Light distribution control system and light distribution control method

Abstract

The present invention carries out external recognition for a vehicle and carries out light distribution control of a headlight installed in the vehicle on the basis of the external recognition for the vehicle. The configuration comprises an external sensor that acquires information about an object in the surroundings of the vehicle and a ground speed acquisition unit that acquires the ground speed of the object on the basis of output from the external sensor, wherein the coverage range of the headlight is set variably on the basis of the ground speed according to the ground speed acquisition unit.

Description

Technical Field

[0001] The present invention relates to a light distribution control system and a light distribution control method.

Background Art

[0002] In recent years, vehicles such as automobiles equipped with a light distribution control system that recognizes the external environment of the vehicle and controls the light distribution of vehicle headlights based on the recognition results have been put into practical use. As light distribution control mounted on vehicles, there is a function called AHB (Auto High Beam), which uses an external environment recognition sensor such as an on-vehicle camera to recognize oncoming vehicles and preceding vehicles, automatically switches to low beam, and automatically switches back to high beam after passing the oncoming vehicle or preceding vehicle. There is also a function called ADB (Adaptive Driving Beam) that shields light from the area of the recognized oncoming vehicle or preceding vehicle, and irradiates high beam to areas other than the shielded area.

[0003] In order to correctly shield light from other vehicles such as oncoming vehicles and preceding vehicles, it is necessary for the external environment recognition sensor to accurately distinguish between vehicles and non-vehicles. Patent Document 1 describes a technology that narrows down light spots such as tail lamps and headlights based on shape, and performs vehicle determination based on movement direction and speed. However, with conventional technologies, it is not always possible to perform appropriate vehicle determination: there are cases where it is difficult to detect light spots due to disturbance, and cases where there are unlit vehicles that cannot be recognized. Therefore, when such vehicles cannot be recognized, there has been a problem that glare is caused to drivers of other vehicles.

Prior Art Literature

Patent Literature

[0004]

Patent Document 1

Summary of the Invention

Problem to be Solved by the Invention

[0005] When performing automatic light distribution control, it is undesirable to shine high beams on other vehicles. However, as mentioned above, if it is difficult to detect the lights of other vehicles due to disturbances, or if a vehicle is not using its lights, it may be impossible to detect the lights of other vehicles, potentially leading to the high beams being shone on. Therefore, there has been a demand for the development of technology that can perform light distribution control using methods other than detecting the lights of other vehicles.

[0006] Furthermore, when recognizing vehicles from images, false detections can occur due to roadside objects and other factors. If a roadside object is mistakenly identified as a vehicle, the system may incorrectly switch to low beams. Therefore, it was desirable to prevent false detections of roadside objects and other factors as much as possible.

[0007] The present invention aims to provide a light distribution control system and a light distribution control method that can effectively detect other vehicles while preventing false detection of objects other than vehicles. [Means for solving the problem]

[0008] To solve the above problems, for example, the configuration described in the claims is adopted. This invention includes several means for solving the above-mentioned problems, but to give one example, the light distribution control system of the present invention recognizes the external environment of the vehicle and controls the light distribution of the headlights mounted on the vehicle based on the recognition of the external environment of the vehicle. Here, an example of the light distribution control system of the present invention includes an external sensor that acquires information about objects around the vehicle, and a ground speed acquisition unit that acquires the ground speed of the object based on the output of the external sensor. The system includes an image acquisition unit that acquires images of the area around the vehicle, and a light spot detection unit that detects light spots contained in the images. It is equipped with a system that allows the illumination range of the headlights to be variably set based on the ground speed acquired by the ground speed acquisition unit. Furthermore, if the light spot detection unit does not detect a light spot, and the ground speed acquisition unit determines that the object's ground speed is equal to or greater than the first threshold, the vehicle's headlights are set to either low beam or high beam with the object obscured. [Effects of the Invention]

[0009] According to the present invention, by acquiring the ground velocity of an object and controlling the illumination of the vehicle's headlights when the ground velocity exceeds a threshold, it is possible to prevent the vehicle from shining its high beams at other vehicles in a manner other than detecting the vehicle's lights. Other issues, configurations, and effects not mentioned above will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0010] [Figure 1] This is a block diagram showing an example configuration of a light distribution control system according to a first embodiment of the present invention. [Figure 2] This is a block diagram showing an example of the hardware configuration of a recognition processing device included in a light distribution control system according to a first embodiment of the present invention. [Figure 3] This is a flowchart showing the light distribution control process according to a first embodiment of the present invention. [Figure 4] Figure 4A shows an example of a candidate for light shielding (Example 1), and Figure 4B shows an example of a state where a light shielding target has been selected from the candidates for light shielding shown in Figure 4A. [Figure 5] Figure 5A shows an example of a candidate for light shielding (Example 2), and Figure 5B shows an example of a state where a target that is not shielded is selected from the candidates for not shielding shown in Figure 5A. [Figure 6] This is a block diagram showing an example configuration of a light distribution control system according to a second embodiment of the present invention. [Figure 7] This is a flowchart showing the light distribution control process according to a second embodiment of the present invention. [Modes for carrying out the invention]

[0011] <Example of the first embodiment> Hereinafter, a light distribution control system and light distribution control method according to a first embodiment of the present invention will be described with reference to Figures 1 to 5.

[0012] [Configuration of the light distribution control system] FIG. 1 shows a configuration of a light distribution control system according to the present embodiment. The light distribution control system 1 shown in FIG. 1 is mounted on a vehicle including a headlight (not shown), and includes a recognition processing device 10 and a vehicle control device 20. Based on the result recognized by the recognition processing device 10, the vehicle control device 20 performs light distribution control for the vehicle's headlight.

[0013] Furthermore, the light distribution control system 1 includes an external sensor 14 installed in front of the vehicle to acquire information ahead of the vehicle. The external sensor 14 is a sensor that performs external sensing processing to acquire information ahead of or around the vehicle. As the external sensor 14, a stereo camera, a monocular camera, a millimeter-wave radar, LiDAR (Light Detection And Ranging), or the like can be used. A stereo camera uses a plurality of cameras to capture images of the area ahead of and around the vehicle. A monocular camera uses a single camera to capture images of the area ahead of and around the vehicle. A millimeter-wave radar acquires distances and angles of objects ahead of and around the vehicle. LiDAR acquires three-dimensional information of the area ahead of and around the vehicle.

[0014] Note that as the external sensor 14, these sensors may be used alone, or a combination of a monocular camera, a millimeter-wave radar, or the like may be used. Additionally, any sensor other than those described above may be used as long as it can perform sensing around the vehicle.

[0015] Both the recognition processing device 10 and the vehicle control device 20 are computers. FIG. 2 shows an example of the hardware configuration of the recognition processing device 10 and the vehicle control device 20 implemented as computers. Although FIG. 2 illustrates the configuration as that of the recognition processing device 10, the vehicle control device 20 may also have the same hardware configuration as the recognition processing device 10. The recognition processing device 10 includes a central processing unit (CPU) 11 serving as a processor, a read-only memory (ROM) 11a, a random access memory (RAM) 11b, a storage device 12, a network interface 11c, an input unit 11d, and an output unit 11e. The CPU 11 may be formed of an MPU (Micro Processor Unit). For the storage device 12, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or a semiconductor memory is used. The ROM 11a may also be configured to serve as the storage device 12.

[0016] The input unit 11d converts various types of information input from an external sensor 14 to the recognition processing device 10 or the vehicle control device 20 into information that can be processed by the CPU 11. The ROM 11a and the storage device 12 are recording media in which a control program for executing arithmetic processing described later as appropriate and various types of information necessary for executing the arithmetic processing are stored.

[0017] In accordance with a control program stored in the ROM 11a, the CPU 11 performs predetermined arithmetic processing on a signal input to the input unit 11d and signals fetched from the ROM 11a, the RAM 11b, or the storage device 12. From the output unit 11e, a command for controlling an output target, information to be used by the output target, and the like are output. Here, in a case where the present device is the recognition processing device 10, the output target is the light distribution control unit 21 of the vehicle control device 20, and in a case where the present device is the vehicle control device 20, the output target is the headlight of a vehicle whose light distribution is controlled.

[0018] Note that configuring the recognition processing device 10 and the vehicle control device 20 with a computer including a CPU is merely an example; for example, part or all of the recognition processing device 10 and the vehicle control device 20 may be implemented by dedicated hardware such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

[0019] Returning to the explanation of the configuration in Figure 1, the recognition processing device 10 includes a ground speed acquisition unit 13. The ground speed acquisition unit 13 performs ground speed acquisition processing to acquire the relative velocity of an object in front of the vehicle based on the information acquired by the external sensor 14. For example, the ground speed acquisition unit 13 uses 3D information acquired by a stereo camera or LiDAR to recognize an obstacle in front of the vehicle as a three-dimensional object, and calculates the ground speed from the time-series data of the distance information of the recognized three-dimensional object.

[0020] Alternatively, the ground velocity acquisition unit 13 uses images captured by a stereo camera or monocular camera to pre-calculate feature quantities such as edges and colors from images of the target to be recognized and other images, thereby training the classifier. Then, the ground velocity acquisition unit 13 uses the trained classifier to identify the newly acquired target to be recognized. For training the classifier, for example, SVM (Support Vector Machine) or AdaBoost can be used.

[0021] Furthermore, the ground velocity acquisition unit 13 may perform deep learning, such as a Convolutional Neural Network (CNN), using images of the target to be recognized and images of other targets in advance. The ground velocity acquisition unit 13 may then use the recognition network trained by the CNN to recognize the target to be recognized.

[0022] Furthermore, after identifying these recognition targets, the ground speed acquisition unit 13 can estimate distance information to the recognition target from the image size of the recognition target, or it can determine the ground speed from time-series data of distance information to the recognition target measured using sensors such as millimeter-wave radar.

[0023] The vehicle control device 20 includes a light distribution control unit 21. The light distribution control unit 21 controls the light distribution of the vehicle's headlights and sets the illumination range of the headlights variably based on the ground speed information of objects such as vehicles around the vehicle acquired by the ground speed acquisition unit 13 of the recognition processing device 10. In other words, for example, in the case of a vehicle equipped with an AHB function that automatically switches to high beams after passing an oncoming or preceding vehicle, the light distribution control unit 21 switches the headlights to low beam when light distribution control is necessary. Also, when light distribution control is not necessary, the light distribution control unit 21 switches the headlights to high beam.

[0024] Furthermore, in the case of a vehicle equipped with an ADB (Adaptive Driving Beam) function that blocks light from the area of ​​a recognized oncoming or preceding vehicle and illuminates the area other than the blocked area with high beams, the light distribution control unit 21 blocks light from the area of ​​the corresponding headlight when light distribution control is required, and illuminates the other areas with high beams.

[0025] [Process flow for controlling the light distribution of headlights] Figure 3 is a flowchart showing the process flow for controlling the light distribution of the headlights in the light distribution control system of this embodiment.

[0026] First, the ground speed acquisition unit 13 of the recognition processing device 10 acquires information from the external sensor 14 and calculates and acquires the ground speed of objects around the vehicle based on the information from the external sensor 14 (step S101). The ground speed acquisition unit 13 acquires the ground speed for each object. For example, if an oncoming vehicle and a preceding vehicle are present in front of the vehicle, the ground speed acquisition unit 13 acquires the ground speed of the oncoming vehicle and the preceding vehicle, respectively.

[0027] Next, the recognition processing device 10 determines whether the ground speed acquired in step S101 is above a threshold for each object (step S102). Here, the threshold in step S102 is set to a constant speed suitable for distinguishing between a moving vehicle and an object on the road, such as 40 km / h. In step S102, to determine whether the ground velocity is above a threshold, a method of using the results of multiple measurements can be considered to prevent misidentification of roadside objects or walls.

[0028] Furthermore, when traveling on the same road, it can be assumed that oncoming and preceding vehicles are traveling at a similar speed to your own vehicle, so the threshold for ground speed and the number of ground speed measurements may be changed based on your own vehicle's speed. Additionally, when there are many roadside objects, such as on a curve, the possibility of misjudgment increases, so the threshold for ground speed and the number of ground speed measurements may be changed based on the curvature of the road you are traveling on.

[0029] Then, if step S102 determines that the ground speed is above a threshold (True in step S102), the recognition processing device 10 determines that the object should have its high beams blocked, and the vehicle control device 20 controls the headlights so as not to dazzle the object (step S103). Controlling the headlights so as not to dazzle objects such as oncoming vehicles is performed, for example, by the AHB function or ADB function.

[0030] If the ground speed is determined to be below a threshold in step S102 (False in step S102), the recognition processing device 10 determines that the vehicle is not a target for shading, and the vehicle control device 20 controls the headlights so as not to change the illumination of the high beams to the target (step S104).

[0031] [Example of the determination state of the recognition target] Figures 4 and 5 are illustrative diagrams showing examples of the determination state according to this embodiment. Figures 4 and 5 are images of the front of the vehicle captured by the camera. The areas enclosed by rectangles in each image indicate the parts detected by the recognition processing device 10 as objects to be shielded from light.

[0032] Figure 4A is an image 100a showing a situation where a preceding vehicle 102, an oncoming vehicle 101, a streetlamp 103, and a sign 104 are visible. Since the preceding vehicle 102 and the oncoming vehicle 101 are traveling on the road, their ground speed can be obtained. Therefore, in step S102, if it can be determined that the obtained ground speed is above a threshold, the recognition processing device 10 sets the preceding vehicle 102 and the oncoming vehicle 101 as targets for light shielding, as shown in Figure 4B, with the recognition result being image 100b. Furthermore, since the sign 104 and the streetlamp 103 are road fixtures, their ground speed is determined to be below a certain level in step S102, so the recognition processing device 10 does not consider them to be objects of light shielding in the recognition result image 100b.

[0033] Figure 5A is an image 200a showing a situation where only roadside objects such as sign 201, a traffic guidance marker (arrow type) 202, and a traffic guidance marker (pole type) 203 are visible.

[0034] In a situation like image 200a shown in Figure 5A, in step S102, the sign 201 and the guideposts 202 and 203 are determined to have a ground speed below a certain level. Therefore, as shown in Figure 5B, the sign 201 and the guideposts 202 and 203 are not considered targets for shading in the recognition result image 200b. Thus, the light distribution control unit 21 can illuminate with high beams when the AHB function is active. Furthermore, even when the ADB function is active, the light distribution control unit 21 can illuminate with high beams without shading.

[0035] The scenes shown in Figures 4 and 5 are just examples of the target scenarios, and the determination state according to this embodiment can handle a variety of other situations.

[0036] As described above, according to this embodiment, the recognition processing device 10 determines an object detected by the external sensor 14 to be a light-blocking target only if its ground speed is above a certain speed. Therefore, it is possible to reliably distinguish between a moving vehicle and an object fixed on the road and determine which object is a light-blocking target. Consequently, it is possible to prevent high beams from shining on other vehicles whose ground speed is above or below a certain speed.

[0037] <Second Embodiment Example> Next, a light distribution control system and light distribution control method according to a second embodiment of the present invention will be described with reference to Figures 6 and 7. In Figures 6 and 7, the same reference numerals are used for the same parts as those described in Figures 1 to 5 of the first embodiment, and redundant explanations are omitted.

[0038] [Configuration of the light distribution control system] Figure 6 shows the configuration of the light distribution control system in this embodiment. As shown in Figure 6, the recognition processing device 10 in the light distribution control system 1' of this embodiment includes, in addition to the ground speed acquisition unit 13 described in the first embodiment, an image acquisition unit 15, a light point detection unit 16, and a vehicle detection unit 17. In other words, the CPU (Central Processing Unit) 11 of the recognition processing device 10 is configured to function as an image acquisition unit 15, a light point detection unit 16, a vehicle detection unit 17, and a ground speed acquisition unit 13. The light distribution control system 1' shown in Figure 6 is also configured by the computer shown in Figure 2.

[0039] The image acquisition unit 15 acquires images captured by the camera (stereo camera or monocular camera) which serves as the external sensor 14. The light spot detection unit 16 detects pixels with a brightness above a certain value from the image acquired by the image acquisition unit 15, and integrates the area adjacent to the pixel to form a light spot region. Furthermore, the light spot detection unit 16 identifies the type of light spot by using the brightness information of the light spot region, the relative distance from the vehicle, and the relative position information from the vehicle to determine whether it is a light spot from the vehicle's headlights or taillights or another type of light spot.

[0040] The vehicle detection unit 17 detects vehicles from image parallax information (in the case of a stereo camera) or image features. When detecting vehicles, the vehicle detection unit 17 can also utilize machine learning. In other words, the vehicle detection unit 17 uses images captured by a stereo camera or monocular camera to identify vehicles using a pre-trained classifier.

[0041] The trained classifier is trained using SVM (Support Vector Machine) or AdaBoost by calculating features such as edges and color from images of the target and non-target objects. Then, the vehicle detection unit 17 uses the trained classifier to calculate features such as edges and color from the images acquired by the image acquisition unit 15 and perform identification.

[0042] Alternatively, the classifier may train a recognition network using deep learning such as a CNN (Convolutional Neural Network) on images of the target to be recognized and other images, and then use the trained network to identify the target to be recognized, such as a vehicle, from the images acquired by the image acquisition unit 15.

[0043] The ground speed acquisition unit 13 acquires the ground speed of objects around the vehicle, similar to the first embodiment example described in Figure 1. Furthermore, the vehicle control device 20 controls the light distribution of the vehicle's headlights based on the information about vehicles and other objects in the vicinity of the vehicle acquired by the recognition processing device 10.

[0044] [Process flow for controlling the light distribution of headlights] Figure 7 is a flowchart showing the process flow for controlling the light distribution of the headlights in the light distribution control system of this embodiment. First, the light spot detection unit 16 recognizes light spots that are candidates for light shielding in the light distribution control unit 21 based on the image acquired by the image acquisition unit 15 (step S201). Then, in step S201, the light spot detection unit 16 determines whether or not any light spots have been detected (step S202).

[0045] If a candidate light point for the light-blocking target is recognized in step S202 (false in step S202), the recognition processing device 10 determines that the recognized light point is a light-blocking target (step S208). The result of determining that the light-blocking target is a target in step S208 is transmitted from the recognition processing device 10 to the vehicle control device 20. As a result, the vehicle control device 20 sets the control state of the headlights to either low beam or high beam with the corresponding light-blocking target blocked.

[0046] Furthermore, if no light point that is a candidate for the light-shielding target is recognized in step S202 (True in step S202), the vehicle detection unit 17 performs vehicle identification on the image acquired by the image acquisition unit 15 (step S203). Furthermore, the ground speed acquisition unit 13 acquires the ground speed of objects around the vehicle (step S204). Ground speed acquisition is performed for each object; for example, if an oncoming vehicle and a preceding vehicle are present in front of the vehicle, the ground speed of the oncoming vehicle and the preceding vehicle are acquired separately.

[0047] Subsequently, the recognition processing device 10 determines in step S203 whether or not vehicle identification was performed (step S205). If it is determined in step S205 that vehicle identification was performed (True in step S205), the recognition processing device 10 determines whether or not the ground speed obtained in step S204 is equal to or greater than threshold a (step S206). Here, threshold a is, for example, 40 km / h.

[0048] If the recognition processing device 10 determines in step S206 that the ground speed is greater than or equal to threshold a (True in step S206), it determines that the corresponding object is a light-blocking target (step S208). This determination of a light-blocking target in step S208 is transmitted from the recognition processing device 10 to the vehicle control device 20, and the vehicle control device 20 controls the headlights to a state corresponding to the determination of a light-blocking target.

[0049] Furthermore, if the ground speed is determined to be less than threshold a in step S206 (False in step S206), the recognition processing device 10 determines that the object is not subject to light shielding (step S209). In this determination in step S209, even if a vehicle has been identified in step S203, the recognition processing device 10 determines that there is a possibility of misidentification of roadside objects, etc., and performs the determination in step S209.

[0050] Furthermore, if it is determined in step S205 that there is no vehicle identification (false in step S205), it is determined whether the ground speed obtained in step S204 is greater than or equal to threshold b (step S207). Here, threshold b is set to a value greater than the threshold a mentioned above, for example, 80 km / h.

[0051] In step S207, if the recognition processing device 10 determines that the ground speed is greater than or equal to threshold b (True in step S207), the recognition processing device 10 determines that the corresponding object is a target for light blocking (step S208). This determination of the target for light blocking is transmitted from the recognition processing device 10 to the vehicle control device 20, and the vehicle control device 20 controls the headlights to the corresponding state.

[0052] Furthermore, if step S207 determines that the ground speed is less than threshold b (False in step S207), the recognition processing device 10 determines that the identified vehicle is not subject to light shielding (step S209).

[0053] As explained above, in this embodiment, if a light spot is detected in step S202, it is immediately determined to be a target for light shielding. If no light spot is detected in step S202, the light shielding determination is made based on a combination of the vehicle identification results in steps S203 and S204 and the ground speed result.

[0054] As a result, even if, for example, in step S203, a portion of the traffic guidance markers 202 and 203, as explained in Figure 5, is mistakenly identified as a vehicle, the ground speed can be checked in step S206, allowing it to be determined in step S209 that the vehicle is not subject to light shielding. Therefore, in scenes where high-beam illumination is required, as shown in Figure 7, it is possible to prevent accidentally switching to low beam.

[0055] Furthermore, in scenes with preceding or oncoming vehicles, as shown in Figure 4, even if light point detection is not performed in step S202 due to false detection in rainy weather, etc., and vehicle identification is not performed in step S205, by checking the ground speed in step S207, the preceding vehicle 102 and oncoming vehicle 101 can be designated as targets for light shielding in step S208. This makes it possible to reliably target the preceding vehicle 102 or oncoming vehicle 101 as objects to be shielded from light, even when it is difficult to detect light points or identify vehicles from images, by utilizing ground speed, thereby preventing glare.

[0056] Furthermore, the vehicle detection unit 17 in this embodiment can detect vehicles more accurately by utilizing machine learning, such as a trained classifier, thereby improving the accuracy of headlight control.

[0057] As described above, according to this embodiment, similar to the first embodiment, it is possible to prevent the illumination of high beams to other vehicles that should not be illuminated. Furthermore, in this embodiment, when detecting light spots or vehicles and deciding on the illumination of the vehicle's headlights based on the detection results, appropriate control can be performed even if the respective recognition results are not obtained. In other words, in this embodiment, by acquiring the ground speed of objects around the vehicle and controlling the illumination of the vehicle's headlights when the ground speed is above a threshold, the illumination of the headlights can be controlled based on the ground speed even when it is difficult to detect light spots or vehicles from the image due to disturbances such as rain.

[0058] Furthermore, in this embodiment, even if a vehicle is detected as a vehicle, the ground speed of the object is obtained, and if the ground speed is below a threshold, it is determined to be a roadside object. As a result, according to this embodiment, even if a roadside object is mistakenly detected as a vehicle, it is possible to drive while maintaining the high beams of the vehicle's headlights without blocking the illumination of the roadside object.

[0059] Furthermore, in this embodiment, the vehicle detection unit 17 uses machine learning to detect vehicles from image parallax information and image features. This allows the vehicle detection unit 17 to detect vehicles more accurately. However, using machine learning is just one example, and vehicles may be detected by other methods.

[0060] [Differentiation] The embodiments described above are explained in detail for the purpose of clearly illustrating the present invention, and are not necessarily limited to those comprising all the configurations described.

[0061] For example, the values ​​of 40 km / h and 80 km / h used as examples of threshold speeds in the above-described embodiments are merely examples, and higher or lower speeds may be used. For instance, if the accuracy of the speed acquired by the ground speed acquisition unit 13 is high and the error is small, the 40 km / h threshold value in the first embodiment and the 40 km / h threshold value a in the second embodiment may be set to lower speed values. Furthermore, these thresholds may be variably set according to the vehicle's driving speed.

[0062] Furthermore, in the embodiments described above, the process of distinguishing a moving vehicle from other objects using the flowcharts in Figures 2 and 7 was applied to the light distribution control of the headlights. However, the recognition results from the recognition processing device 10 may also be used by the vehicle control device 20 for control other than the headlights.

[0063] For example, when the vehicle control device 20 performs control as an autonomous driving system (AD) or an advanced driver-assistance system (ADAS), the recognition processing device 10 can perform headlight distribution control by performing identification processing between the moving vehicle and other objects. Furthermore, AD and ADAS can be one of the pieces of information necessary for vehicle recognition processing.

[0064] In this way, by applying the recognition results from the recognition processing device 10 to functions other than headlight light distribution control, it becomes possible to perform autonomous driving and advanced driver assistance more accurately and precisely.

[0065] Furthermore, the configuration diagrams shown in Figures 1 and 6 only show control lines and information lines that are deemed necessary for explanation, and do not necessarily show all control lines and information lines in the actual product. In reality, it is reasonable to assume that almost all components are interconnected.

[0066] Furthermore, in the example described in this embodiment, the vehicle control device 20, which acquires the recognition result from the recognition processing device 10, controls the headlights. Alternatively, the vehicle control device 20 may perform part or all of the recognition processing. Or, the recognition processing device 10 may directly control the control of the high beam and low beam of the headlights, or set the light-shielding range when the high beam is on.

[0067] Furthermore, when the recognition processing unit 10 and the vehicle control unit 20 are configured as information processing devices such as computers, the programs that implement the recognition processing unit 10 and the vehicle control unit 20 may be stored not only in non-volatile storage or memory within the computer, but also on external memory, IC cards, SD cards, optical discs, or other recording media, and transferred for use. [Explanation of symbols]

[0068] 1,1′…Light distribution control system, 10…Recognition processing unit, 11…Central processing unit (CPU), 11a…Read-only memory (ROM), 11b…Random access memory (RAM), 11c…Network interface, 11d…Input unit, 11e…Output unit, 12…Storage device, 13…Ground speed acquisition unit, 14…Outside sensor, 15…Image acquisition unit, 16…Light point detection unit, 17…Vehicle detection unit, 20…Vehicle control device, 21…Light distribution control unit, 100a,100b…Image, 101…Oncoming vehicle, 102…Preceding vehicle, 103…Streetlight, 104…Sign, 200a,200b…Image, 201…Sign, 202,203…Lane guidance marker

Claims

1. A light distribution control system that recognizes the external environment of a vehicle and controls the light distribution of the headlights mounted on the vehicle based on the recognition of the external environment of the vehicle, An external sensor that acquires information about objects around the vehicle, A ground velocity acquisition unit acquires the ground velocity of the object based on the output of the external sensor, An image acquisition unit that acquires images of the area around the vehicle, The system includes a light spot detection unit that detects light spots included in the aforementioned image, Based on the ground speed acquired by the ground speed acquisition unit, the illumination range of the headlights is variably set. If the light spot detection unit does not detect a light spot, and the ground speed acquisition unit determines that the object's ground speed is equal to or greater than the first threshold, the vehicle's headlights are set to either low beam or high beam, which shields the object from light. Light distribution control system.

2. In the ground speed acquisition unit, if the ground speed of the object is less than a first threshold, the direction of the vehicle's headlights is set to high beam. The light distribution control system according to claim 1.

3. Furthermore, it includes a vehicle detection unit that detects vehicles included in the aforementioned image, If the light spot detection unit does not detect a light spot, and the vehicle detection unit detects a vehicle, and the ground speed acquisition unit determines that the ground speed of the object is equal to or greater than the first threshold, the vehicle's headlights are set to either low beam or high beam, which obscures the object. The light distribution control system according to claim 1.

4. Furthermore, it includes a vehicle detection unit that detects vehicles included in the aforementioned image, If the light spot detection unit does not detect a light spot, and the vehicle detection unit does not detect a vehicle, and the ground speed acquisition unit determines that the ground speed of the object is greater than or equal to a second threshold that is faster than the first threshold, the vehicle's headlights are set to either low beam or high beam that obscures the object. The light distribution control system according to claim 1.

5. The aforementioned vehicle detection unit detects vehicles from image disparity information and image features obtained through machine learning. The light distribution control system according to claim 3.

6. In the ground speed acquisition unit, the ground speed information of the object is also used as object information for the purpose of autonomous driving or advanced driver assistance of the vehicle. The light distribution control system according to claim 1.

7. A light distribution control method that performs external environment recognition of a vehicle and controls the light distribution of headlights mounted on the vehicle based on external environment recognition of the vehicle, External sensing processing to acquire information about objects around the vehicle, Based on the information obtained by the external sensing process, a ground velocity acquisition process is performed to acquire the ground velocity of the object, Image acquisition process to obtain images of the area around the vehicle, This includes a light spot detection process that detects light spots included in the aforementioned image, Based on the ground speed acquired in the ground speed acquisition process, the illumination range of the headlights is set to be variable. If no light spot is detected in the light spot detection process, and the ground speed of the object is equal to or greater than the first threshold in the ground speed acquisition process, the direction of the vehicle's headlights is set to low beam or high beam that shields the object from light. Light distribution control method.

Images