example, when there is a possibility of collision with an object in front of the vehicle, the driver is alerted by an alarm, and when the collision becomes unavoidable Pre-crash safety systems that reduce the damage to passengers by automatic braking have been put into practical use.

  Here, there is an external recognition device that determines the possibility of collision between an object ahead of the host vehicle detected by a radar or a camera and the host vehicle (see Patent Document 1). Patent Document 1 predicts the course of the host vehicle as a turning radius using the vehicle speed, the steering angle, and the yaw rate, and can collide with the host vehicle according to the predicted path and the relative position of the object detected by the external recognition sensor. And a second collision determination method for calculating a relative velocity vector from the amount of change in the relative position of the object and determining a collision possibility with the own vehicle according to the relative velocity vector. Is described. According to this method, the first collision determination result is emphasized for an object having a small vertical relative velocity, and the second collision determination result is performed for an object having a large vertical relative velocity. It is possible to accurately determine the possibility of collision in a scene in which approaching with a low-speed vehicle traveling ahead occurs after avoiding steering of a stopped vehicle in front of the host vehicle.

JP 2005-25458 A

  However, in the first collision determination method, sensor information such as the vehicle speed, the steering angle, and the yaw rate is used. Therefore, when an abnormality occurs in such information, for example, the vehicle speed sensor, the steering angle sensor, and the yaw rate sensor The accuracy of the first collision determination may be significantly reduced at the time of failure or when communication failure occurs when such information is received from another control device. As a result, the entire system may be reduced. Must be invalidated.

  Therefore, an object of the present invention is to generate an abnormality in the information of the own vehicle used when determining the possibility of collision with the own vehicle according to the predicted course of the own vehicle and the relative position of the object detected by the external recognition sensor. In such a case, it is an object of the present invention to provide an external recognition apparatus that minimizes the performance degradation of the entire system.

  In order to solve the above problems, one of the desirable embodiments of the present invention is as follows.

  The external environment recognition device includes an own vehicle information acquisition unit that acquires information on the own vehicle, an object information acquisition unit that acquires information on an object detected by the external environment recognition sensor, and an own vehicle information acquired by the own vehicle information acquisition unit. Based on the predicted course setting means for setting the predicted course of the own vehicle, the object information acquired by the object information acquisition means, and the predicted course set by the predicted course setting means, and the object detected by the external recognition sensor and the own vehicle A first collision determination unit that determines the possibility of collision with the vehicle, and a second collision determination unit that determines a collision possibility between the object detected by the external recognition sensor and the host vehicle based on the object information acquired by the object information acquisition unit. Collision determination means, own vehicle information determination means for determining abnormality of own vehicle information acquired by own vehicle information acquisition means, and own vehicle information determination means determine that own vehicle information acquired by own vehicle information acquisition means is abnormal The second collision determination means And a collision determination selecting means for-option.

  According to the present invention, even when an abnormality occurs in the information of the own vehicle used when determining the possibility of collision with the own vehicle according to the predicted course of the own vehicle and the relative position of the object detected by the external recognition sensor. Thus, it is possible to provide an external recognition apparatus that minimizes the performance degradation of the entire system.

  Hereinafter, embodiments will be described in detail with reference to FIGS.

  FIG. 1 is a block diagram of the external environment recognition apparatus 100.

  The contents of the control shown below are programmed in the external recognition apparatus 100 and are repeatedly executed at a predetermined cycle.

  The own vehicle information acquisition unit 1 acquires own vehicle information such as the vehicle speed Vsp, the steering angle α, and the yaw rate γ in accordance with detection signals from the vehicle speed sensor, the steering angle sensor, and the yaw rate sensor. These vehicle information may be acquired by directly inputting the signals of the sensors to the external recognition device 100, and when the sensor signals are input to other control devices, You may acquire by performing communication using LAN (Local Area Network).

  The object information acquisition unit 2 acquires a relative distance PY [i], a lateral position PX [i], and a width WD [i] of an object around the vehicle according to a detection signal of an external recognition sensor such as a radar or a camera. Here, i is an object ID number when a plurality of objects are detected. The object information may be acquired by directly inputting a signal from the external recognition sensor to the external recognition device 100, or may be acquired by performing communication with the external recognition sensor using a LAN (Local Area Network). You may do it.

  The predicted course setting means 3 calculates the predicted course of the host vehicle according to the host vehicle information (vehicle speed Vsp, steering angle α, yaw rate γ) acquired by the host vehicle information acquiring unit 1. Here, the turning radius R is calculated as the predicted course of the host vehicle. A method for calculating the turning radius R will be described later.

  The first collision determination means 4 includes the predicted course of the host vehicle calculated by the predicted course setting means 3, the object information acquired by the object information acquisition means 2 (relative distance PY [i], lateral position PX [i], width The risk level 1DREC1 [i] is calculated according to WD [i]). Here, i is an object ID number when a plurality of objects are detected. A method of calculating the risk level 1DREC1 [i] will be described later.

  The second collision determination unit 5 sets the risk 2DREC2 [i] according to the object information (relative distance PY [i], lateral position PX [i], width WD [i]) acquired by the object information acquisition unit 2. Calculate. Here, i is an object ID number when a plurality of objects are detected. A method for calculating the risk level 2DREC2 [i] will be described later.

  The own vehicle information determination means 6 determines whether each information is normal or not according to the own vehicle information (vehicle speed Vsp, steering angle α, yaw rate γ) acquired by the own vehicle information acquisition means 1, and a diagnosis NG determination flag fDGNVCAN. Is calculated. A method for calculating the diagnostic NG determination flag fDGNVCAN will be described later.

  The collision determination selection means 7 places importance on the risk 1DREC1 [i], which is the result of the first collision determination means 4, according to the diagnosis result of the vehicle information determination means 6, or the second collision determination means 5 It is determined whether or not the risk level 2DREC2 [i] as a result is important, and the integrated risk level DRECI [i] is calculated. Note that these determination methods and the calculation method of the integrated risk DRECI [i] will be described later.

  Next, referring to FIG. 2, the above-described pre-crash safety system is taken as an example, and a system operation method of outputting an alarm according to the integrated risk DRECI [i] or automatically controlling the brake Will be described. FIG. 2 is a flowchart showing a method of operating the pre-crash safety system.

  First, in step 201, object information is read. Next, in step 201, the collision prediction time TTC [i] of each object detected by the external recognition sensor is calculated using equation (1). Here, the relative velocity VY [i] is obtained by pseudo-differentiating the relative distance PY [i] of the object.

TTC [i] = PY [i] ÷ VY [i] (1)
Further, in step 203, the integrated risk degree DRECI [i] calculated by the external recognition apparatus 100 is read. In addition, the process of steps 201-203 is set as the structure which performs a loop process according to the detected number of objects.

  In step 204, an object satisfying the condition of the expression (2) is selected according to the integrated risk DRECI [i] read in step 203, and the predicted collision time TTC [i] is selected among the selected objects. ] Is the smallest object k.

DRECI [i] ≧ cDRECI # (2)
Here, the predetermined value cDRECI # is a threshold value for determining whether or not the vehicle collides. Next, in step 205, it is determined whether or not the brake is automatically controlled in accordance with the predicted collision time TTC [k] of the selected object k. If Expression (3) is established, the process proceeds to step 206, brake control is executed, and the process is terminated. On the other hand, if Expression (3) is not established, the process proceeds to Step 207.

TTC [k] ≦ cTTCBRK # (3)
In step 207, it is determined whether or not the alarm is output in accordance with the predicted collision time TTC [k] of the selected object k. If the expression (4) is established, the process proceeds to step 208, an alarm is output, and the process is terminated. In addition, when Expression (4) is not established, neither the brake control nor the alarm is executed, and the process is terminated.

TTC [k] ≦ cTTCALM # (4)
As described above, the integrated risk DRECI [i] indicating the result of the collision determination is an important parameter that affects the operation of the system (automatic braking, alarm).

  Next, the predicted course setting means 3 will be described with reference to FIG. FIG. 3 is a schematic diagram showing the processing contents of the predicted course setting means 3.

  If the vehicle position is the origin O, the predicted course can be approximated by an arc of a turning radius R passing through the origin O. Here, the turning radius R is expressed by Expression (5) using the steering angle α, the speed Vsp, the stability factor A, the wheel base L, and the steering gear ratio Gs of the host vehicle.

R = (1 + AV ² ) × (L · Gs / α) (5)
The stability factor is an important value that determines the magnitude of the change depending on the speed of the steady circular turning of the vehicle. As can be seen from the equation (5), the turning radius R changes in proportion to the square of the speed Vsp of the host vehicle with the stability factor A as a coefficient. Further, the turning radius R can be expressed by Expression (6) using the vehicle speed Vsp and the yaw rate γ.

R = V / γ (6)
As described above, the predicted course of the host vehicle can be approximated by an arc of the turning radius R by using the host vehicle information on the vehicle speed, the steering angle, and the yaw rate.

  Next, the processing content of the first collision determination means 4 will be described with reference to FIG.

  When the relative distance of an object detected by an external recognition sensor such as a radar or a camera is PY [i] and the lateral position is PX [i], the distance r from the center of the arc drawn by the turning radius R to the object is expressed by the following equation (7). It is represented by

(PX [i] -R) ² + PY [i] ² = r (7)
Further, the difference d between the turning radius R and the distance r can be obtained by Expression (8).

d = | R−r | (8)
From FIG. 3, it can be determined that the degree of risk is high because the detected object approaches the predicted course of the vehicle as the radius difference d in equation (8) decreases. For example, an object whose difference d is about half or less of the vehicle width is calculated so that the risk 1DREC1 [i] is equal to or greater than a predetermined value cDRECI # so that the pre-crash safety system described in FIG. 2 operates. Set to.

  Next, the processing content of the second collision determination means 5 will be described with reference to FIG. FIG. 4 is a schematic diagram showing the processing contents of the second collision determination means 5.

  When the relative distance of the object m detected by an external recognition sensor such as a radar or a camera is PY [m] and the lateral position is PX [m], the relative velocity vector (VX [m] ], VY [m]) are directed in the direction of the vehicle and are expected to intersect the X axis within the vehicle width. Therefore, it can be determined that the detected object m has a high possibility of collision. When the relative distance of the sensing object n is PY [n] and the lateral position is PX [n], the relative velocity vectors (VX [n], VY [n]) of the sensing object n obtained by these pseudo-differentiations are obtained. It is expected that the vehicle does not intersect the X axis within the vehicle width and does not intersect the side surface of the vehicle. Therefore, it can be determined that the detected object n has a low possibility of collision.

  As described above, the possibility of collision can be determined by obtaining the relative velocity vector from the object information detected by the external recognition sensor and obtaining the intersection between the relative velocity vector and the own vehicle. For example, an object whose relative velocity vector is between points a and b in the figure is calculated so that the risk 2DREC2 [i] is equal to or greater than a predetermined value cDRECI #, and the pre-crash Set the safety system to operate.

  Next, the processing content of the own vehicle information determination means 6 is demonstrated using FIG. FIG. 5 is a flowchart showing the processing contents of the vehicle information determination means 6.

  Here, the value of each bit #xDGNVSP, #xDGNSTR, #xDGNYAW of the diagnosis NG determination flag fDGNVCAN represents normality / abnormality of own vehicle information corresponding to the vehicle speed, steering angle, and yaw rate, and is diagnosed as abnormal The corresponding bit is set in.

  First, in step 501, the own vehicle information (vehicle speed Vsp, steering angle α, yaw rate γ) is read. In step 502, the previous value (value in the previous calculation cycle) of these pieces of information is used as the vehicle speed previous value Vspz, steering. Stored as the previous angle value αz and the previous yaw rate value γz. Next, in step 503, a difference between the vehicle speed previous value Vspz and the vehicle speed Vsp is calculated, and it is determined whether or not a state in which the difference is larger than a predetermined value (threshold value that can be determined to be abnormal) has elapsed for a predetermined time. If it is determined in step 503 that this condition is not satisfied and the vehicle speed Vsp is not abnormal, the process proceeds to step 505. In step 503, if the difference is greater than the predetermined value and the predetermined time has elapsed and it is diagnosed that the vehicle speed Vsp is abnormal, the process proceeds to step 504, and the bit #xDGNVSP of the diagnosis NG determination flag fDGNVCAN is set. Set and go to step 505.

  In step 505, a difference between the previous steering angle value αz and the steering angle α is calculated, and it is determined whether or not a state in which the difference is larger than a predetermined value (threshold value that can be determined to be abnormal) has elapsed for a predetermined time. If it is determined in step 505 that this condition is not satisfied and the steering angle α is not abnormal, the process proceeds to step 507. If it is determined in step 505 that the difference is larger than the predetermined value and the predetermined time has elapsed and the steering angle α is diagnosed to be abnormal, the process proceeds to step 506 and bit #xDGNSTR of the diagnostic NG determination flag fDGNVCAN. And proceed to step 507.

  In step 507, the difference between the previous yaw rate value γz and the yaw rate γ is calculated, and it is determined whether or not a predetermined time has elapsed when the difference is larger than a predetermined value (threshold value that can be determined to be abnormal). If it is determined in step 507 that this condition is not satisfied and the yaw rate γ is not abnormal, the process ends. If it is determined in step 507 that the difference is greater than the predetermined value and the predetermined time has elapsed and the yaw rate γ is diagnosed to be abnormal, the process proceeds to step 508 and the bit #xDGNYAW of the diagnostic NG determination flag fDGNVCAN is set. Set and finish the process.

  As described above, it is possible to determine whether or not the vehicle speed, the steering angle, and the yaw rate, which are the vehicle information, are abnormal by the vehicle information determination unit 6.

  Next, processing contents of the collision determination selection means 7 will be described with reference to FIG. FIG. 6 is a flowchart showing the processing contents of the collision determination selection means 7.

  Note that the processing in steps 602 to 607 is configured to perform loop processing according to the number of detected objects.

  First, in step 601, a diagnostic NG determination flag fDGNVCAN is read, and in step 602, object information is read. Next, in step 603, the relative speed VY [i] is calculated by pseudo-differentiating the relative distance PY [i] of the object, and the process proceeds to step 604. In step 604, the risk level 2 weighting coefficient KW2 [i] is calculated according to the equation (9) according to the relative speed VY [i] calculated in step 603.

KW2 [i] = function tKW2 (VY [i]) (9)
Next, in Step 605, it is determined whether or not the first collision determination using the predicted course of the host vehicle set according to the host vehicle information such as the vehicle speed Vsp, the steering angle α, and the yaw rate γ can be applied. . Specifically, referring to each bit of the diagnostic NG determination flag fDGNVCAN and if all the bits are cleared (conditions C1 to C3 described in step 605 in the figure), the process proceeds to step 606, which is dangerous. The integrated risk degree DRECI [i] is calculated according to the equation (10) using the degree 2 weight coefficient KW2 [i].

DRECI [i] = (1-KW2 [i]) × DREC1 [i]
+ KW2 [i] × DREC2 [i] (10)
If at least one bit of the diagnostic NG determination flag fDGNVCAN is set in step 605, the process proceeds to step 607, and the risk level 2DREC2 [i] is assigned to the integrated risk level DRECI [i].

  As described above, when the vehicle information such as the vehicle speed, the steering angle, and the yaw rate is normal by the collision determination selection unit 7, the risk 1DREC1 [ The integrated risk DRECI [i] can be calculated by weighting the i] and the risk 2DREC2 [i] well. In addition, when the vehicle information such as the vehicle speed, the steering angle, and the yaw rate is abnormal, it is possible to minimize the degradation of the performance of the entire system by using the risk 2DREC2 [i] that does not use such information. It becomes possible. In addition, referring to each bit of the diagnostic NG determination flag fDGNVCAN, the method for setting the predicted course described in FIG. 3 may be changed. For example, when it is determined that only the steering angle is abnormal, the turning radius R is calculated according to the vehicle speed and the yaw rate using Expression (6), and when only the yaw rate is determined to be abnormal, Expression (5) ) Can be used to calculate the turning radius R according to the vehicle speed and the steering angle.

  Next, another embodiment of the external recognition apparatus will be described with reference to FIGS.

  FIG. 7 is a processing configuration diagram of a sensor fusion system that improves the detection accuracy and reliability of an object by combining information of a radar that is a first external field recognition sensor and a camera that is a second external field recognition sensor.

  The vehicle recognition device 100a receives the vehicle information such as the vehicle speed, the steering angle, and the yaw rate and the radar information such as the relative distance, lateral position, and width of the object detected by the radar. By executing the processing, the radar integration risk RDRECI [i] is calculated. Further, in the external environment recognition apparatus 100a, the relative distance, the lateral position, and the width, which are object information of the radar detection object, are changed to PYR [i], PXR [i], and WDR [i] by the first object information acquisition unit (not shown). Is getting as.

  In the image processing means 710, a high-risk object among a plurality of radar detection objects according to the radar object information (PYR [i], PXR [i], WDR [i]) and the radar integrated risk RDRECI [i]. Is selected and image processing is performed on the object selected according to the information of the image captured by the camera, thereby outputting camera information such as the relative distance, lateral position, and width of the object.

  Here, the processing content of the image processing means 710 will be described with reference to FIG. FIG. 8 is a flowchart showing the processing contents of the image processing means 710.

  First, in step 801, the radar integration risk RDRECI [i] is read. In step 802, it is determined whether the radar integration risk RDRECI [i] is equal to or greater than a predetermined value cRDRECI #. If the condition of step 802 is satisfied, that is, if it is determined that the object detected by the radar and the own vehicle are likely to collide, the process proceeds to step 803. If it is determined that the condition in step 802 is not satisfied, that is, it is determined that the object detected by the radar and the own vehicle are unlikely to collide, the process is terminated.

  Next, in step 803, the radar object information (PYR [i], PXR [i], WDR [i]) is read. In step 804, the radar object information and the camera geometric model (the position on the image and the actual position) are read. The processing area on the image is set based on After setting the image processing area, the process proceeds to step 805, and image processing such as pattern matching that scans the area is executed to detect the object. Thereafter, the process proceeds to step 806, and when an object is detected, object information such as a relative distance, a lateral position, and a width is output, and the process ends.

  Returning to FIG. 7, the external vehicle recognition device 100 b receives the vehicle information such as the vehicle speed, the steering angle, and the yaw rate and the camera information such as the relative distance, the lateral position, and the width of the object detected by the camera. The integrated risk degree DRECI [i] is calculated by executing the same processing as in FIG. In the external environment recognition apparatus 100b, the relative distance, the lateral position, and the width, which are object information of the camera detection object, are set to PY [i], PX [i], and WD [i] by the second object information acquisition unit (not shown). Is getting as.

  The control means 720 outputs an alarm according to the camera object information (PY [i], PX [i], WD [i]) and the integrated risk DRECI [i], or automatically executes brake control. The command value is calculated according to the flowchart shown in FIG.

  As described above, an object that is highly likely to collide with the vehicle by performing collision determination according to the vehicle information such as the vehicle speed, the steering angle, and the yaw rate and the information of the object detected by the radar that is the first external recognition sensor. The detection accuracy / reliability can be improved by selecting and performing object detection by image processing using the camera which is the second external recognition sensor. At this time, by applying the external recognition device 100a to the process of selecting an object that is highly likely to collide with the vehicle according to the radar information, there is an abnormality in the information on the vehicle used for the first collision determination. Even if it occurs, it is possible to minimize the performance degradation of the entire system.

  Next, the processing contents of the vehicle information determination unit when the vehicle information is acquired by a predetermined communication unit will be described with reference to FIG. FIG. 9 is a flowchart showing the processing contents of the vehicle information determination means for diagnosing whether or not the communication means has failed.

  As an example, a case where the external recognition device receives vehicle speed data will be described.

  First, in step 901, it is determined whether or not vehicle speed data has been received by referring to a reception buffer corresponding to the vehicle speed data. When vehicle speed data is received, the process proceeds to step 902, where a vehicle speed data reception flag fVCANRCV (#xVSPRCV) is set. Set and go to step 904. In step 904, the value of the reception buffer corresponding to the vehicle speed data is substituted into the vehicle speed Vsp, and the process proceeds to step 905. In step 901, when the vehicle speed data is not received, the process proceeds to step 903, the vehicle speed data reception flag fVCANRCV (#xVSPRCV) is cleared, and the process proceeds to step 905.

  Next, in step 905, it is determined whether or not a predetermined time has elapsed in which the vehicle speed data reception flag fVCANRCV (#xVSPRCV) is cleared. If the predetermined time has elapsed, the communication means has failed. The process proceeds to step 906, where the diagnosis NG determination flag fDGNVCAN (#xDGNVSP) is set and the process is terminated. In step 905, if the vehicle speed data reception flag fVCANRCV (#xVSPRCV) is not cleared for a predetermined time, the process is terminated.

  As described above, the external recognition device measures the interval at which the vehicle information such as the vehicle speed, the steering angle, and the yaw rate is received, and determines whether or not the reception interval has passed a predetermined time (≈set communication cycle). Thus, it is possible to diagnose an abnormality in the communication means.

  Next, a method of performing lane recognition by image processing and setting a predicted course according to the result of lane recognition when the abnormality of the own vehicle information is determined by the own vehicle information determination will be described using FIG. . FIG. 10 is a schematic diagram showing a lane recognition result (vehicle position in a lane) by image processing. FIG. 10 shows a relative coordinate system with the own vehicle as the origin O, and the coordinates of the road center line obtained from the lane recognition result are defined as (X, Y).

  Assuming that the road center line passing through the origin O is the predicted course (turning radius R) of the host vehicle, this equation is expressed by Expression (11) using the turning radius R.

(X−R) ² + Y ² = R ² (11)
Therefore, the turning radius R is expressed by Expression (12) using the coordinates (X, Y) of the road center line.

R = (X ² + Y ² ) / (2X) (12)
As described above, the predicted course (turning radius R) of the own vehicle can be obtained from the coordinates (X, Y) of the road center line obtained from the result of lane recognition, and the vehicle's own vehicle speed, steering angle, yaw rate, etc. When an abnormality occurs in the information, it is possible to suppress a significant decrease in the function of the system by substituting the road center line obtained from the lane recognition result.

  FIG. 11 is a diagram comparing the present invention with a conventional system.

  When the vehicle information such as the vehicle speed, the steering angle, and the yaw rate is normal, the possibility of collision between the object detected by the external recognition sensor and the vehicle can be accurately determined by the method described in Patent Document 1. Safety systems such as a pre-crash safety system can be operated without any performance degradation in the operating range. According to the present invention, even when an abnormality occurs in the own vehicle information, the second collision determination using only the object information is selected, or the road center line obtained from the lane recognition result of the camera or the like is used as the predicted course. By making the first collision determination available by setting to, the system can be operated in the whole area with a slight performance degradation.

The block diagram of an external field recognition apparatus. The flowchart which shows the operation | movement method of a pre-crash safety system. The schematic diagram which shows the processing content of a prediction course setting means and a 1st collision determination means. The schematic diagram which shows the processing content of a 2nd collision determination means. The flowchart which shows the processing content of the own vehicle information determination means. The flowchart which shows the processing content of a collision determination selection means. The processing block diagram of the sensor fusion system by the radar which is another embodiment, and a camera. The flowchart which shows the processing content of an image processing means. The flowchart which shows the processing content of the own vehicle information determination means which diagnoses whether the communication means has failed. The schematic diagram which shows the lane recognition result (the own vehicle position in a lane) by image processing. The figure which compares this invention and a conventional system.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Own vehicle information acquisition means 2 Object information acquisition means 3 Predicted course setting means 4 First collision determination means 5 Second collision determination means 6 Own vehicle information determination means 7 Collision determination selection means 100 External recognition apparatus 100a External recognition apparatus a
100b External recognition device b
710 Image processing means 720 Control means

Claims (6)

Own vehicle information acquisition means for acquiring own vehicle information;
Object information acquisition means for acquiring information of an object detected by an external recognition sensor;
Predicted course setting means for setting a predicted course of the own vehicle based on the own vehicle information acquired by the own vehicle information acquisition means;
Based on the object information acquired by the object information acquisition means and the predicted course set by the predicted course setting means, first collision determination means for determining the possibility of collision between the object detected by the external field recognition sensor and the own vehicle. When,
Based on the object information acquired by the object information acquisition means, second collision determination means for determining the possibility of collision between the object detected by the external recognition sensor and the host vehicle;
Vehicle information determination means for determining abnormality of the vehicle information acquired by the vehicle information acquisition means;
An external environment recognition apparatus comprising: a host vehicle information determination unit that includes a collision determination selection unit that selects the second collision determination unit when the host vehicle information acquired by the host vehicle information acquisition unit is determined to be abnormal. Own vehicle information acquisition means for acquiring own vehicle information;
First object information acquisition means for acquiring information of an object detected by the first external recognition sensor;
Predicted course setting means for setting a predicted course of the own vehicle based on the own vehicle information acquired by the own vehicle information acquisition means;
Based on the object information acquired by the first object information acquisition means and the predicted course set by the predicted course setting means, the possibility of collision between the object detected by the first external recognition sensor and the host vehicle is determined. First collision determination means;
Based on the object information acquired by the first object information acquisition means, second collision determination means for determining the possibility of collision between the object detected by the first external recognition sensor and the own vehicle;
Second object information acquisition means for acquiring object information using the second external recognition sensor for an object determined by the first collision determination means to have a high possibility of collision;
Vehicle information determination means for determining abnormality of the vehicle information acquired by the vehicle information acquisition means;
An external environment recognition device comprising: a collision determination selection unit that selects the second collision determination unit when the vehicle information determination unit determines that the vehicle information acquired by the vehicle information acquisition unit is abnormal.   The external environment recognition device according to claim 1, wherein the own vehicle information includes at least one of a vehicle speed, a steering angle, and a yaw rate, and the object information indicates a relative position with respect to the own vehicle.   The external environment recognition device according to claim 1, wherein the own vehicle information acquisition unit acquires own vehicle information by a predetermined communication unit, and the own vehicle information determination unit determines whether or not the communication unit has failed. . Own vehicle information acquisition means for acquiring own vehicle information;
An external recognition sensor that recognizes the lane,
Object information acquisition means for acquiring information of an object detected by the external world recognition sensor;
Predicted course setting means for setting a predicted course of the host vehicle according to the host vehicle information acquired by the host vehicle information acquiring unit;
Based on the object information acquired by the object information acquisition means and the predicted course set by the predicted course setting means, first collision determination means for determining the possibility of collision between the object detected by the external field recognition sensor and the own vehicle. When,
Vehicle information determination means for determining abnormality of the vehicle information acquired by the vehicle information acquisition means;
When the own vehicle information determining means determines that the own vehicle information acquired by the own vehicle information acquiring means is abnormal, the predicted course setting means is based on the road center line obtained from the lane recognized by the external recognition sensor. An external recognition device that sets the predicted course of the vehicle. Own vehicle information acquisition means for acquiring own vehicle information;
First and second external recognition sensors for recognizing a lane;
First object information acquisition means for acquiring information of an object detected by the first external recognition sensor;
Predicted course setting means for setting a predicted course of the host vehicle according to the host vehicle information acquired by the host vehicle information acquiring unit;
Based on the object information acquired by the first object information acquisition means and the predicted course set by the predicted course setting means, the possibility of collision between the object detected by the first external recognition sensor and the host vehicle is determined. First collision determination means;
Second object information acquisition means for acquiring object information using the second external field recognition sensor for an object determined by the first collision determination means to have a high possibility of collision;
Vehicle information determination means for determining abnormality of the vehicle information acquired by the vehicle information acquisition means;
When the vehicle information determination unit determines that the vehicle information acquired by the vehicle information acquisition unit is abnormal, the predicted course setting unit calculates the road center line obtained from the lane recognized by the second external sensor. An external recognition device that sets the predicted course of the vehicle according to the vehicle.

Images