Cleaning system for imaging device

Abstract

To suppress degradation of the recognition performance of an imaging device after cleaning compared to the recognition performance of the imaging device before cleaning in a cleaning system for the imaging device.SOLUTION: A cleaning system 21 for cleaning imaging devices 12A-12C provided in a vehicle 1 comprises a cleaning device 22 that cleans the imaging devices, and a control device 20 that controls cleaning of the imaging devices by the cleaning device. The control device acquires a captured image 24 from the imaging device, calculates a current value obtained by converting a current state of contamination 26 of the imaging device into a numerical value on the basis of the captured image, calculates a predicted value obtained by converting a future state of contamination of the imaging device after operation of the cleaning device into a numerical value, and determines whether or not to operate the cleaning device on the basis of the current value and the predicted value.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a washing system for washing an imaging device provided in a vehicle. [Background technology]

[0002] In recent years, efforts to provide access to sustainable transportation systems that take into consideration vulnerable traffic participants have been gaining momentum. To achieve this, efforts are being made to further improve traffic safety and convenience through research and development of preventive safety technologies. For example, Patent Document 1 discloses a lens cleaning device that cleans the lenses of cameras (imaging devices) that capture images of the area around a vehicle. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-27539 Summary of the Invention [Problem to be solved by the invention]

[0004] After cleaning the image capture device, some of the cleaning fluid or dirt that could not be removed by cleaning may remain on the image capture device. In this case, depending on the type of dirt that remains, the recognition performance of the image capture device after cleaning may actually be reduced compared to the recognition performance of the image capture device before cleaning.

[0005] In view of the above background, an object of the present invention is to provide a cleaning system for an imaging device that suppresses a decrease in the recognition performance of the imaging device after cleaning compared to the recognition performance of the imaging device before cleaning, thereby contributing to the development of a sustainable transportation system. [Means for solving the problem]

[0006] In order to solve the above problem, one aspect of the present invention is a washing system (21) for washing an imaging device (12A to 12C) provided in a vehicle (1), comprising a washing device (22) that washes the imaging device, and a control device (20) that controls the washing of the imaging device by the washing device, wherein the control device acquires an image (24) from the imaging device, calculates a current value by converting the current state of dirt (26) of the imaging device into a numerical value based on the image, calculates a predicted value by converting the future state of dirt of the imaging device after the washing device is operated into a numerical value, and determines whether to operate the washing device based on the current value and the predicted value.

[0007] According to this aspect, when it is expected that the recognition performance of the imaging device after cleaning will be improved compared to the recognition performance of the imaging device before cleaning, the cleaning device can be operated. On the other hand, when it is expected that the recognition performance of the imaging device after cleaning will be lower than the recognition performance of the imaging device before cleaning, the operation of the cleaning device can be suppressed. Therefore, it is possible to suppress a decrease in the recognition performance of the imaging device after cleaning relative to the recognition performance of the imaging device before cleaning.

[0008] In the above aspect, the control device may divide the captured image into multiple regions, set weight values ​​corresponding to each of the multiple regions, calculate the current value by summing the weight values ​​corresponding to the regions among the multiple regions where it is determined that the current dirt exists, calculate the predicted value by summing the weight values ​​corresponding to the regions among the multiple regions where it is predicted that the future dirt will occur, and determine to operate the cleaning device if the current value is greater than the predicted value.

[0009] According to this aspect, by setting weight values ​​according to the importance of a plurality of regions that make up the captured image, it is possible to appropriately determine whether or not to operate the cleaning device.

[0010] In the above aspect, the multiple regions may be arranged in three or more columns vertically and three or more columns horizontally, and the control device may set the largest weight value for the central region among the multiple regions.

[0011] According to this aspect, by setting the largest weight value to the central area having the highest importance, it is possible to more appropriately determine whether or not to operate the cleaning device.

[0012] In the above aspect, the weight value used to calculate the current value and the weight value used to calculate the predicted value may be the same value.

[0013] According to this aspect, the current value and the predicted value can be easily calculated.

[0014] In the above aspect, the control device may be configured to be able to execute a plurality of driving assistance functions, and may change the weight values ​​corresponding to the plurality of regions according to the driving assistance function being executed.

[0015] According to this aspect, it is possible to more appropriately determine whether or not to operate the washing device depending on the driving assistance function being executed.

[0016] In the above aspect, the control device may set a first weight value which is the weight value when the driving assistance function is being executed, and a second weight value which is the weight value when the driving assistance function is not being executed, and may set the second weight value to the same value for all of the areas, and may set the second weight value to be smaller than the lowest value of the first weight value.

[0017] According to this aspect, the weight value can be set to an appropriate value depending on whether the driving assistance function is being executed, thereby making it possible to more appropriately determine whether to operate the cleaning device.

[0018] In the above aspect, the control device may change the weight value corresponding to each of the plurality of regions depending on the position of the imaging device.

[0019] According to this aspect, it is possible to more appropriately determine whether or not to operate the cleaning device depending on the position of the imaging device.

[0020] In the above aspect, the plurality of regions may be arranged in three or more vertical rows, and the control device may predict that the future stain will occur in the region in the bottommost row of the plurality of regions.

[0021] Since cleaning fluid typically flows downward, it is likely to remain at the bottom of the imaging device. According to the above aspect, it is possible to accurately predict the area where future contamination will occur, taking this tendency into consideration.

[0022] In the above aspect, the control device may store a predicted value map showing the distribution of the areas where future dirt is predicted to occur, calculate the predicted value based on the predicted value map, and update the predicted value map based on the captured image after the cleaning device is operated.

[0023] According to this aspect, the actual state of the imaging device after cleaning is used to calculate the predicted value, so that the predicted value can be calculated with high accuracy.

[0024] In the above aspect, the control device may update the predicted value map based on the captured image taken after a predetermined time has elapsed since the cleaning device was started.

[0025] Generally, cleaning fluid is removed from the imaging device over time. In accordance with the above aspect, taking into account this tendency, the actual state of the imaging device after a predetermined time has elapsed since cleaning is performed is used to calculate the predicted value, thereby enabling more accurate calculation of the predicted value. [Effects of the Invention]

[0026] According to the above configuration, in the cleaning system for an imaging device, it is possible to suppress a decrease in the recognition performance of the imaging device after cleaning relative to the recognition performance of the imaging device before cleaning. [Brief explanation of the drawings]

[0027] [Figure 1] FIG. 1 is a plan view of a vehicle according to a first embodiment; [Figure 2] FIG. 1 is a block diagram showing a vehicle to which a washing system according to a first embodiment is applied; [Figure 3] 1 is a flowchart showing cleaning control in a cleaning system according to a first embodiment; [Figure 4] FIG. 1 is a diagram showing an example of a state in which a captured image according to the first embodiment is divided into a plurality of regions; [Figure 5] FIG. 1A is a diagram showing an example of a first weight value set for each region when driving assist control (LKAS) according to the first embodiment is being executed; FIG. 1B is a diagram showing an example of a first weight value set for each region when driving assist control (ACC) according to the first embodiment is being executed; and FIG. 1C is a diagram showing an example of a second weight value set for each region when driving assist control according to the first embodiment is not being executed. [Figure 6] FIG. 1 is a diagram showing an example of a predicted value map according to the first embodiment; [Figure 7] FIG. 1A shows an example of a first weight value set for each area of ​​a second imaging device provided on the left side of a vehicle according to the first embodiment, and FIG. 1B shows an example of a first weight value set for each area of ​​a second imaging device provided on the right side of a vehicle according to the first embodiment. [Figure 8] 10 is a flowchart showing cleaning control in a cleaning system according to a second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0028] (First embodiment) Hereinafter, an embodiment of a vehicle 1 to which a washing system 21 for imaging devices 12A to 12C according to the present invention is applied will be described with reference to the drawings.

[0029] As shown in Fig. 1, the vehicle 1 is, for example, a four-wheeled automobile. The vehicle 1 has a body 2 forming a framework, and a plurality of wheels 3 supported by the body 2. A plurality of entrance / exit doors 5 for passengers to enter and exit the vehicle 1 are provided on each of the left and right sides of the body 2.

[0030] 2, the vehicle 1 includes a plurality of imaging devices 12A to 12C, a drive unit 14, a braking unit 16, a steering unit 18, a control unit 20, and a cleaning unit 22. The control unit 20 and the cleaning unit 22 configure a cleaning system 21 for cleaning the imaging devices 12A to 12C provided on the vehicle 1.

[0031] Each of the imaging devices 12A to 12C is a device that detects the state of the surroundings of the vehicle 1. Each of the imaging devices 12A to 12C captures images of targets (neighboring vehicles such as a vehicle in front, pedestrians, structures on the road, lane markings, etc.) present around the vehicle 1. In this embodiment, the multiple imaging devices 12A to 12C include a first imaging device 12A (see FIG. 1) provided at the front end of the vehicle body 2, a pair of second imaging devices 12B (see FIG. 1) provided at the left and right sides of the vehicle body 2, and a third imaging device 12C (see FIG. 1) provided at the rear end of the vehicle body 2. The first imaging device 12A captures images of targets present in front of the vehicle 1. Each of the second imaging devices 12B captures images of targets present in front of the vehicle 1 and to the left and right outside. The third imaging device 12C captures images of targets present behind the vehicle 1. Hereinafter, images of targets captured by each of the imaging devices 12A to 12C will be referred to as "captured images."

[0032] The drive device 14 is a device that applies a drive force to the vehicle 1. The drive device 14 includes a drive source that generates a drive force for running the vehicle 1. The drive source is configured by, for example, an internal combustion engine and / or an electric motor.

[0033] The braking device 16 is a device that applies braking force to the vehicle 1. The braking device 16 includes, for example, a brake caliper that presses a pad against a brake rotor, and an electric cylinder that supplies hydraulic pressure to the brake caliper.

[0034] The steering device 18 is a device that changes the steering angle of the wheels 3 by steering the wheels 3. The steering device 18 includes, for example, a rack-and-pinion mechanism connected to the wheels 3 and an electric motor that drives the rack-and-pinion mechanism.

[0035] The control device 20 is an electronic control unit (ECU) consisting of a computer configured to execute various processes. The control device 20 includes an arithmetic processing unit (a processor such as a CPU or MPU) and a storage device (memory such as a ROM or RAM). The arithmetic processing unit reads necessary software from the storage device and executes predetermined arithmetic processing in accordance with the read software. The control device 20 may be configured as a single piece of hardware, or as a unit consisting of multiple pieces of hardware. The control device 20 is connected to each component of the vehicle 1 via a communication network such as a CAN (Controller Area Network), and controls each component of the vehicle 1.

[0036] The control device 20 acquires captured images from each of the imaging devices 12A to 12C. The control device 20 is configured to be able to execute a plurality of driving assistance functions based on the captured images from each of the imaging devices 12A to 12C. For example, the plurality of driving assistance functions are included in the advanced driver assistance control (ADAS: Advanced Driver Assistance Systems) of the vehicle 1. The advanced driver assistance control is a control equivalent to SAE autonomous driving levels 1 to 2. When the advanced driver assistance control is executed, the driver is the main driver of the vehicle 1 and has the authority to drive the vehicle 1.

[0037] The control device 20 is configured to be able to execute a lane keeping assistance system (LKAS) as one of a plurality of driving assistance functions. When the LKAS is executed, the control device 20 controls the steering device 18 to assist the driver in steering the vehicle 1 so that the vehicle 1 maintains its position within the lane.

[0038] The control device 20 is also configured to be able to execute adaptive cruise control (ACC) as another of the driving assistance functions. When the ACC is executed, the control device 20 controls the drive device 14 and the brake device 16 so that the vehicle 1 follows the vehicle ahead while maintaining a predetermined distance therebetween.

[0039] The cleaning device 22 includes, for example, a tank (not shown) that stores cleaning liquid, a plurality of nozzles 23A to 23C (see FIG. 1) connected to the tank, and a pump (not shown) connected to each of the nozzles 23A to 23C. Each of the nozzles 23A to 23C is provided to spray cleaning liquid toward the corresponding imaging devices 12A to 12C. Each of the imaging devices 12A to 12C is cleaned by spraying the cleaning liquid.

[0040] The control device 20 is configured to be able to control the cleaning of each of the imaging devices 12A to 12C by the cleaning device 22. In other words, the control device 20 operates the cleaning device 22, thereby cleaning each of the imaging devices 12A to 12C. The control device 20 may selectively operate pumps connected to each of the nozzles 23A to 23C of the cleaning device 22, thereby cleaning predetermined imaging devices 12A to 12C.

[0041] An example of cleaning control in the cleaning system 21 configured as above will be described. The cleaning control is control for cleaning each of the imaging devices 12A to 12C by the cleaning device 22. The cleaning control for the first imaging device 12A will be described below as an example.

[0042] As shown in FIG. 3, first, the control device 20 acquires the captured image 24 (see FIG. 4) from the first imaging device 12A (ST1).

[0043] Next, the control device 20 divides the acquired captured image 24 into a plurality of regions (ST2). As shown in FIG. 4, in this embodiment, the control device 20 divides the acquired captured image 24 into three vertical columns and three horizontal columns. As shown in FIG. 5, the regions arranged side by side in the top row are referred to as first to third regions, the regions arranged side by side in the center row are referred to as fourth to sixth regions, and the regions arranged side by side in the bottom row are referred to as seventh to ninth regions. Of the regions arranged side by side in the top row, the left region is referred to as the first region, the center region is referred to as the second region, and the right region is referred to as the third region. Of the regions arranged side by side in the center row, the left region is referred to as the fourth region, the center region is referred to as the fifth region, and the right region is referred to as the sixth region. Of the regions arranged side by side in the bottom row, the left region is referred to as the seventh region, the center region is referred to as the eighth region, and the right region is referred to as the ninth region. In FIG. 5, numbers indicating the regions are circled.

[0044] 3, next, the control device 20 determines whether or not a driving assistance function is being executed (ST3). If a driving assistance function is being executed (ST3: Yes), the control device 20 sets a first weight value for each region according to the driving assistance function being executed (ST4).

[0045] For example, as shown in FIG. 5A, during execution of LKAS, the control device 20 sets the first weight values ​​set for the regions arranged side by side in the vertical center column to be larger than the first weight values ​​set for the regions arranged side by side in the vertical column. That is, the control device 20 sets the first weight values ​​set for the fourth to sixth regions to be larger than the first weight values ​​set for the first to third regions and the first weight values ​​set for the seventh to ninth regions. At this time, the control device 20 sets the largest first weight value for the fifth region (the central region) among the multiple regions. Furthermore, the control device 20 sets the first weight values ​​set for the fourth and sixth regions (regions adjacent to the fifth region on the left and right) to be second largest after the first weight value set for the fifth region.

[0046] For example, as shown in FIG. 5B, during ACC execution, the control device 20 sets the first weight values ​​set for the regions arranged vertically in the center columns on the left and right to be greater than the first weight values ​​set for the regions arranged vertically in the left and right columns. That is, the control device 20 sets the first weight values ​​set for the second, fifth, and eighth regions to be greater than the first weight values ​​set for the first, fourth, and seventh regions and the first weight values ​​set for the third, sixth, and ninth regions. At this time, the control device 20 sets the largest first weight value for the fifth region (the central region) among the multiple regions. Furthermore, the control device 20 sets the first weight values ​​set for the second and eighth regions (regions adjacent above and below the fifth region) to be second largest after the first weight value set for the fifth region.

[0047] As shown in Fig. 3, when the driving support function is not being executed (ST3: No), the control device 20 sets a second weight value to each region (ST5). The second weight value is a weight value when the driving support function is not being executed. At this time, as shown in Fig. 5(C), the control device 20 sets the second weight value to the same value for all regions and sets the second weight value to be smaller than the smallest value of the first weight value.

[0048] As shown in FIG. 3, next, the control device 20 calculates the current value and the predicted value based on the captured image 24 acquired from the first imaging device 12A and the first weight value or the second weight value (ST6).

[0049] The current value is a value obtained by converting the current state of the stain 26 (see FIG. 4) on the first imaging device 12A into a numerical value. The control device 20 calculates the current value by summing up the first weight values ​​or second weight values ​​corresponding to the areas determined to have the current stain 26 among the multiple areas.

[0050] For example, referring to Figures 4 and 5(A), when the current dirt 26 is included in the fifth and sixth regions during LKAS execution, the current value is 16 points, which is the sum of the first weight values ​​corresponding to the fifth and sixth regions. Referring to Figures 4 and 5(B), when the current dirt 26 is included in the fifth and sixth regions during ACC execution, the current value is 13 points, which is the sum of the first weight values ​​corresponding to the fifth and sixth regions. Referring to Figures 4 and 5(C), when the driving assistance function is not being executed, when the current dirt 26 is included in the fifth and sixth regions during LKAS execution, the current value is 2 points, which is the sum of the second weight values ​​corresponding to the fifth and sixth regions.

[0051] In another embodiment, the control device 20 may calculate, for each region, the ratio between the area of ​​the region and the area of ​​the current stain 26 contained in the region. Then, the control device 20 may calculate the current value by summing up the first weight values ​​corresponding to the regions among the multiple regions where the calculated ratio is equal to or greater than a predetermined threshold.

[0052] The predicted value is a value obtained by converting the future state of dirt on the first imaging device 12A after the cleaning device 22 is operated into a numerical value. With reference to FIG. 6, the control device 20 stores a predicted value map M showing the distribution of areas where future dirt is predicted to occur (hereinafter referred to as "dirt prediction areas"). In the example of FIG. 6, "X" is displayed in the seventh to ninth areas, which are dirt prediction areas. The control device 20 calculates the predicted value by summing the first weight value or the second weight value corresponding to the dirt prediction area. The first weight value and the second weight value used to calculate the current value are the same as the first weight value and the second weight value used to calculate the predicted value.

[0053] For example, with reference to Figures 6 and 5(A), when future contamination is predicted to occur in the 7th to 9th regions during LKAS execution (i.e., when the 7th to 9th regions are contamination prediction regions), the predicted value is 8 pt, which is the sum of the first weight values ​​corresponding to the 7th to 9th regions. With reference to Figures 6 and 5(B), when future contamination is predicted to occur in the 7th to 9th regions during ACC execution, the predicted value is 11 pt, which is the sum of the first weight values ​​corresponding to the 7th to 9th regions. With reference to Figures 6 and 5(C), when future contamination is predicted to occur in the 7th to 9th regions during non-execution of the driving assistance function, the predicted value is 3 pt, which is the sum of the second weight values ​​corresponding to the 7th to 9th regions.

[0054] As shown in FIG. 3, the control device 20 then determines whether to operate the cleaning device 22 based on the current value and the predicted value. Specifically, the control device 20 compares the current value calculated as described above with the predicted value (ST7). If the current value is greater than the predicted value (ST7: Yes), the control device 20 determines to operate the cleaning device 22 (ST8). That is, the cleaning device 22 cleans the first imaging device 12A. If the current value is equal to or less than the predicted value (ST7: No), the control device 20 determines not to operate the cleaning device 22 (ST9). That is, the cleaning device 22 does not clean the first imaging device 12A.

[0055] Next, the control device 20 determines whether a predetermined time has elapsed since the cleaning device 22 was operated (ST10). If the control device 20 determines that the predetermined time (for example, about 10 seconds) has elapsed (ST10: Yes), it updates the predicted value map M based on the image captured after the predetermined time has elapsed (ST11). Specifically, first, the control device 20 acquires an image captured by the first imaging device 12A after the predetermined time has elapsed. Then, the control device 20 identifies the predicted dirt area at the current time (when the predetermined time has elapsed) based on the acquired image captured by the first imaging device 12A. Furthermore, the control device 20 updates the predicted value map M by replacing the predicted value map M used in calculating the predicted value (ST6) with the predicted value map M in which the predicted dirt area at the current time has been recorded. This completes the cleaning control.

[0056] In the cleaning system 21 configured as described above, if the current value is greater than the predicted value, the recognition performance of the first image capturing device 12A after cleaning is expected to be improved compared to the recognition performance of the first image capturing device 12A before cleaning, and the control device 20 can operate the cleaning device 22. On the other hand, if the current value is equal to or less than the predicted value, the recognition performance of the first image capturing device 12A after cleaning is expected to be lower than the recognition performance of the first image capturing device 12A before cleaning, and the control device 20 can suppress the operation of the cleaning device 22. Therefore, it is possible to suppress a decrease in the recognition performance of the first image capturing device 12A after cleaning relative to the recognition performance of the first image capturing device 12A before cleaning.

[0057] The control device 20 divides the captured image 24 acquired from the first imaging device 12A into three columns vertically and three columns horizontally, and sets a first weight value corresponding to each of the nine regions. At this time, the control device 20 sets the largest first weight value to the fifth region (the central region) among the nine regions. The fifth region (the central region) is more important in detecting the situation around the vehicle 1 than the first to fourth regions and the sixth to ninth regions (regions other than the central region). Therefore, by setting the largest first weight value to the fifth region (the central region), which has the highest importance, it is possible to more appropriately determine whether to operate the washing device 22.

[0058] The first and second weighting values ​​used to calculate the predicted value are the same as the first and second weighting values ​​used to calculate the current value, which makes it possible to easily calculate the current value and the predicted value.

[0059] The control device 20 changes the first weight value corresponding to each of the nine areas depending on the driving assistance function being executed (see FIGS. 5(A) and 5(B)). The importance of each area changes depending on the driving assistance function being executed by the control device 20. By setting the first weight value based on the importance of each area, the control device 20 can more appropriately determine whether to operate the cleaning device 22 depending on the driving assistance function being executed.

[0060] When the driving assistance function is not being executed, the control device 20 sets a second weight value, which is a weight value when the driving assistance function is not being executed. At this time, the control device 20 sets the second weight value to the same value for all areas and sets the second weight value to be smaller than the smallest value of the first weight value. This makes it possible to set the first weight value and the second weight value to appropriate values ​​depending on whether the driving assistance function is being executed. Therefore, it is possible to more appropriately determine whether to operate the cleaning device 22.

[0061] The control device 20 updates the predicted value map M based on the image captured after a predetermined time has elapsed since the first imaging device 12A was cleaned. Due to aging or other reasons of the coating applied to the first imaging device 12A, cleaning liquid tends to remain on the first imaging device 12A. Furthermore, cleaning liquid remaining on the first imaging device 12A is removed from the first imaging device 12A over time. In consideration of these trends, the actual state of the first imaging device 12A after a predetermined time has elapsed since cleaning is used to calculate the predicted value, so that the predicted value can be calculated with high accuracy.

[0062] The control device 20 may change the first weight value corresponding to each of the multiple regions depending on the position of each of the image capturing devices 12A to 12C. For example, in the image captured by each of the second image capturing devices 12B provided on the left and right sides of the vehicle 1, the first weight value set for the outer left and right regions of the vehicle 1 may be larger than the first weight value set for the inner left and right regions of the vehicle 1. For example, as shown in FIG. 7A, in the image captured by the second image capturing device 12B provided on the left side of the vehicle 1, the first weight value set for the fourth region (the outer left and right regions of the vehicle 1) is larger than the first weight value set for the sixth region (the inner left and right regions of the vehicle 1). For example, as shown in FIG. 7B, in the image captured by the second image capturing device 12B provided on the right side of the vehicle 1, the first weight value set for the sixth region (the outer left and right regions of the vehicle 1) is larger than the first weight value set for the fourth region (the inner left and right regions of the vehicle 1). The importance of each region changes depending on the position of each of the image capturing devices 12A to 12C. By setting the first weight value in consideration of the importance of each area, the control device 20 can more appropriately determine whether or not to operate the cleaning device 22 depending on the position of each of the imaging devices 12A to 12C.

[0063] The first imaging device 12A, which is provided at the front end of the vehicle body 2, is more susceptible to wind from traveling than the third imaging device 12C, which is provided at the rear end of the vehicle body 2. Therefore, the distribution of dirt in the image captured by the first imaging device 12A is more susceptible to the influence of vehicle speed than the distribution of dirt in the image captured by the third imaging device 12C. In light of this, the control device 20 may change the first weight value in the image captured by the first imaging device 12A relative to the first weight value in the image captured by the third imaging device 12C.

[0064] (Second embodiment) Next, the wash control according to a second embodiment of the present invention will be described with reference to Fig. 8. The wash control according to the second embodiment differs from the wash control according to the first embodiment in the calculation method of the predicted value in ST26. The wash control according to the second embodiment also differs from the wash control according to the first embodiment in that the predicted value map M is not updated. Other than that, the wash control according to the second embodiment is the same as the wash control according to the first embodiment, and therefore description thereof will be omitted.

[0065] In ST26, the control device 20 predicts that dirt will occur in the seventh to ninth regions (the regions in the bottom row) of the nine regions in the future. Then, the control device 20 calculates a prediction value by summing up the weight values ​​corresponding to the seventh to ninth regions.

[0066] Since cleaning liquid normally flows downward, cleaning liquid tends to remain in the lowest part of the first imaging device 12A. Based on this tendency, by predicting that future contamination will occur in the seventh to ninth areas (the areas in the lowest row), it is possible to accurately predict the areas where future contamination will occur. Furthermore, since there is no need to update the prediction value map M (see the first embodiment), the cleaning control can be simplified.

[0067] Although the description of the specific embodiment has been completed above, the present invention is not limited to the above embodiment and can be widely modified and implemented. For example, in this embodiment, the control device 20 divides the captured image 24 into three columns vertically and three columns horizontally. In other embodiments, the control device 20 may divide the captured image 24 into two or four or more columns vertically or two or four or more columns horizontally. Furthermore, in other embodiments, the control device 20 may not divide the captured image 24. In the second embodiment, the predicted value is calculated based on the tendency of the cleaning liquid to flow downward, but the predicted value may also be calculated based on the degree of dirt, the type of dirt, weather, etc. Furthermore, the first embodiment and the second embodiment may be combined to update and calculate the predicted value as appropriate. [Explanation of symbols]

[0068] 1: Vehicle 12A to 12C: Imaging device 21: Cleaning system 22: Cleaning equipment 24: Captured image 26: Current stains

Claims

1. A cleaning system for cleaning an imaging device provided in a vehicle, comprising: a cleaning device for cleaning the imaging device; a control device that controls cleaning of the imaging device by the cleaning device, The control device Acquiring a captured image from the imaging device; Calculating a current value obtained by converting a current state of dirt on the imaging device into a numerical value based on the captured image; calculating a predicted value obtained by converting a future state of contamination of the imaging device after the cleaning device is operated into a numerical value; The cleaning system for an imaging device determines whether or not to operate the cleaning device based on the current value and the predicted value.

2. The control device Dividing the captured image into a plurality of regions; setting a weight value corresponding to each of the plurality of regions; calculating the current value by summing the weight values ​​corresponding to the areas determined to have the current stain among the plurality of areas; calculating the predicted value by summing the weight values ​​corresponding to the areas where the future contamination is predicted to occur among the plurality of areas; The cleaning system for an imaging device according to claim 1 , wherein the cleaning device is determined to be in operation when the current value is greater than the predicted value.

3. The plurality of regions are provided in three or more rows in the vertical direction and three or more rows in the horizontal direction, The cleaning system for an imaging device according to claim 2 , wherein the control device sets the largest weight value for a central region among the plurality of regions.

4. 3. The cleaning system for an imaging device according to claim 2, wherein the weight value used to calculate the current value and the weight value used to calculate the predicted value are the same value.

5. The control device A plurality of driving assistance functions are executable, The cleaning system for an imaging device according to claim 2 , wherein the weight values ​​corresponding to the plurality of regions are changed according to the driving assistance function being executed.

6. The control device a first weight value that is the weight value when the driving assistance function is being executed, and a second weight value that is the weight value when the driving assistance function is not being executed; 6. The cleaning system for an imaging device according to claim 5, wherein the second weight value is set to the same value for all the regions and the second weight value is set to be smaller than the smallest value of the first weight values.

7. The cleaning system for an imaging device according to claim 2 , wherein the control device changes the weight value corresponding to each of the plurality of regions depending on the position of the imaging device.

8. The plurality of regions are arranged in three or more rows in the vertical direction, The cleaning system for an imaging device according to claim 2 , wherein the control device predicts that the future contamination will occur in the area in the lowest row among the plurality of areas.

9. The control device storing a predicted value map showing a distribution of the areas where future contamination is predicted to occur; calculating the predicted value based on the predicted value map; The system for cleaning an imaging device according to claim 2 , wherein the predicted value map is updated based on the captured image after the cleaning device is operated.

10. The cleaning system for an imaging device according to claim 9 , wherein the control device updates the predicted value map based on the captured image after a predetermined time has elapsed since the cleaning device was put into operation.

Images