JP2014101068A - False determination device and vehicle control system - Google Patents

Abstract

An error detection determination apparatus and a vehicle control system for determining in a short time whether or not camera data based on imaging by a camera mounted on a vehicle is erroneous detection data.
An error detection determination device acquires camera information based on imaging by a camera mounted on a vehicle, and performs control related to the vehicle based on the camera information acquired by the camera information acquisition unit. Based on the behavior information acquired by the vehicle control unit that generates and outputs the control signal for performing, the behavior information acquisition unit that acquires the behavior information indicating the behavior of the vehicle, and the behavior information acquisition unit, the camera information acquisition unit A false detection determination unit that determines whether or not the acquired camera information is false detection information.
[Selection] Figure 1

Description

  The present invention relates to a technique for determining erroneous detection by a camera mounted on a vehicle.

  Conventionally, image data sent from a camera mounted on the host vehicle, or data such as lane information, forward vehicle position information, and pedestrian position information obtained by analyzing the image data (hereinafter referred to as these data) A system for performing vehicle control (for example, “camera data” collectively) is input to an ECU (control unit) and control data generated based on the camera data is output to a control target such as a lamp unit (for example, (Night vision, auto high beam, lane keep, ADB, obstacle / human detection system, etc.) are known (see, for example, Patent Document 1).

  These systems basically perform the necessary control based on the correct camera data input from the camera. For example, in the ADB system, the position information of the preceding vehicle is input from a camera, and the light irradiation pattern from the host vehicle is controlled so that the position of the preceding vehicle falls within the light shielding range. In the marking light system, a pedestrian is detected from the camera data, and the marking light is emitted from the own vehicle to the position of the pedestrian, thereby prompting the driver to detect the pedestrian early.

  However, since not only normal data but also erroneous data due to erroneous detection may be sent from the camera, it is desirable that the ECU does not perform erroneous vehicle control based on erroneous camera data. In this regard, for example, in Patent Document 2, the pitch angle (tilt angle in the front-rear direction) of the host vehicle is detected, and the position coordinates of the preceding vehicle detected by the camera are corrected using the pitch angle. A light distribution control system for calculating the exact position coordinates of the vehicle ahead is described.

  However, due to camera vibration caused by the vibration of the vehicle body, erroneous detection data that is difficult to correct accurately by pitch angle correction may occur unexpectedly. Some high-performance cameras have their own anti-blur mechanism, and others have a mechanism that reduces false detection by image processing, but the ECU cannot determine the correctness of the camera data. In order to eliminate such erroneous detection data on the ECU side, conventionally, for example, moving average processing such as position information of a preceding vehicle input in time series has been performed.

  However, in a system that controls a vehicle by detecting a moving object (vehicle, pedestrian, etc.), the vehicle control (turns off the light of the lamp, emits a warning sound, brakes, etc.) after detecting the object in front of the vehicle. If there is a time lag before the application, the vehicle control effect is greatly reduced. Therefore, it is necessary to obtain control data from input camera data in a short time. On the other hand, when moving average processing is performed as in the prior art, there is an inconvenience that a time lag is promoted.

JP 2011-57160 A JP 2011-31808 A

One specific object of the present invention is to provide a false detection determination device that can quickly determine whether camera data is false detection data.
Another aspect of the specific aspect of the present invention is to provide a vehicle control system capable of reducing inappropriate control based on erroneous detection data.

  An error detection determination apparatus according to an aspect of the present invention provides a vehicle information based on a camera information acquisition unit that acquires camera information based on imaging by a camera mounted on a vehicle, and camera information acquired by the camera information acquisition unit. Based on the behavior information acquired by the vehicle control unit that generates and outputs a control signal for performing related control, the behavior information acquisition unit that acquires behavior information indicating the behavior of the vehicle, and the behavior information acquisition unit, the camera information And a false detection determination unit that determines whether the camera information acquired by the acquisition unit is false detection information.

  According to the above configuration, in order to determine whether the camera information acquired by the camera information acquisition unit is false detection information based on the behavior information, for example, when a significant behavior change occurs in the vehicle, based on that Thus, it can be determined in a short time that the camera information is false detection information.

  In the above-described erroneous detection determination device, it is preferable that the vehicle control unit generates and outputs a control signal based on the camera information only when the camera information is determined not to be erroneous detection information by the erroneous detection determination unit.

  According to the said structure, the control regarding a vehicle can be performed more appropriately.

  In the erroneous detection determination apparatus, the behavior information acquisition unit can acquire vehicle height information as behavior information, for example.

  According to the above configuration, the behavior of the vehicle can be easily acquired. Further, in many cases, a general vehicle is provided with a vehicle height sensor, and in that case, it is not necessary to newly provide a vehicle height sensor, which contributes to the cost reduction of the erroneous detection determination device.

  In the above erroneous detection determination device, the erroneous detection determination unit determines that the height difference of the vehicle height value within a certain time of the vehicle exceeds a predetermined threshold based on the behavior information acquired by the behavior information acquisition unit. In such a case, it is also preferable to determine that the camera information acquired within a certain period of time is erroneous detection information.

  When the change in vehicle height within a certain period of time is large, the camera vibrates due to the vertical movement of the vehicle body, and it is considered that the camera information is likely to be false detection information. Can be determined appropriately.

  A vehicle control system according to an aspect of the present invention includes the above-described erroneous detection determination device and a vehicle-mounted device that operates according to a control signal output from the erroneous detection determination unit. Here, the vehicle mounting device is mounted at an arbitrary position of the vehicle and performs an operation related to the operation of the vehicle. As an example of the vehicle-mounted device, there is a vehicle lamp that performs light irradiation by variably setting a light irradiation range according to a control signal.

  As a result, a vehicle control system is provided that can control the vehicle-mounted device with less errors within a short time according to the situation around the vehicle.

FIG. 1 is a block diagram illustrating a configuration of a vehicle control system according to an embodiment. FIG. 2A is a graph showing an example of a change in vehicle height over time when the vertical vibration of the vehicle body is small. FIG. 2B is a graph showing an example of a change in vehicle height over time when the vertical vibration of the vehicle body is large. FIG. 2C is a graph showing an example of a change in the height of the road surface on which the vehicle travels and a change in the vehicle height due to a change in the height of the road surface. FIG. 3 is a flowchart showing an operation procedure of the control unit of the vehicle control system.

  Embodiments of the present invention will be described below with reference to the drawings. As an example, the vehicle control system detects the positions of the vehicle and the pedestrian by analyzing image data obtained by imaging the space ahead of the host vehicle, and controls the light irradiation from the lamp unit based on the detection result. The case will be described.

  FIG. 1 is a block diagram illustrating a configuration of a vehicle control system according to an embodiment. The vehicle control system shown in FIG. 1 includes an imaging unit 10, a vehicle height sensor 11, a control unit 12, and a lamp unit 13. The imaging unit 10 includes a camera 101 and a vehicle / pedestrian detection unit 102. The control unit 12 includes a camera data acquisition unit 121, a sensor data acquisition unit 122, an erroneous detection determination unit 123, and a vehicle control unit (light distribution control unit) 124. The control unit 12 corresponds to “an erroneous detection determination device”.

  The camera 101 is installed at a predetermined position (for example, in the vicinity of an indoor rearview mirror) of the host vehicle, captures a space in front of the host vehicle at a predetermined cycle, and generates and outputs image data indicating the image.

  The vehicle / pedestrian detection unit 102 performs image processing using the image data output from the camera 101 to detect whether each of the forward vehicle and the pedestrian exists in the captured front space. If it exists, the position information is output to the control unit 12. Here, the “front vehicle” means a preceding vehicle or an oncoming vehicle. The vehicle / pedestrian detection unit 102 is realized by executing a predetermined operation program in a computer system having, for example, a CPU, a ROM, a RAM, and the like.

  Note that the vehicle / pedestrian detection unit 102 may be configured integrally with the camera 101. Further, the function of the vehicle / pedestrian detection unit 102 may be realized in the control unit 12.

  The vehicle height sensor 11 is disposed, for example, near the suspension in front of or behind the vehicle body, and detects the vehicle height (for example, the height of the vehicle body from the road surface). As the vehicle height sensor 11, any known sensor can be used. For example, the relative stroke between the cylinder and the piston rod may be detected as the vehicle height, or the road surface is irradiated with a laser beam from the vehicle body, and the reflected light is regarded as a light spot deviation on the photo detector. May calculate the displacement from the road surface.

  The control unit 12 is a computer that includes a CPU, ROM, RAM, and the like. When the CPU executes a predetermined operation program, the camera data acquisition unit 121, the sensor data acquisition unit 122, and the functional blocks illustrated in FIG. An erroneous determination detection unit 123 and a vehicle control unit 124 are realized. The camera data acquisition unit 121 corresponds to a “camera information acquisition unit”, and the sensor data acquisition unit 122 corresponds to a “behavior information acquisition unit”.

  The camera data acquisition unit 121 acquires vehicle and pedestrian position information from the vehicle / pedestrian detection unit 102 of the imaging unit 10. In the present embodiment, at least the position information of the vehicle corresponds to “camera data”.

  The sensor data acquisition unit 122 acquires sensor data (vehicle height information) indicating the vehicle height that is output from the vehicle height sensor 11. This sensor data corresponds to “behavior information”.

  When the erroneous detection determination unit 123 determines that the vehicle height difference in a certain time (for example, 1 second) exceeds a predetermined threshold based on the sensor data acquired by the sensor data acquisition unit 122, the error detection determination unit 123 The camera data acquired by the camera data acquisition unit 121 is determined as erroneous detection data.

  FIG. 2A is a graph showing an example of a change in vehicle height over time when the vertical vibration of the vehicle body is small. In this graph, the height difference of the vehicle body due to the change with time is small, and it is considered that the erroneous detection by the camera 101 hardly occurs. FIG. 2B is a graph showing an example of a change in vehicle height over time when the vertical vibration of the vehicle body is large. In this graph, the vehicle height difference is large in the time zones T1 and T2, and it is considered that erroneous detection by the camera 101 is likely to occur. The above “predetermined time” and “threshold value” can be determined in advance by analyzing the actual measurement values shown in such a graph.

  The conditions for determining the erroneous detection data are not limited to the above. For example, when the change degree of the vehicle height is larger than a predetermined change degree, the camera data acquired at that timing is erroneously detected. You may determine with detection data.

  FIG. 2C is a graph showing an example of a change in the height of a road (road surface) on which the vehicle travels and a change in the vehicle height due to a change in the height of the road. As indicated by portions R1, R2, R3, and R4 surrounded by dotted lines in this graph, the amount of change in vehicle height is large at the position where the slope of the road surface changes, and it is considered that erroneous detection by the camera 101 occurs. It is done. Therefore, when the slope of the road surface, that is, the degree of change in the pitch angle of the vehicle is measured and the slope exceeds a predetermined value, the acquired camera data can be determined as erroneous detection data. .

  When the erroneous detection determination unit 123 determines that the camera data is not erroneous detection data, the vehicle control unit 124 controls the light distribution of the lamp unit 13 of the vehicle based on the camera data. Specifically, the vehicle control unit 124 determines a region that should be the light irradiation range / non-irradiation range corresponding to the position of the pedestrian and the position of the preceding vehicle, and based on the determined light irradiation range. The light distribution pattern is determined, and a control signal corresponding to the light distribution pattern is generated and output to the lamp unit 13.

  The lamp unit 13 is installed on each of the front left and right sides of the host vehicle, and is used for irradiating light that illuminates the front of the host vehicle. The lamp unit 13 includes an LED driver 131 and a matrix LED 132. Here, only one lamp unit 13 is illustrated.

  The LED driver 131 selectively lights each LED by supplying a drive signal to a plurality of LEDs (light emitting diodes) included in the matrix LED 132 based on a control signal output from the vehicle control unit 124. .

  The matrix LED 132 includes a plurality of LEDs arranged in a matrix, and the plurality of LEDs are selectively lit based on a drive signal supplied from the LED driver 131. The matrix LED 22 can turn on a plurality of LEDs independently and control the irradiation intensity (brightness).

  Here, for example, when the input period of camera data from the imaging unit 10 to the control unit 12 is 40 milliseconds, and the calculation time required to generate control data from the camera data is 10 milliseconds, When the moving average process is performed with the camera data for 10 times to make a false detection determination, the camera data input time for 10 times takes 400 milliseconds (40 milliseconds × 10 times), so the camera for 10 times It takes a maximum of 450 (= 40 + 400 + 10) milliseconds to perform calculation by inputting data and outputting control data.

  On the other hand, in the above-described embodiment, when the control unit 12 receives the input of the camera data and the sensor data, it can perform the false detection determination in almost real time, so that the control data is output after the camera data is input. The time required for this is 50 (= 40 + 10) milliseconds at the maximum, and the response speed is 9 times that of the conventional case.

  Next, with reference to the flowchart shown in FIG. 3, the operation | movement procedure of the vehicle control system of this embodiment is demonstrated.

  First, the camera 101 of the imaging unit 10 images the front of the host vehicle at a predetermined cycle, and outputs image data generated by the imaging to the vehicle / pedestrian detection unit 102. The vehicle / pedestrian detection unit 102 of the imaging unit 10 analyzes the image data input from the camera 101, detects the positions of the preceding vehicle and the pedestrian, and controls the camera data indicating the detection contents (position information). To the unit 12.

  The camera data acquisition unit 121 of the control unit 12 acquires camera data output from the imaging unit 10 (step S11). Moreover, the sensor data acquisition part 122 of the control part 12 acquires sensor data from the vehicle height sensor 11 (step S12).

  The erroneous detection determination unit 123 analyzes the sensor data acquired by the sensor data acquisition unit 122, and detects the maximum value and the minimum value of the vehicle height within a certain past time from the present time (step S13).

  Next, the erroneous detection determination unit 123 determines whether the height difference of the vehicle height within a certain time exceeds a predetermined threshold value, thereby detecting the camera data acquired within the certain time as erroneous detection data. It is determined whether or not (step S14). If the vehicle height difference exceeds the threshold and it is determined that the camera data is erroneous detection data (step S14: YES), the process returns to step S11.

  On the other hand, when it is determined that the camera data acquired within a predetermined time is not false detection data (step S14: NO), the vehicle control unit 124 controls the light distribution of the lamp unit 13 based on the camera data. A signal is generated and output to the lamp unit 13 (step S15). Specifically, the vehicle control unit 124 sets a region to be a light irradiation range corresponding to the position of a pedestrian or a preceding vehicle indicated by the camera data, and sets a light distribution pattern based on the light irradiation range. Then, a control signal corresponding to the light distribution pattern is generated and output to the lamp unit 13. The LED driver 131 of the lamp unit 13 controls the light distribution pattern of the matrix LED 132 based on the control signal.

  As described above, in this embodiment, in order to determine whether or not the camera data is false detection data based on the vehicle height information which is an example of the behavior information, if a significant behavior change occurs in the vehicle, Based on this, it can be determined in a short time that the camera data is erroneous detection data. And it becomes possible to perform control of the lamp unit which is an example of a vehicle mounting apparatus without an error within a short time according to the surrounding situation of a vehicle.

  In addition, since a vehicle height sensor necessary for erroneous detection determination of camera data is mounted in advance in many vehicles, the cost can be reduced by using it. In addition, since an erroneous detection prevention function is unnecessary on the camera side, an inexpensive camera that does not have such a function can be used.

  In addition, this invention is not limited to the content of each embodiment mentioned above, In the range of the summary of this invention, it can change and implement variously.

  For example, in the above-described embodiment, the lamp unit has been described as an example of the vehicle-mounted device. However, the vehicle-mounted device may be any device that is mounted on the vehicle and performs some operation related to the operation of the vehicle. It may be a brake.

  Further, the vehicle behavior information is not limited to the vehicle height information obtained by the vehicle height sensor. For example, the vehicle orientation in the horizontal plane obtained by the gyro sensor and the inclination angle with respect to the vertical direction (for example, the vehicle center of gravity A pitch angle and a pitch angle, which are rotation angles around the horizontal axis of the center of gravity of the vehicle obtained by the tilt sensor, and the like. The vehicle speed and acceleration obtained by a speed sensor and an acceleration sensor may be used. That is, any vehicle behavior information that can be detected using various sensors may be used.

DESCRIPTION OF SYMBOLS 10: Imaging part 101: Camera 102: Vehicle / pedestrian detection part 11: Vehicle height sensor 12: Control part 121: Camera data acquisition part 122: Sensor data acquisition part 123: Misjudgment detection part 124: Vehicle control part 13: Lamp Unit 131: LED driver 132: Matrix LED

Claims (5)

A camera information acquisition unit for acquiring camera information based on imaging by a camera mounted on a vehicle;
A vehicle control unit that generates and outputs a control signal for performing control related to the vehicle, based on the camera information acquired by the camera information acquisition unit;
A behavior information acquisition unit for acquiring behavior information indicating the behavior of the vehicle;
Based on the behavior information acquired by the behavior information acquisition unit, an erroneous detection determination unit that determines whether the camera information acquired by the camera information acquisition unit is erroneous detection information;
Including a false detection determination device. The vehicle control unit generates and outputs the control signal based on the camera information only when the camera information is determined not to be the error detection information by the error detection determination unit.
The false detection determination device according to claim 1. The behavior information acquisition unit acquires vehicle height information of the vehicle as the behavior information.
The false detection determination device according to claim 1 or 2. The erroneous detection determination unit is configured to perform the predetermined time when the height difference of the vehicle height value within a predetermined time of the vehicle exceeds a predetermined threshold based on the behavior information acquired by the behavior information acquisition unit. Determining the camera information acquired within as false detection information,
The false detection determination device according to any one of claims 1 to 3. The false detection determination device according to any one of claims 1 to 4,
A vehicle-mounted device that operates according to a control signal output from the erroneous detection determination unit;
Including a vehicle control system.

Images