JP2020154665A - Travel environment evaluation system - Google Patents

Abstract

To evaluate travel environment in automatic driving even when a vehicle traveling on a candidate area has no equivalent device for acquiring peripheral information necessary for traveling by automatic driving.SOLUTION: A travel environment evaluation system 100 for evaluating travel environment of an automatic driving vehicle includes: an environmental condition recording image database 130 for storing a plurality of environmental condition recording images in advance; a related-image database 140 storing a plurality of related-images in advance; an image receiving unit 110 receiving a candidate area image; a weather information receiving unit 120 for receiving weather information in a candidate area; an image processing unit 150 for processing the candidate area image using the amount of change in environmental conditions; a future environment estimating unit 160 for estimating an effective degree of information on surrounding environment acquired when the automatic driving vehicle travels the candidate area in the future; and an information transmitting unit 110 for transmitting the effective degree estimated by the future environment estimating unit to the automatic driving vehicle.SELECTED DRAWING: Figure 2

Description

The present disclosure relates to a driving environment evaluation system.

The driving environment evaluation system is a system that evaluates the ease of driving of a vehicle. In such a driving environment evaluation system, future poor visibility due to rainfall or snowfall is calculated and predicted based on an image taken by a camera provided in the preceding vehicle, and the vehicle runs according to the prediction result. There is something that evaluates ease (for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2016-181061

In such a driving environment evaluation system, the degree of poor visibility is used as a criterion for evaluating the driving environment, and the driving environment for driving a vehicle by manual driving by a person is evaluated. On the other hand, since a vehicle driven by autonomous driving travels using information on the surrounding environment acquired by millimeter waves or a laser in addition to a camera, a system for evaluating whether or not the driving environment can be recognized regarding the driving of the vehicle by autonomous driving Has not been fully considered so far. Further, if the preceding vehicle does not have an equivalent device for acquiring information on the surrounding environment necessary for driving by automatic driving, there is also a problem that it is not possible to evaluate whether or not the driving environment can be recognized in automatic driving. In order to solve such problems, even if the preceding vehicle does not have an equivalent device that acquires information on the surrounding environment necessary for driving by autonomous driving, whether or not the driving environment can be recognized in autonomous driving is determined. It is desired to build a system that can be evaluated.

According to one form of the present disclosure, a driving environment evaluation system is provided. This driving environment evaluation system is a driving environment evaluation system (100) that evaluates the driving environment of an automatically driving vehicle including a peripheral information acquisition unit that acquires information on the surrounding environment necessary for driving by automatic driving, and is photographed by a camera. An environmental condition recording image database (130) that stores in advance a plurality of environmental condition recorded images in which environmental conditions including weather information and illuminance at the time of shooting are recorded, and a camera at a point where information on the surrounding environment is acquired. A related image database (140) that stores a plurality of related images that have been photographed and associated with the validity of information about the surrounding environment in advance, and the automatic driving vehicle will run in the future between the present and a predetermined time in the past. An image receiving unit (110) for receiving a candidate area image captured in the candidate area, a weather information receiving unit (120) for receiving weather information in the candidate area, and a plurality of the cameras existing in the candidate area. Of the environmental condition recorded images of, the environmental condition recorded image in which the environmental condition closest to the current environmental condition in the candidate area is recorded, and the time required for the automatic driving vehicle to move from the current position to the candidate area. An image processing unit that processes the candidate region image using the environmental condition recorded image in which the environmental condition closest to the environmental condition in the candidate region predicted based on is recorded and the amount of change in the environmental condition extracted from the image processing unit ( The future environment estimation unit (150) is compared with the plurality of related images and the processed candidate region image to estimate the effectiveness acquired when the autonomous vehicle travels in the candidate region in the future. 160) and an information transmission unit (110) that transmits the effectiveness estimated by the future environment estimation unit to the automatic driving vehicle. According to the driving environment evaluation system of this form, the surrounding environment acquired when the autonomous driving vehicle travels in the candidate area in the future by comparing the processed candidate area image, which is a captured image, with a plurality of related images. You can estimate the validity of the information about. Therefore, even if the vehicle traveling in the candidate area does not have an equivalent device for acquiring information on the surrounding environment necessary for driving by automatic driving, it is possible to evaluate whether or not the driving environment can be recognized in automatic driving. ..

The present disclosure can also be realized in various forms other than the driving environment evaluation system. For example, it can be realized in the form of a driving environment evaluation device, a driving environment evaluation method, or the like.

It is explanatory drawing which shows the outline of the driving environment evaluation system of 1st Embodiment. It is explanatory drawing which shows the structure of the driving environment evaluation system, a vehicle and the like. It is explanatory drawing which shows the structure of the information gathering vehicle. It is a conceptual diagram which shows the processing flow of a candidate area image. This is a flow showing future environment estimation processing. It is explanatory drawing which shows the structure of the driving environment evaluation system, a vehicle and the like. This is a flow showing future environment estimation processing. It is explanatory drawing explaining the method of estimating the effectiveness in a wide area.

A. First Embodiment:
FIG. 1 shows an outline of the driving environment evaluation system 100 of the first embodiment. The vehicle 50 is a vehicle that travels by manual driving. "Manual driving" refers to operations for engine control (depressing the accelerator pedal), operations for brake control (depressing the brake pedal), and operations for steering control (rotation of the steering wheel). It means the operation performed by the passenger who boarded the 50. The autonomous driving vehicle 70 is a vehicle that travels by automatic driving. The "automatic driving" means a driving in which the passengers in the autonomous driving vehicle 70 automatically perform all of the engine control, the brake control, and the steering control without performing the driving operation. In order to drive by autonomous driving, it is necessary to acquire information on the surrounding environment, which will be described later. The autonomous driving vehicle 70 is a vehicle capable of switching between automatic driving and manual driving. The driver (driver) of the autonomous driving vehicle 70 can switch between automatic driving and manual driving by a predetermined switch provided on the instrument panel or the like.

In FIG. 1, the roads R shown in a grid pattern are roads on which the vehicle 50 and the self-driving vehicle 70 can travel. The vehicle 50 is traveling in the candidate region DP in which the autonomous driving vehicle 70 will travel in the future. The traveling environment evaluation system 100 receives a candidate region image which is a captured image taken in the candidate region DP by the camera 52 (shown in FIG. 2) provided in the vehicle 50. The driving environment evaluation system 100 estimates the validity of the information about the surrounding environment acquired when the autonomous driving vehicle 70 travels in the candidate area DP in the future based on the candidate area image, and determines the effectiveness of the information about the surrounding environment. Send to. The validity of the information regarding the surrounding environment referred to here is a numerical value of the degree of recognition of the surrounding environment by the autonomous driving vehicle 70, and is expressed in the range of 0 to 1. The closer the effectiveness is to 1, the higher the accuracy of recognition of the surrounding environment by the autonomous driving vehicle 70. In this embodiment, the effectiveness is defined with the F value as an index. In other embodiments, effectiveness may be defined by other indicators.

FIG. 2 shows a schematic configuration of the traveling environment evaluation system 100, the vehicle 50, and the autonomous driving vehicle 70. The vehicle 50 includes a camera 52 and a communication unit 54. The camera 52 captures the surroundings of the vehicle 50 to acquire an image. In this embodiment, a drive recorder attached to the vehicle 50 is used as the camera 52. In another embodiment, as the camera 52, a smartphone of a passenger boarding the vehicle 50 may be attached to the vehicle 50, or a so-called automatic method that detects an obstacle in front of the vehicle 50 to reduce the damage of a collision. A camera provided in the vehicle 50 may be used for braking. The communication unit 54 sends the captured image captured by the camera 52 to the traveling environment evaluation system 100. At this time, information indicating the time at the time of photographing and information indicating the specifications of the camera 52 are added to the captured image transmitted via the communication unit 54. When the camera 52 is a camera capable of recording the environmental conditions including the weather information and the illuminance at the time of shooting in the captured image, the information indicating the environmental conditions is added to the captured image transmitted via the communication unit 54. May be good. The information indicating the specifications of the camera 52 referred to here may include at least one of a lens type, a horizontal angle of view, a minimum subject illuminance, and a focal length, or may include a model number of the camera.

The autonomous driving vehicle 70 includes a peripheral information acquisition unit 72, a communication unit 74, and a recognition unit 76. The communication unit 74 receives a signal indicating the effectiveness estimated by the traveling environment evaluation system 100. The peripheral information acquisition unit 72 acquires information on the surrounding environment necessary for traveling by automatic driving. The peripheral information acquisition unit 72 includes a camera, a millimeter wave radar, and LiDAR. The information about the surrounding environment refers to the information acquired by the peripheral information acquisition unit 72. The recognition unit 76 determines the validity of the information acquired by the peripheral information acquisition unit 72. As described above, the effectiveness here is expressed in the range of 0 to 1, and the closer it is to 1, the higher the accuracy of recognition of the surrounding environment by the autonomous driving vehicle 70.

The performance of the camera 52 included in the vehicle 50 is lower than the performance of the camera included in the peripheral information acquisition unit 72. Further, the vehicle 50 does not include the millimeter wave radar and LiDAR included in the peripheral information acquisition unit 72. Therefore, the vehicle 50 does not have a device equivalent to the peripheral information acquisition unit 72 for acquiring information on the surrounding environment necessary for traveling by automatic driving.

The driving environment evaluation system 100 includes a communication unit 110, a weather information receiving unit 120, an environmental condition recording image database 130, a related image database 140, an image processing unit 150, and a future environment estimation unit 160. The communication unit 110 functions as an image receiving unit that receives the candidate region image sent from the vehicle 50, and also transmits the validity of the information about the surrounding environment estimated by the future environment estimation unit 160, which will be described later, to the autonomous driving vehicle 70. Functions as a transmitter.

The weather information receiving unit 120 receives the weather information. The meteorological information referred to here includes information such as precipitation and date and time. More specifically, the weather information receiving unit 120 acquires the weather information from the server that provides the weather information via the network.

The information collecting vehicle 300 shown in FIG. 3 is a vehicle that collects information for constructing the environmental condition recording image database 130 and the related image database 140. The information collecting vehicle 300 includes a camera 352, a peripheral information acquisition unit 372, a communication unit 374, and a recognition unit 376.

The camera 352 is a camera having the same performance as the camera 52 provided in the vehicle 50. The camera 352 is preferably the same product as the camera 52 provided in the vehicle 50, but may be any product as long as it has the same performance as the camera 52.

The peripheral information acquisition unit 372 has the same configuration as the peripheral information acquisition unit 72 provided in the autonomous driving vehicle 70. The peripheral information acquisition unit 372 preferably has the same configuration as the peripheral information acquisition unit 72 provided in the autonomous driving vehicle 70, but has an arbitrary configuration as long as it has the same performance as the peripheral information acquisition unit 72. May be good. The recognition unit 376 has the same configuration as the recognition unit 76 provided in the autonomous driving vehicle 70. The recognition unit 376 preferably has the same configuration as the recognition unit 76 provided in the autonomous driving vehicle 70, but may have any configuration as long as it has the same performance as the recognition unit 76.

The communication unit 374 records the validity of the captured image acquired by the camera 352, the information on the surrounding environment acquired by the peripheral information acquisition unit 372, and the information on the peripheral environment determined by the recognition unit 76 in the environmental condition recording image database. It is sent to a system having a function of constructing 130 and constructing a related image database 140. Such a system may be included in the driving environment evaluation system 100, or a construction system different from the driving environment evaluation system 100 may exist independently. When using different construction systems, after the environmental condition recording image database 130 and the related image database 140 are constructed, the environmental condition recording image database 130 and the related image database are applied to the driving environment evaluation system 100 from the construction system. 140 is sent.

The information collecting vehicle 300 collects information by traveling in various areas under various weather conditions. In traveling for information collection, the information collecting vehicle 300 captures the surroundings of the information collecting vehicle 300 by the camera 352 and acquires a photographed image, and the peripheral information acquisition unit 372 relates to the surrounding environment necessary for traveling by automatic driving. Get information. Further, the validity of the acquired information is determined by the recognition unit 376.

The environmental condition recording image database 130 is constructed by using the captured image acquired by the camera 352. By associating the photographed image with the weather information at the time of shooting and the environmental conditions including the illuminance, an environmental condition recorded image which is a photographed image in which the weather information at the time of shooting and the environmental conditions including the illuminance are recorded is created. The environmental condition recording image database 130 is constructed by storing the plurality of environmental condition recording images created in this manner in advance.

The environmental condition recording image database 130 constructed by the above steps is a plurality of environmental condition recording images that are captured images taken by the camera 352 and are labeled with environmental conditions including weather information and illuminance at the time of shooting. Is stored in advance.

The related image database 140 is constructed by using the validity of the captured image acquired by the camera 352, the information on the surrounding environment acquired by the peripheral information acquisition unit 372, and the information on the peripheral environment determined by the recognition unit 76. .. Since the information collecting vehicle 300 includes the camera 352, the peripheral information acquisition unit 372, and the recognition unit 376, the camera 352, the peripheral information acquisition unit 372, and the recognition unit 376 are traveling when the information collection vehicle 300 is traveling at a certain point. Information on the surrounding environment of the captured image and the validity of the information can be acquired at the same time. The features of the image (for example, edges of objects, running paths, signs, signals, etc.) are extracted from the captured image acquired at this time, and the extraction result is associated with the validity. As a pattern recognition model used at this time, a support vector machine (SVM) can be mentioned. Shooting associated with the validity of information about the surrounding environment by associating the validity of the information about the surrounding environment acquired at the same point with the captured images simultaneously acquired at the point where the information about the surrounding environment was acquired. A related image, which is an image, is created. The related image database 140 is constructed by storing the plurality of related images created in this manner in advance.

The related image database 140 constructed by the above steps is a plurality of captured images taken by the camera 352 at the point where the information about the surrounding environment is acquired and which are the captured images associated with the validity of the information about the surrounding environment. Store related images in advance.

Returning to the description of FIG. 2, the image processing unit 150 processes the candidate region image received by the communication unit 110. The image processing unit 150 moves the image PG in which the environmental condition closest to the current environmental condition in the candidate area DP among the plurality of environmental condition recorded images is labeled, and the autonomous driving vehicle 70 from the current position to the candidate area DP. Candidates are used by using the image FG labeled with the environmental conditions closest to the environmental conditions predicted based on the weather information in the candidate region DP after the time required for the elapse, and the amount of change in the environmental conditions extracted from the image FG. Process the area image. FIG. 4 shows a processing flow of the candidate region image. For example, when the amount of change in environmental conditions extracted from the image PG and the image FG indicates an increase in precipitation and a decrease in illuminance, the candidate region image is processed to indicate an increase in precipitation and a decrease in illuminance. Will be done. The image PG and the image FG do not have to be the captured images obtained by capturing the candidate region DP, and may be captured images obtained by capturing another region. The image PG and the image FG are selected based on the current environmental conditions in the candidate region DP and the environmental conditions predicted in the future in the candidate region DP. It is preferable that the image PG and the image FG are, of course, captured images obtained by capturing the candidate region DP, and are conditions close to the current environmental conditions in the candidate region DP and the environmental conditions predicted in the future in the candidate region DP. ..

The criteria for selecting the environmental condition recorded image used by the image processing unit 150 to extract the amount of change in the environmental condition includes information indicating the specifications of the camera 52 added to the captured image transmitted via the communication unit 54. Is done. If the environmental condition recorded image stored in the environmental condition recorded image database 130 contains an environmental condition recorded image taken by a camera having the same specifications as the camera 52, the environmental condition recorded image is used as the environmental condition recorded image. Select the image used to extract the amount of change. If the environmental condition recorded image stored in the environmental condition recorded image database 130 does not contain an environmental condition recorded image taken by a camera having the same specifications as the camera 52, the specifications closest to the specifications of the camera 52 are used. An image to be used for extracting the amount of change in the environmental conditions is selected from the environmental condition recorded images taken by the camera.

The future environment estimation unit 160 compares the processed candidate area image with a plurality of related images, and estimates the validity of the information regarding the surrounding environment acquired when the autonomous driving vehicle 70 travels in the candidate area DP in the future. .. When this comparison is made, the features of the image are extracted from the processed candidate region image and compared with a plurality of related images. Both the processed candidate region image and the related image are photographed images. The future environment estimation unit 160 determines the validity of the information about the surrounding environment associated with the selected related image by selecting the related image of the environmental condition closest to the processed candidate area image, and the autonomous driving vehicle 70 determines the validity. It is used for estimation as the validity of the information about the surrounding environment acquired when the candidate area DP is driven in the future. Here, the environmental condition that serves as a reference when the related image is selected is a condition related to an element that affects the accuracy of recognition by the autonomous driving vehicle 70, and is, for example, rain, fog, or dust in the captured image. It refers to the turbidity and illuminance of the atmosphere due to such factors. The communication unit 110 transmits the validity of the information regarding the surrounding environment estimated by the future environment estimation unit 160 to the autonomous driving vehicle 70.

The driving environment evaluation system 100 executes the future environment estimation process described in FIG. The future environment estimation process is executed when a request is received from the autonomous driving vehicle 70. When the future environment estimation process is started, in step S110, the communication unit 110 acquires a candidate region image sent from the vehicle 50. At this time, the candidate area image acquired by the communication unit 110 is taken by the camera 52 provided in the vehicle 50 from the present to the predetermined time in the past when there is a request from the autonomous driving vehicle 70. The past predetermined time referred to here may be arbitrarily set, but the shorter the time width, the higher the accuracy of the future environment estimation processing. The camera 52 provided in the vehicle 50 acquires a candidate area image in response to a request from the communication unit 110 after the future environment estimation process is started, and transmits the candidate area image to the communication unit 110 via the communication unit 54. You may send it. In step S120, the image processing unit 150 selects the image PG and the image FG, which are the environmental condition recording images used by the image processing unit 150 to extract the amount of change in the environmental conditions.

In step S130, the image processing unit 150 processes the candidate region image using the amount of change in the environmental conditions extracted from the image PG and the image FG. In step S140, the future environment estimation unit 160 compares the processed candidate area image with a plurality of related images, and the information regarding the surrounding environment acquired when the autonomous driving vehicle 70 travels in the candidate area DP in the future is valid. The degree is estimated.

In step S150, the communication unit 110 transmits the validity of the information regarding the surrounding environment estimated by the future environment estimation unit 160 to the autonomous driving vehicle 70. After that, the future environment estimation process is completed.

According to the first embodiment described above, by comparing the processed candidate region image with a plurality of related images, the valid information on the surrounding environment acquired when the autonomous driving vehicle 70 travels in the candidate region DP in the future. The degree can be estimated. Therefore, even if the vehicle 50 traveling in the candidate region DP does not have an equivalent device for acquiring information on the surrounding environment necessary for traveling by automatic driving, whether or not the driving environment can be recognized in automatic driving can be recognized. Can be evaluated.

In addition, the traveling environment evaluation system 100 acquires captured images from each vehicle that is ahead in the direction in which the autonomous driving vehicle 70 is heading, such as the vehicle 50, and each vehicle travels using the captured images. It is possible to estimate the validity of the information regarding the surrounding environment acquired when the autonomous driving vehicle 70 travels in the area in the future. Therefore, the autonomous driving vehicle 70, which has received the estimated effectiveness for the area in which each vehicle is traveling from the traveling environment evaluation system 100, may set a route passing through the highly effective area as a planned traveling route. it can.

B. Second embodiment:
The driving environment evaluation system 100a shown in FIG. 6 includes a current environment estimation unit 170 and an effectiveness correction unit 180 instead of the environmental condition recording image database 130 and the image processing unit 150, and the future environment estimation unit of the first embodiment. The device configuration is the same as that of the driving environment evaluation system 100a of the first embodiment, except that the future environment estimation unit 160a different from the 160 is provided. The same reference numerals as those in the first embodiment indicate the same configuration, and the preceding description is referred to.

The current environment estimation unit 170 compares the candidate area image received by the communication unit 110 with a plurality of related images, and is valid for information on the surrounding environment acquired when the autonomous driving vehicle 70 is currently traveling in the candidate area DP. Estimate the degree.

The effectiveness correction unit 180 predicts based on the current environmental conditions in the candidate area DP and the candidate area DP after the time required for the autonomous driving vehicle 70 to move from the current position to the candidate area DP has elapsed. The effectiveness estimated by the current environment estimation unit 170 is corrected by using the environmental conditions to be obtained and the amount of change in the environmental conditions extracted from.

The future environment estimation unit 160a regards the validity corrected by the validity correction unit 180 as the validity of the information on the surrounding environment acquired when the autonomous driving vehicle 70 travels in the candidate area DP in the future, and the autonomous driving vehicle Estimate the validity of the information about the surrounding environment that 70 acquires when traveling in the candidate area DP in the future. The communication unit 110 transmits the validity of the information regarding the surrounding environment estimated by the future environment estimation unit 160a to the autonomous driving vehicle 70.

The driving environment evaluation system 100a executes the future environment estimation process described in FIG. 7. The future environment estimation process is executed when a request is received from the autonomous driving vehicle 70. When the future environment estimation process is started, in step S210, the communication unit 110 acquires a candidate region image sent from the vehicle 50. In step S220, the current environment estimation unit 170 estimates the validity of the information regarding the surrounding environment acquired when the autonomous driving vehicle 70 currently travels in the candidate area DP.

In step S230, the validity estimated in step S220 by the validity correction unit 180 using the change amount of the current environmental condition in the candidate region DP and the environmental condition extracted from the environmental condition predicted in the future in the candidate region DP. The degree is corrected. In step S240, the future environment estimation unit 160a estimates the validity of the information about the surrounding environment acquired when the autonomous driving vehicle 70 travels in the candidate area DP in the future by using the validity corrected in step S230. ..

In step S250, the communication unit 110 transmits the validity of the information regarding the surrounding environment estimated by the future environment estimation unit 160a to the autonomous driving vehicle 70. After that, the future environment estimation process is completed.

According to the second embodiment described above, the self-driving vehicle 70 acquires the candidate region DP when it travels in the future by correcting the validity estimated by comparing the candidate region image with the plurality of related images. The validity of information about the surrounding environment can be estimated. Therefore, even if the vehicle 50 traveling in the candidate region DP does not have an equivalent device for acquiring information on the surrounding environment necessary for traveling by automatic driving, whether or not the driving environment can be recognized in automatic driving can be recognized. Can be evaluated.

C. Other embodiments:
In the description of the first embodiment, the information collecting vehicle 300 has been used for collecting information for constructing the environmental condition recording image database 130 and the related image database 140, but the present disclosure is not limited to this. For example, the information gathering for constructing the environmental condition recording image database 130 and the related image database 140 may be performed by photographing in the artificial weather room. In such a form, in a state where the camera 352 and the peripheral information acquisition unit 372 are arranged in the artificial weather room, the environmental condition recording image database 130 and related by changing the environmental conditions including the weather information and the illuminance to various conditions Information gathering for constructing the image database 140 is executed. In the artificial weather room, it is preferable that an object to be a target for extracting image features from the information about the surrounding environment acquired by the peripheral information acquisition unit 372 is arranged.

In the description of the first embodiment, the captured image taken in the candidate region DP was taken by the camera 52 provided in the vehicle 50 leading in the direction in which the autonomous driving vehicle 70 is heading, but the present disclosure is limited to this. I can't. The captured image captured in the candidate region DP may be captured by a surveillance camera arranged in the candidate region DP.

In the first embodiment, the traveling environment evaluation system 100 acquires an image taken from each vehicle leading in the direction in which the autonomous driving vehicle 70 is heading, and each vehicle is traveling using the captured image. Although it has been explained that the validity of the information regarding the surrounding environment acquired when the autonomous driving vehicle 70 travels in the area in the future can be estimated, the present disclosure is not limited to this. For example, the traveling environment evaluation system 100 may estimate the effectiveness in a wider area by using the validity of the information about the surrounding environment estimated in the area in which each vehicle is traveling.

A method of estimating the effectiveness in a wider area will be described with reference to FIG. FIG. 8 shows the respective vehicles 50a to 50d leading in the direction in which the autonomous driving vehicle 70 is heading. The vehicles 50a to 50d include a camera and a communication unit as in the vehicle 50 described in the first embodiment. The vehicles 50a to 50d are traveling in the candidate regions DPa to DPd to which the autonomous driving vehicle 70 will travel in the future, respectively. The driving environment evaluation system 100 first acquires captured images taken by the cameras provided in the vehicles 50a to 50d in the candidate areas DPa to DPd, and automatically uses the captured images to automatically determine the region in which the vehicles 50a to 50d are traveling. The validity of the information about the surrounding environment acquired when the driving vehicle 70 travels in the future is estimated. At this time, the traveling environment evaluation system 100 creates a Voronoi diagram with the positions of the vehicles 50a to 50d as the mother points. The creation of the Voronoi diagram complements the validity of the information about the surrounding environment of the region at a position different from the candidate regions DPa to DPd. In such a Voronoi diagram, for example, when the estimated validity for the candidate region DPa is 0.9 and the estimated validity for the candidate region DPc is 0.5, the candidate region DPa and the candidate region DPc The effectiveness in the region between is distributed to decrease from 0.9 to 0.5 as the candidate region DPa goes to the candidate region DPc. The same applies to the effectiveness in the area between other candidate areas. By creating such a Voronoi diagram, the traveling environment evaluation system 100 may estimate the effectiveness in a region wider than the candidate regions DPa to DPd. In such an estimation method, the vehicle traveling in each of the candidate regions in which the autonomous driving vehicle 70 will travel in the future is not limited to the vehicle traveling by manual driving, and may be an autonomous driving vehicle traveling by automatic driving. ..

The present disclosure is not limited to the above-described embodiments and modifications, and can be realized with various configurations within a range not deviating from the gist thereof. For example, the embodiments corresponding to the technical features in each of the embodiments described in the column of the outline of the invention, the technical features in the modified examples are used to solve some or all of the above-mentioned problems, or the above-mentioned above. It is possible to replace or combine them as appropriate to achieve some or all of the effects. Further, if the technical feature is not described as essential in the present specification, it can be appropriately deleted.

100 ... Driving environment evaluation system, 110 ... Communication unit, 120 ... Weather information receiving unit, 130 ... Environmental condition recording image database, 140 ... Related image database, 150 ... Image processing unit, 160 ... Future environment estimation unit

Claims (2)

It is a driving environment evaluation system (100) that evaluates the driving environment of an autonomous driving vehicle equipped with a peripheral information acquisition unit that acquires information on the surrounding environment necessary for driving by autonomous driving.
An environmental condition recording image database (130) that stores a plurality of environmental condition recorded images in advance, which are photographed by a camera and record environmental conditions including weather information and illuminance at the time of shooting.
A related image database (140) that stores a plurality of related images taken by a camera at a point where information about the surrounding environment is acquired and associated with the validity of the information about the surrounding environment in advance.
An image receiving unit (110) that receives a candidate area image taken in the candidate area by a camera existing in the candidate area in which the autonomous driving vehicle will travel in the future from the present to a predetermined time in the past.
A weather information receiving unit (120) that receives weather information in the candidate area, and
Among the plurality of environmental condition recorded images, an environmental condition recorded image in which the environmental condition closest to the current environmental condition in the candidate area is recorded, and an image required for the autonomous driving vehicle to move from the current position to the candidate area. Image processing to process the candidate region image using the environmental condition recorded image in which the environmental condition closest to the environmental condition in the candidate region predicted based on time is recorded and the amount of change in the environmental condition extracted from the image. Part (150) and
A future environment estimation unit (160) that estimates the effectiveness acquired when the autonomous driving vehicle travels in the candidate region in the future by comparing the plurality of related images with the processed candidate region image.
A driving environment evaluation system including an information transmission unit (110) that transmits the effectiveness estimated by the future environment estimation unit to the autonomous driving vehicle. It is a driving environment evaluation system (100a) that evaluates the driving environment of an autonomous driving vehicle equipped with a peripheral information acquisition unit that acquires information on the surrounding environment necessary for driving by autonomous driving.
A related image database (140) that stores a plurality of related images taken by a camera at a point where information about the surrounding environment is acquired and associated with the validity of the information about the surrounding environment in advance.
An image receiving unit (110) that receives a candidate area image taken in the candidate area by a camera existing in the candidate area in which the autonomous driving vehicle will travel in the future from the present to a predetermined time in the past.
A weather information receiving unit (120) that receives weather information in the candidate area, and
The current environment estimation unit (170) that estimates the effectiveness acquired when the autonomous driving vehicle currently travels in the candidate region by comparing the plurality of related images with the candidate region image.
Changes in environmental conditions extracted from the current environmental conditions in the candidate area and the environmental conditions in the candidate area predicted based on the time required for the autonomous vehicle to move from the current position to the candidate area. An effectiveness correction unit (180) that corrects the estimated effectiveness using a quantity, and
The corrected effectiveness is regarded as the effectiveness acquired when the autonomous vehicle travels in the candidate area in the future, and the effectiveness acquired when the autonomous vehicle travels in the candidate region in the future. Future environment estimation unit (160a) that estimates
A driving environment evaluation system including an information transmission unit (110) that transmits the effectiveness estimated by the future environment estimation unit to the autonomous driving vehicle.

Images