JP2024030012A - Self-location estimation system - Google Patents

Abstract

An object of the present invention is to suppress deterioration in stability of self-position estimation of a vehicle. A self-position estimation system 1000 capable of communicating with a plurality of vehicles M includes an environment recognition section 210 that is installed in each of the plurality of vehicles and that acquires environmental information, and a map information acquisition section that acquires map information. a position estimating unit 250 that acquires the environmental information 240 and the detected information, and estimates the self-position of each of the plurality of vehicles using the environmental information, the map information, and the detected information; A state determination unit 111 that determines whether or not a determination target feature is in a passable state using the self-position estimated by each of the plurality of vehicles and acquired environmental information; and a status notification unit 113 that notifies a predetermined destination when the status determination unit determines that the status is impossible. [Selection diagram] Figure 1

Description

The present disclosure relates to a self-location estimation system.

Conventionally, information on road features acquired by information acquisition devices installed in multiple vehicles is collected, and changes in this information over a predetermined period of time are compared. A technique for detecting the appearance, change, removal, etc. of features such as signs is known (Patent Document 1). The information processing system described in Patent Document 1 suppresses a decrease in stability of vehicle position estimation by updating map information according to appearance, change, removal, etc. of features.

JP2019-164840A

However, even if a feature is deleted from the map information because it has been removed, in reality it is simply impassable due to a malfunction of the feature, and if the feature itself still exists. There is. In such a case, there may be a difference between the map information and the actual state of the feature, which may reduce the stability of vehicle position estimation.

The present disclosure can be realized as the following forms.

According to one aspect of the present disclosure, a self-location estimation system (1000) is provided. This self-position estimation system is a self-position estimation system capable of communicating with a plurality of vehicles (M), is installed in each of the plurality of vehicles, and acquires environmental information that is information about the environment around the vehicle. an environment recognition unit (210), a map information acquisition unit (240) that acquires map information, and a map information acquisition unit (240) that acquires the environmental information about the vehicle from the environment recognition units installed in each of the plurality of vehicles; Acquire detection information, which is information about the driving state and position of the vehicle, from sensors mounted on each of the plurality of vehicles, and use the environmental information, the map information, and the detection information, A position estimating unit (250) that estimates the self-position of each of the plurality of vehicles, and the self-position estimated in each of the plurality of vehicles and the acquired environmental information, and uses a state determination unit (111) that determines whether or not a determination target feature existing in the area is in a passable state; and a state determination unit (111) that determines whether the determination target feature existing in , and a status notification unit (113) that notifies a predetermined destination.

According to this form of self-position estimation system, environmental information and detection information about each of a plurality of vehicles are used to determine whether or not a determination target feature existing on the travel route of each vehicle is in a passable state. If the status determination unit determines that the target feature is impassable, it will notify a predetermined destination, giving an opportunity to repair the target feature and updating the map information. It is possible to suppress a decrease in stability of self-position estimation due to a difference between the state of the object and the state of the actual feature.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is an explanatory diagram showing a schematic configuration of a self-position estimation system according to a first embodiment. FIG. 1 is a block diagram showing a schematic configuration of a vehicle according to a first embodiment. FIG. 2 is a side view showing an imaging area of a camera mounted on a vehicle. FIG. 3 is a top view showing an imaging area of a camera mounted on a vehicle. 1 is a block diagram showing a schematic configuration of a server according to a first embodiment. FIG. It is a flowchart which shows the procedure of self-position estimation processing of a 1st embodiment. It is a flowchart which shows the procedure of map information update processing of a 1st embodiment. It is a flowchart which shows the procedure of map information update processing of a 2nd embodiment. It is a flowchart which shows the procedure from step S242A to step S248A in the map information update process of 2nd Embodiment.

A. First embodiment:
A-1. System configuration:
As shown in FIG. 1, the self-position estimation system 1000 of this embodiment includes a plurality of vehicles M and a server 100 that can communicate with the plurality of vehicles M via a network. The self-position estimation system 1000 uses information acquired from a plurality of vehicles M to determine whether features on the travel route of each vehicle M are passable or impassable. is determined, and the map information stored in the server 100 is updated according to the state of the feature. "A state in which a feature is impassable" means, for example, a state in which a railroad crossing barrier remains down due to a failure. Further, when the self-position estimation system 1000 determines that the feature is in an impassable state, it notifies the feature manager Ad of this fact via the network. For example, when a railroad crossing is in an impassable state, the "manager Ad" corresponds to the railway company that manages the railroad crossing. Note that the target of notification by the self-position estimation system 1000 is not limited to the administrator Ad, but may be a predetermined destination such as a local government or a police station in the area where the feature exists.

Each of the plurality of vehicles M can execute automatic driving, and is configured to be able to switch between automatic driving and manual driving. "Automatic driving" refers to driving in which engine control, brake control, and steering control are automatically executed in place of the driver. "Manual driving" refers to operations for engine control (depressing the accelerator pedal), brake control (depressing the brake pedal), and steering control (rotating the steering wheel). means the operation carried out by a person. Note that the vehicle M is not limited to a vehicle equipped with an engine, and may be an electric vehicle (EV) or a fuel cell vehicle (FCV).

As shown in FIG. 2, each of the plurality of vehicles M includes an environment recognition section 210, an own vehicle state quantity sensor section 220, a satellite positioning acquisition section 230, a map information acquisition section 240, a position estimation section 250, and a vehicle state sensor section 220. A control unit 260 is provided. In this embodiment, at least one of the environment recognition section 210, the own vehicle state quantity sensor section 220, the satellite positioning acquisition section 230, the map information acquisition section 240, the position estimation section 250, and the vehicle control section 260 The unit is a functional unit realized by a computer including a CPU, ROM, and RAM. An example of such a computer is an ECU (Electronic Control Unit).

The environment recognition unit 210 detects the positions and states of features around the vehicle M. In this embodiment, the vehicle M includes a camera 211 shown in FIGS. 3 and 4, a lidar device 212, and a millimeter wave radar 213 as the environment recognition unit 210. In the present embodiment, the camera 211 images a radial area in front of the vehicle M, which is indicated by hatching, and acquires image data. The Lidar device 212 detects the position and distance of a feature by emitting a laser beam and receiving a reflected wave reflected by the object. The millimeter wave radar 213 detects the position and distance of a feature by emitting millimeter waves and receiving reflected waves reflected by the feature. Note that the vehicle M only needs to include at least one of the Lidar device 212 and the millimeter wave radar 213. In the following description, the image data and the position and distance of a feature acquired by the environment recognition unit 210 will also be referred to as "environmental information." In other words, the environmental information can be said to be information regarding the environment around the vehicle M.

The own vehicle state quantity sensor unit 220 shown in FIG. 2 detects own vehicle state quantities representing the running state of the vehicle M when it is running, such as vehicle speed and yaw rate. In this embodiment, the vehicle M includes a vehicle speed sensor and a yaw rate sensor as the vehicle state quantity sensor section 220. Note that the own vehicle state quantity sensor section 220 is not limited to a vehicle speed sensor and a yaw rate sensor. The vehicle M may be equipped with a sensor that detects the acceleration, pitch angle, and roll angle of the vehicle M as the own vehicle state quantity sensor section 220.

The satellite positioning acquisition unit 230 detects the current position (longitude/latitude) of the vehicle M based on navigation signals received from artificial satellites forming a GNSS (Global Navigation Satellite System). The current position of the vehicle M acquired by the satellite positioning acquisition unit 230 and the vehicle state quantity detected by the vehicle state quantity sensor unit 220 described above correspond to “detection information” in the present disclosure.

The map information acquisition unit 240 acquires map information from the server 100. Map information includes not only static map information such as road width and number of lanes, but also dynamic information such as traffic congestion information. In addition, in the present embodiment, the map information includes information indicating whether or not the vehicle M can pass through a feature (hereinafter referred to as "traffic (also called “approval/disapproval information”).

The position estimating unit 250 estimates the self-position of the vehicle M on the map and the positions of features existing around the vehicle M using the above-mentioned environmental information, detection information, and map information. Further, the position estimating unit 250 transmits the estimated self-position and the acquired environmental information to the server 100.

The vehicle control unit 260 controls acceleration, deceleration, and steering of the vehicle M according to the self-position estimated by the position estimation unit 250 and the position of features existing around the vehicle M recognized by the environment recognition unit 210. Execute. In addition, if the passability information indicates that the planned feature is impassable, the vehicle control unit 260 switches the driving state from automatic driving to manual driving, changes the driving route, and allows the driver to Notify and carry out. For example, if the ETC gate at which the vehicle was scheduled to travel is closed, the vehicle control unit 260 changes the travel route so that the vehicle avoids the closed ETC gate and travels through an ETC gate that is passable. In addition, since all ETC gates are closed and the driver needs to pay the toll when driving through the general gate, the vehicle control unit 260 notifies the driver that the driver will be driving through the general gate. At the same time, the operating state is switched from automatic operation to manual operation. The vehicle control unit 260 corresponds to the “compatible processing unit” in the present disclosure. Note that the vehicle control unit 260 executes any one of switching the driving state from automatic driving to manual driving, changing the driving route, and notifying the driver depending on the state of the feature. It's okay.

As shown in FIG. 5, the server 100 is configured as a computer including a CPU 110, a storage device 120, a ROM 130, and a RAM 140. The CPU 110, the storage device 120, the ROM 130, and the RAM 140 are communicably connected to each other via a bus.

The CPU 110 functions as a status determining unit 111, a status updating unit 112, and a status notification unit 113 by loading a program stored in the ROM 130 into the RAM 140 and executing the program. The state determination unit 111 acquires environmental information and self-position from each vehicle M, and determines whether a feature existing on the travel route of each vehicle M is in a passable state. The status update unit 112 updates the map information according to the result of the determination by the status determination unit 111. When it is determined that the feature is in an impassable state, the status notification unit 113 notifies the manager Ad of the feature to that effect. Specific processing in each functional unit will be explained in the map information update processing described later.

The storage device 120 stores a map information database 121 that is a database that stores the above-mentioned map information. In this embodiment, the map information stored in the map information database 121 is the map information to be acquired by the above-mentioned map information acquisition unit 240. Hereinafter, the map information stored in the map information database 121 will also be referred to as "master map information."

A-2. Self-position estimation processing:
Each vehicle M repeatedly executes the self-position estimation process shown in FIG. 6 while driving, estimates its own position, and transmits the estimated self-position and the acquired environmental information to the server 100.

In each vehicle M, the environment recognition section 210 acquires environmental information (step S110), the own vehicle state quantity sensor section 220 and the satellite positioning acquisition section 230 acquire detection information (step S112), and the map information acquisition section 240 acquires detection information (step S112). Acquisition of map information (step S114) is executed in parallel.

In step S120, the position estimation unit 250 of each vehicle M estimates the own position of the own vehicle using the acquired environmental information, detection information, and map information.

In each vehicle M, the position estimating unit 250 transmits the own position and environmental information to the server 100 (step S130), and the vehicle control unit 260 controls the driving state of the own vehicle according to the own position and the environmental information. (Step S132). After step S130 and step S132 are completed, step S110, step S112, and step S114 are executed again in each vehicle M. As described above, steps S110 to S132 are repeatedly executed in each vehicle M.

A-3. Map information update processing:
The server 100 repeatedly executes the map information update process shown in FIG. 7 while the server 100 is in operation, and updates the master map information according to the determination result of the passability state of each feature existing on the travel route of each vehicle M. Update and notify the feature manager Ad.

In step S210, the state determination unit 111 acquires the self-position and environmental information transmitted from each vehicle M.

In step S220, the state determination unit 111 uses the acquired self-position and environmental information of each vehicle M to determine the features (hereinafter also referred to as "determination target features") that exist on the travel route of each vehicle M. Determine whether or not the road is passable. The state determining unit 111 determines that the target feature is in an impassable state when information indicating that the target feature is in an untraversable state is read from the self-position and environmental information of each vehicle M. do. Taking an ETC gate as an example, "information indicating that the target feature is in a state where it is impossible to drive" includes the fact that the opening/closing bar of the ETC gate is down, the presence of an "x" sign indicating road closure, etc. There is a red light on, a string of characters such as "Closed", etc. In addition, when it is read from the traveling trajectory information of each vehicle M that a plurality of vehicles M are traveling avoiding a specific ETC gate, the state determining unit 111 determines that the ETC gate is in an impassable state. judge. Note that the traveling trajectory information means information indicating the locus of the traveling route of each vehicle M obtained from the history of the self-position of each vehicle M acquired from each vehicle M.

If it is determined that the determination target feature is passable (step S220: Yes), the state determining unit 111 matches the determination result in step S220 with the passability information regarding the determination target feature in the master map information. It is determined whether or not the conditions are met (step S230). When it is determined that the determination result and the passability information match (step S230: Yes), the state determination unit 111 executes step S210 again.

If it is determined that the determination result and the passability information do not match (step S230: No), the state determination unit 111 determines whether the number of times it has been determined that the vehicle is passable is equal to or greater than a preset threshold. (Step S240). If it is determined that the number of times it has been determined that the vehicle is passable is less than the threshold (step S240: No), the state determination unit 111 executes step S210 again.

If it is determined that the number of times it has been determined that the feature is passable is equal to or greater than the threshold (step S240: Yes), the state updating unit 112 updates the master map information by determining that the target feature is in a passable state (step S240: Yes). S250). More specifically, the status update unit 112 updates the passability information in the master map information so that it matches the determination result in step S220 by the status determination unit 111. If the number of times it has been determined that it is passable is still small, the target feature may not be in a stable passable state, and if you update the master map information in such a case, the master map information will be higher. It may become unstable due to frequent updates. Therefore, in this embodiment, when the number of times it has been determined that the vehicle is passable is equal to or greater than the threshold, the status update unit 112 updates the master map information.

After completing step S250, the state determining unit 111 executes step S210 again.

On the other hand, if it is determined that the determination target feature is in an impassable state (step S220: No), the state determination unit 111 uses the determination result in step S220 by the state determination unit 111 and the determination target feature in the master map information. It is determined whether the traffic permission information regarding the vehicle matches the traffic permission information (step S232). If it is determined that the determination result and the passability information match (step S232: Yes), the state determination unit 111 executes step S210 again.

If it is determined that the determination result and the passability information do not match (step S232: No), the state determination unit 111 determines whether the number of times the vehicle is determined to be impassable is equal to or greater than a preset threshold. (Step S242). If it is determined that the number of times the vehicle has been determined to be impassable is less than the threshold (step S242: No), the state determination unit 111 executes step S210 again. The intention to perform the determination in step S242 is the same as the intention to perform the determination in step S240 described above.

If it is determined that the number of times the feature has been determined to be impassable is equal to or greater than the threshold (step S242: Yes), the state updating unit 112 updates the master map information by determining that the target feature is in an impassable state (step S242: Yes). S252). Furthermore, in step S262, the state notification unit 113 notifies the administrator Ad that the determination target feature is in an impassable state. After completing step S262, the state determining unit 111 executes step S210 again. As described above, in the server 100, steps S210 to S262 are repeatedly executed for each feature to be determined.

According to the self-position estimation system 1000 of the first embodiment described above, it is determined whether the determination target feature is in a passable state by using the environmental information and detection information about each of the plurality of vehicles M. However, when the state determination unit 111 determines that the target feature is in an impassable state, it notifies the administrator of the target feature, etc., so that an opportunity to repair the feature can be given. , it is possible to suppress a decrease in the stability of self-position estimation due to differences between map information and the state of actual features.

Furthermore, when the state determination unit 111 determines that the target feature is in an impassable state, the self-position estimation system 1000 switches the driving state from automatic driving to manual driving and changes the driving route. , and notifying the driver, it is possible to drive while avoiding features that are impassable.

In addition, the self-position estimation system 1000 updates the passability information in the master map information when the passability information and the determination result by the state determining unit 111 are different from each other, so that the master map information corresponds to the state of the actual feature. It is possible to modify the content according to the information provided, thereby suppressing a decrease in the accuracy of self-position estimation.

B. Second embodiment:
As shown in FIG. 8, the self-position estimation system 1000 of the second embodiment executes steps S242A to S248A shown in FIG. 9 instead of step S242 of the map information update process shown in FIG. This is different from the self-position estimation system 1000 of the first embodiment. The system configuration and other procedures in the map information update process of the self-position estimation system 1000 of the second embodiment are the same as those of the self-position estimation system 1000 of the first embodiment. The detailed explanation will be omitted.

In step S242A shown in FIG. 9, the state determination unit 111 acquires the duration of the impassable state for the determination target feature. The duration of the impassable state can be determined by taking the difference between the time when the determination target feature is first determined to be impassable and the current time.

In step S244A, the state determination unit 111 measures the number of detours of the vehicle M at the determination target feature. The "number of detours" means the number of times that a determination result that the determination target feature is impassable was obtained and the vehicle M detoured around the determination target feature. The number of detours can be determined from the travel trajectory information of each vehicle M described above.

In step S246A, the state determining unit 111 calculates the impassability possibility Pc from the duration and the number of detours. "Possibility of impassability Pc" means a numerical value representing the possibility that the target feature is actually impassable. In this embodiment, the impassability possibility Pc is calculated by referring to a table created in advance that shows the relationship between the duration time, the number of detours, and the impassability possibility Pc. This table is created so that the possibility of impassability Pc increases as the duration becomes longer and the number of detours increases. This is because if the duration is long or the number of detours is large, there is a high possibility that the impassable state of the determination target feature occurs continuously rather than accidentally.

In step S248A, the state determining unit 111 determines whether the calculated impassability possibility Pc is greater than or equal to a preset threshold. When the impassability possibility Pc is less than the threshold value (step S248A: No), the state determination unit 111 executes step S210 shown in FIG. 8 again. On the other hand, if the impassability possibility Pc is equal to or greater than the threshold value in FIG. 9 (step S248A: Yes), the state updating unit 112 executes step S252 shown in FIG. 8. In other words, if the determination result that the target feature is impassable does not change after a preset period of time has elapsed, or the determination result that the target feature is impassable is set in advance. If the number of times obtained is equal to or greater than the specified number of times, the state updating unit 112 executes step S252.

According to the self-position estimation system 1000 of the second embodiment described above, when the determination target feature continues to be in an impassable state for more than a preset time, or the determination target feature is in an impassable state. When the determination result is obtained more than a preset number of times, the possibility of impassability Pc is increased, and when the possibility of impassability Pc is greater than or equal to a preset threshold, the target feature becomes impassable. The system updates the passability information in the master map information and notifies the administrator of the target feature to that effect, giving an opportunity to repair the feature and reducing the accuracy of the passability information. can be suppressed.

C. Other embodiments:
(C1) In the embodiment described above, each vehicle M includes the position estimation unit 250, but the present disclosure is not limited thereto. The server 100 includes a map information acquisition section 240 and a position estimation section 250, and estimates the self-position of each vehicle M using the map information possessed by the server 100 and the environmental information and detection information acquired from each vehicle M. The vehicle M may be notified. Also in this embodiment, the same effects as in the above embodiment can be achieved.

(C2) In the above embodiment, when updating the master map information, the status updating unit 112 sets the condition that the determination result by the status determining unit 111 and the passability information in the master map information do not match. It is not limited to this. The status update unit 112 may update the master map information every time a new determination result is obtained. According to this embodiment, the same effects as those of the embodiment described above can be achieved, and it is possible to suppress the map information update process from becoming complicated.

(C3) In the above embodiment, the state determining unit 111 calculates the impassability possibility Pc using the duration of the impassable state and the number of detours of the vehicle M, but the present disclosure is not limited to this. . The state determining unit 111 may calculate the impassability possibility Pc using only one of the duration and the number of detours. Also in this embodiment, the same effects as in the above embodiment can be achieved.

The present disclosure is not limited to the embodiments described above, and can be implemented in various configurations without departing from the spirit thereof. For example, the technical features in each embodiment that correspond to the technical features in the form described in the summary column of the invention may be used to solve some or all of the above-mentioned problems, or to achieve one of the above-mentioned effects. In order to achieve some or all of the above, it is possible to replace or combine them as appropriate. Further, unless the technical feature is described as essential in this specification, it can be deleted as appropriate.

The self-localization system 1000 and techniques thereof described in this disclosure are implemented using a dedicated computer provided by configuring a processor and memory programmed to perform one or more functions embodied by a computer program. It may be realized by. Alternatively, the self-localization system 1000 and techniques described in this disclosure may be implemented by a dedicated computer provided by a processor configured with one or more dedicated hardware logic circuits. Alternatively, the self-localization system 1000 and its method described in the present disclosure may include a processor configured with a processor and memory programmed to perform one or more functions and one or more hardware logic circuits. It may be realized by one or more dedicated computers configured by a combination of the following. The computer program may also be stored as instructions executed by a computer on a computer-readable non-transitory tangible storage medium.

M...Vehicle, 111...State determination section, 113...State notification section, 210...Environment recognition section, 240...Map information acquisition section, 250...Position estimation section, 1000...Self position estimation system

Claims (5)

A self-position estimation system (1000) capable of communicating with a plurality of vehicles (M),
an environment recognition unit (210) that is installed in each of the plurality of vehicles and that acquires environmental information that is information about the environment surrounding the vehicle;
a map information acquisition unit (240) that acquires map information;
The environmental information about the vehicle is acquired from the environment recognition unit mounted on each of the plurality of vehicles, and information about the driving state and position of the vehicle is obtained from the sensor mounted on each of the plurality of vehicles. a position estimating unit (250) that acquires certain detection information and estimates the self-position of each of the plurality of vehicles using the environmental information, the map information, and the detection information;
Determining whether or not a determination target feature existing on the travel route of each of the vehicles is in a passable state by using the estimated self-position of each of the plurality of vehicles and the acquired environmental information. a state determination unit (111) that
a state notification unit (113) that notifies a predetermined destination when the state determination unit determines that the determination target feature is in an impassable state;
Equipped with
Self-localization system. The self-position estimation system according to claim 1,
Each of the plurality of vehicles includes the map information acquisition section and the position estimation section,
The self-position estimation system stores passability information stored as part of master map information, which is the map information to be acquired by the map information acquisition unit of each of the vehicles, which indicates the passability state of a feature. further comprising a status update unit (112) that updates passability information, which is information indicating
The state updating unit updates the passability information in the master map information to the passability information in the master map information when the passability state indicated by the passability information in the master map information and the determination result by the state determination unit are different from each other. update to match the results,
Self-localization system. The self-position estimation system according to claim 2,
Each of the plurality of vehicles has a corresponding processing unit (260),
The response processing unit is configured to indicate that the passability information included in the map information acquired by the map information acquisition unit indicates that a feature existing on the travel route of the vehicle is impassable. In this case, performing at least one of the following: switching the driving state of the vehicle from automatic driving to manual driving, changing the driving route of the vehicle, and notifying the driver.
Self-localization system. The self-position estimation system according to claim 2 or 3,
The state determination unit determines that the determination target feature is in an impassable state when the determination result of the passability state of the determination target feature remains impassable even after a preset time has elapsed. Increasing the numerical value (Pc) representing the possibility that
The state updating unit updates the passability information by determining that the determination target feature is in an impassable state when the numerical value is greater than or equal to a preset threshold;
The state notification unit notifies the destination of information indicating that the determination target feature is in an impassable state when the numerical value is greater than or equal to a preset threshold;
Self-localization system. The self-position estimation system according to claim 4,
The state determination unit increases the numerical value when the same determination result that the determination target feature is impassable is obtained a preset number of times or more within a preset time. ,
Self-localization system.

Images