JP7616175B2 - Vehicle position estimation method - Google Patents

Description

This disclosure relates to a localization technique that uses a camera mounted on a vehicle to estimate the vehicle's position.

Patent Document 1 discloses a technology for estimating vehicle position based on the positions of feature points on an image captured by an on-board camera. The amount of error in the estimated vehicle position is estimated based on the vehicle roll angle and the height of the camera. The height of the camera is estimated from the amount of luggage carried and the control amount of the active suspension.

JP 2019-100970 A

Localization is a technology that uses a camera mounted on a vehicle to estimate the vehicle's position with high accuracy. If the vehicle tilts due to being loaded with luggage, etc., the height and orientation (optical axis direction) of the camera may deviate from the reference state, which may reduce the accuracy of localization. Changes in the height and orientation of the camera from the reference state can be determined based on the amount of suspension stroke of each wheel. Therefore, it is conceivable to perform localization by correcting the image captured by the camera based on the amount of suspension stroke.

On the other hand, when a vehicle is traveling, unevenness on the road surface causes vertical vibrations in the vehicle. When large vertical vibrations occur, it takes time for the vibrations (expansion and contraction) of the suspension springs to converge. Such vibrations of the suspension springs lead to a decrease in the accuracy of the above correction based on the suspension stroke amount. A decrease in correction accuracy leads to a decrease in localization accuracy.

One objective of the present disclosure is to provide technology that can further improve the accuracy of localization for estimating the vehicle's position.

The first aspect relates to a vehicle position estimation method for estimating the position of a vehicle in a predetermined area in which markers are arranged.
The vehicle position estimation method is
Recognizing the positions of markers around the vehicle using a camera mounted on the vehicle;
correcting the position of the recognized marker based on a stroke amount of a suspension of the vehicle;
Estimating a position of the vehicle in the predetermined area based on the corrected positions of the markers;
The damping force of the suspension is set higher within the predetermined area compared to outside the predetermined area.

The second aspect relates to a vehicle position estimation method for estimating the position of a vehicle equipped with an automatic valet parking function.
The vehicle position estimation method is
Recognizing the positions of markers around the vehicle using a camera mounted on the vehicle;
correcting the position of the recognized marker based on a stroke amount of a suspension of the vehicle;
Estimating a position of the vehicle based on the corrected positions of the markers;
The method includes setting a damping force of the suspension higher while automatic valet parking is being performed, compared to when automatic valet parking is not being performed.

The third aspect relates to a vehicle position estimation method for estimating a position of a vehicle.
The vehicle position estimation method is
Recognizing the positions of markers around the vehicle using a camera mounted on the vehicle;
correcting the position of the recognized marker based on a stroke amount of a suspension of the vehicle;
Estimating a position of the vehicle based on the corrected positions of the markers;
Obtaining information on a first position where a degree of unevenness of a road surface exceeds a threshold;
When the vehicle passes through the first position, a damping force of the suspension is set higher than that at positions other than the first position.

According to the present disclosure, the positions of markers around the vehicle are recognized by a camera mounted on the vehicle. Furthermore, the recognized positions of the markers are corrected based on the stroke amount of the suspension of the vehicle. Localization is then performed based on the corrected positions of the markers. Since localization is performed taking into account the stroke amount of the suspension, a decrease in localization accuracy due to the inclination of the vehicle, etc. is suppressed.

Furthermore, if necessary, the damping force of the vehicle's suspension is set higher than the default. By increasing the damping force of the suspension, the vibration (compression) of the suspension spring converges quickly. In other words, the vibration of the suspension stroke amount converges quickly. Therefore, the accuracy of the above correction based on the suspension stroke amount is improved. As a result, the localization accuracy is improved.

1 is a conceptual diagram for explaining an overview of a vehicle control system according to an embodiment; FIG. 1 is a conceptual diagram for explaining an overview of automatic valet parking according to an embodiment. FIG. 1 is a conceptual diagram for explaining the effect of vehicle tilt on localization. 1 is a conceptual diagram for explaining a configuration of a vehicle according to an embodiment. FIG. 2 is a conceptual diagram for explaining an example of suspension control according to the embodiment. FIG. 11 is a conceptual diagram for explaining another example of the suspension control according to the embodiment. 1 is a block diagram showing an example of the configuration of a vehicle control system according to an embodiment; FIG. 2 is a block diagram showing an example of various information according to an embodiment.

An embodiment of the present disclosure will be described with reference to the attached drawings.

1. Overview of the Vehicle Control System Fig. 1 is a conceptual diagram for explaining an overview of a vehicle control system 10 according to the present embodiment. The vehicle control system 10 controls a vehicle 1. For example, the vehicle control system 10 is mounted on the vehicle 1. Alternatively, at least a part of the vehicle control system 10 may be disposed in an external device outside the vehicle 1 and control the vehicle 1 remotely.

The vehicle control system 10 controls the driving of the vehicle 1 at least in a specified area AR. Examples of the specified area AR include a parking lot, a city (smart city), etc. For example, the vehicle control system 10 controls the driving of the vehicle 1 so that the vehicle 1 follows a target route TP in the specified area AR. The target route TP may be determined in advance or may be calculated by the vehicle control system 10.

Information on the current position of vehicle 1 is required for vehicle driving control. Therefore, the vehicle control system 10 performs "localization" to estimate the position of vehicle 1 in a specified area AR. Localization is also called self-position estimation processing. Hereinafter, the position of vehicle 1 will be simply called "vehicle position."

In localization, the vehicle control system 10 calculates the amount of movement (displacement) of the vehicle 1 based on the steering angle and vehicle speed of the vehicle 1, and estimates the approximate vehicle position based on the amount of movement. The method of estimating the vehicle position based on the steering angle and vehicle speed is also called "dead reckoning."

To further improve the accuracy of estimating the vehicle position, markers M (landmarks) arranged in the specified area AR are also used. More specifically, the vehicle 1 is equipped with a camera 22 that captures the surroundings of the vehicle 1. The vehicle control system 10 acquires images captured by the camera 22, recognizes the markers M around the vehicle 1 based on the images, and grasps the relative positional relationship between the vehicle 1 and the markers M. The vehicle control system 10 also holds map information indicating the installation positions of the markers M in the specified area AR. The vehicle control system 10 corrects the vehicle position obtained by dead reckoning by comparing the recognized relative position of the marker M with the installation position of the marker M in the specified area AR. By repeatedly performing position estimation by dead reckoning and position correction using the markers M and the camera 22, a highly accurate vehicle position can be continuously obtained.

The localization described above is performed, for example, in "Automated Valet Parking (AVP)" in a parking lot.

Figure 2 is a conceptual diagram for explaining an overview of automatic valet parking. The specified area AR is a parking lot PL. A plurality of markers M are placed in the parking lot PL. Vehicle 1 is an AVP vehicle equipped with a function for automatic valet parking in the parking lot PL, and can automatically drive at least within the parking lot PL. Vehicle 1 may be an autonomous vehicle.

The vehicle control system 10 controls the vehicle 1 in the parking lot PL. More specifically, the vehicle control system 10 includes an on-board system 10V mounted on the vehicle 1 and a management system 10M external to the vehicle 1.

The management system 10M manages automatic valet parking in the parking lot PL. The management system 10M can communicate with each vehicle (vehicle 1, parked vehicle 3) in the parking lot PL. For example, the management system 10M may issue entry and exit instructions to the in-vehicle system 10V. The management system 10M may provide map information of the parking lot PL to the in-vehicle system 10V. The management system 10M may assign a parking space to the vehicle 1. The management system 10M may generate a target route TP from the entry area to the assigned parking space and provide information on the target route TP to the in-vehicle system 10V. The management system 10M may remotely operate each vehicle (vehicle 1, parked vehicle 3) in the parking lot PL.

The in-vehicle system 10V controls the automatic driving of the vehicle 1 in the parking lot PL. For example, the in-vehicle system 10V uses the camera 22 mounted on the vehicle 1 to recognize markers M around the vehicle 1. The in-vehicle system 10V then performs the above-mentioned localization based on the recognition result of the markers M, and estimates the vehicle position in the parking lot PL with high accuracy. The in-vehicle system 10V also receives information on the target route TP from the management system 10M. The in-vehicle system 10V then controls the driving of the vehicle 1 so that it follows the target route TP based on the vehicle position and the target route TP. This enables the vehicle 1 to automatically move from the entrance area to the target parking space.

As a variant, the management system 10M may perform localization. In that case, the in-vehicle system 10V transmits information necessary for localization, such as the steering angle of the vehicle 1, the vehicle speed, and images captured by the camera 22, to the management system 10M. The management system 10M performs localization based on the information received from the in-vehicle system 10V, and estimates the vehicle position in the parking lot PL with high accuracy.

2. Localization Considering the Stroke Amount of the Suspension Fig. 3 is a conceptual diagram for explaining the influence of the tilt of the vehicle 1 on localization. The reference state is a state in which the vehicle 1 is not loaded with luggage or the like and is horizontal. There is a possibility that the vehicle 1 may tilt from the horizontal due to loading luggage or the like on the vehicle 1. For example, the vehicle 1 tilts in the pitch direction. When the vehicle 1 tilts, the height and orientation (optical axis direction) of the camera 22 mounted on the vehicle 1 deviate from the reference state. When the height and orientation of the camera 22 deviate, the position of the marker M in the image captured by the camera 22 changes, and there is a risk of the localization accuracy decreasing.

Even if the vehicle 1 does not tilt, it is possible that the vehicle 1 may sink as a whole due to heavy luggage. Even in this case, the height of the camera 22 will at least deviate from the reference state, and localization accuracy may decrease.

Therefore, in this embodiment, localization is performed taking into account the tilt and sinking of the vehicle 1. The tilt and sinking of the vehicle 1 can be grasped based on the amount of suspension stroke of each wheel of the vehicle 1.

Figure 4 is a conceptual diagram for explaining the configuration of the vehicle 1. The vehicle 1 is equipped with wheels 2 and a suspension 30. The wheels 2 include a left front wheel 2FL, a right front wheel 2FR, a left rear wheel 2RL, and a right rear wheel 2RR. A suspension 30 is provided for each of the left front wheel 2FL, the right front wheel 2FR, the left rear wheel 2RL, and the right rear wheel 2RR.

The suspension 30 is provided to connect between the sprung structure and the unsprung structure of the vehicle 1. The unsprung structure includes the wheels 2. The suspension 30 includes a spring and a shock absorber (damper). The spring and the shock absorber are provided in parallel between the sprung structure and the unsprung structure. The damping force of the shock absorber is variable. In the following description, the "damping force of the shock absorber" may simply be referred to as the "damping force of the suspension 30". The vehicle control system 10 can variably control the damping force of the suspension 30 for each wheel 2 of the vehicle 1.

The stroke amount of the suspension 30 is the relative displacement between the sprung structure and the unsprung structure. The stroke amount of each suspension 30 can be detected by a sensor 24. For example, the sensor 24 is a stroke sensor installed on each suspension 30. As another example, the sensor 24 may be a sprung acceleration sensor provided on the sprung structure above the position of each wheel 2. The sprung acceleration sensor detects the sprung acceleration (vertical acceleration) of the sprung structure. It is also possible to estimate the stroke amount from the sprung acceleration by using an observer configured based on a single wheel two-degree-of-freedom model.

The vehicle control system 10 uses the sensor 24 to acquire the stroke amount of the suspension 30 of each wheel 2. The tilt and sinking of the vehicle 1 are reflected in the stroke amount of the suspension 30 of each wheel 2. The vehicle control system 10 corrects the position of the marker M recognized from the image captured by the camera 22 based on the stroke amount of each suspension 30. More specifically, the vehicle control system 10 corrects the position of the marker M recognized from the image so that it is in the reference state. In other words, the vehicle control system 10 cancels out the deviation of the position of the marker M from the reference state based on the stroke amount of each suspension 30.

Then, the vehicle control system 10 performs localization based on the corrected position of the marker M, i.e., the position of the marker M corresponding to the reference state. This makes it possible to suppress a decrease in localization accuracy caused by the tilt or sinking of the vehicle 1.

3. Suspension Control As described above, the vehicle control system 10 recognizes the position of the marker M around the vehicle 1 using the camera 22 mounted on the vehicle 1. Furthermore, the vehicle control system 10 corrects the recognized position of the marker M based on the stroke amount of the suspension 30 of the vehicle 1. Then, the vehicle control system 10 performs localization based on the corrected position of the marker M, and estimates the vehicle position in the predetermined area AR.

On the other hand, when the vehicle 1 is traveling, the vehicle 1 vibrates up and down due to unevenness of the road surface. When large up and down vibrations occur, it takes time for the vibration (expansion and contraction) of the spring of the suspension 30 to converge. The vibration of the spring of the suspension 30 leads to a decrease in the accuracy of the above correction based on the stroke amount of the suspension 30. The decrease in correction accuracy leads to a decrease in localization accuracy.

Therefore, the vehicle control system 10 according to the present embodiment controls the suspension 30 to suppress a decrease in localization accuracy. More specifically, when a predetermined condition is met, the vehicle control system 10 sets the damping force of the suspension 30 of each wheel 2 higher than the default. By increasing the damping force of the suspension 30, the vibration (expansion and contraction) of the spring of the suspension 30 converges quickly. In other words, the vibration of the stroke amount of the suspension 30 converges quickly. Therefore, the accuracy of the above correction based on the stroke amount of the suspension 30 is improved. As a result, the localization accuracy is improved.

There are various possible examples of the specified conditions that increase the damping force of the suspension 30. Various examples of the specified conditions are described below.

3-1. First Example FIG. 5 is a conceptual diagram for explaining a first example of suspension control. A marker M is arranged within the predetermined area AR. Within the predetermined area AR, the vehicle control system 10 performs localization using the marker M. Within the predetermined area AR where localization is performed in this manner, the vehicle control system 10 sets the damping force of the suspension 30 higher than the default. On the other hand, since localization is not performed outside the predetermined area AR, the vehicle control system 10 maintains the damping force of the suspension 30 at the default. That is, within the predetermined area AR, the damping force of the suspension 30 is set higher than outside the predetermined area AR. This improves the accuracy of localization performed in the predetermined area AR.

3-2. Second Example In FIG. 5, the predetermined area AR may be a parking lot PL. Then, automatic valet parking of the vehicle 1 may be performed in the parking lot PL (see FIG. 2). When automatic valet parking of the vehicle 1 is being performed, the vehicle control system 10 sets the damping force of the suspension 30 higher than the default. On the other hand, when automatic valet parking of the vehicle 1 is not being performed, the vehicle control system 10 maintains the damping force of the suspension 30 at the default. That is, when automatic valet parking is being performed, the damping force of the suspension 30 is set higher than when automatic valet parking is not being performed. This improves the localization accuracy during automatic valet parking.

The state in which the damping force of the suspension 30 is high corresponds to the so-called "stiff suspension 30 state." However, automatic valet parking of the vehicle 1 is generally performed with no occupants in the vehicle 1. Since there are no occupants in the vehicle 1, even if the suspension 30 becomes stiff, there is no impact on the ride comfort.

3-3. Third Example FIG. 6 is a conceptual diagram for explaining a third example of suspension control. In the third example, consider a situation in which there are large irregularities on the road surface within the specified area AR. For example, as shown in FIG. 6, there is a case in which a bump for suppressing speed is installed on the road surface within a parking lot PL. As another example, there is a possibility that a depression exists on the road surface within the specified area AR.

The "first position P1" is a position where the unevenness of the road surface exceeds a threshold. For example, the first position P1 is a position where a bump exists. As another example, the first position P1 is a position where a depression exists. The vehicle control system 10 acquires information on such a first position P1. For example, the first position P1 is registered in advance in map information of the specified area AR. In this case, the vehicle control system 10 acquires information on the first position P1 from the map information of the specified area AR. As another example, the vehicle control system 10 may acquire an image captured by an infrastructure camera installed in the specified area AR, and recognize the first position P1 based on the acquired image. As yet another example, the vehicle control system 10 may acquire an image captured by a camera 22 mounted on the vehicle 1, and recognize the first position P1 based on the acquired image.

When the vehicle 1 passes through the first position P1, large vertical vibrations may occur. Therefore, when the vehicle 1 passes through the first position P1, the vehicle control system 10 sets the damping force of the suspension 30 to be higher than the default. On the other hand, at positions other than the first position P1, the vehicle control system 10 maintains the damping force of the suspension 30 at the default. In other words, when the vehicle 1 passes through the first position P1, the damping force of the suspension 30 is set higher compared to positions other than the first position P1. This improves the localization accuracy when the vehicle 1 passes through the first position P1 and during the period following that.

4. Effects As described above, according to this embodiment, the camera 22 mounted on the vehicle 1 recognizes the positions of the markers M around the vehicle 1. Furthermore, the recognized positions of the markers M are corrected based on the stroke amount of the suspension 30 of the vehicle 1. Then, localization is performed based on the corrected position of the markers M. Since localization is performed taking into account the stroke amount of the suspension 30, a decrease in localization accuracy caused by tilting or sinking of the vehicle 1 is suppressed.

Furthermore, if necessary, the damping force of the suspension 30 of the vehicle 1 is set higher than the default. By increasing the damping force of the suspension 30, the vibration (compression) of the spring of the suspension 30 converges quickly. In other words, the vibration of the stroke amount of the suspension 30 converges quickly. Therefore, the accuracy of the above correction based on the stroke amount of the suspension 30 is improved. As a result, the localization accuracy is improved.

The process of setting the damping force of the suspension 30 high may be performed when there are no occupants in the vehicle 1. Since there are no occupants in the vehicle 1, even if the suspension 30 becomes stiff, there is no effect on the ride comfort.

7 is a block diagram showing a configuration example of a vehicle control system 10 according to the present embodiment. The vehicle control system 10 includes an on-vehicle sensor 20, a suspension 30, a communication device 40, a driving device 50, and a control device 100.

The on-board sensor 20 is mounted on the vehicle 1. The on-board sensor 20 includes a recognition sensor 21 and a vehicle condition sensor 23.

The recognition sensor 21 recognizes (detects) the situation around the vehicle 1. The recognition sensor 21 includes a camera 22 that captures the situation around the vehicle 1. The camera 22 may include a camera for localization and a camera for vehicle driving control. The camera for localization is, for example, a PVM (panoramic view monitor) camera equipped with a fisheye lens. The recognition sensor 21 may further include a LIDAR (Laser Imaging Detection and Ranging), radar, sonar, etc.

The vehicle state sensor 23 detects the state of the vehicle 1. Examples of the vehicle state sensor 23 include a vehicle speed sensor (wheel speed sensor), a steering angle sensor, a yaw rate sensor, a lateral acceleration sensor, and the like. The vehicle state sensor 23 also includes a sensor 24 (see FIG. 4) for acquiring the stroke amount of the suspension 30 of each wheel 2 of the vehicle 1. For example, the sensor 24 is a stroke sensor installed on each suspension 30. As another example, the sensor 24 may be a sprung acceleration sensor provided on a sprung structure above the position of each wheel 2.

The suspension 30 is provided on each wheel 2 of the vehicle 1 (see FIG. 4). The suspension 30 includes a spring and a shock absorber (damper). The damping force of the shock absorber is variable.

The communication device 40 communicates with the outside of the vehicle control system 10. For example, the communication device 40 communicates with an infrastructure camera installed within the specified area AR.

The traveling device 50 is mounted on the vehicle 1. The traveling device 50 includes a steering device, a drive device, and a braking device. The steering device steers the wheels of the vehicle 1. For example, the steering device includes an electric power steering (EPS) device. The drive device is a power source that generates a driving force. Examples of the drive device include an engine, an electric motor, an in-wheel motor, etc. The braking device generates a braking force.

The control device 100 controls the vehicle 1. The control device 100 includes one or more processors 110 (hereinafter simply referred to as processor 110) and one or more storage devices 120 (hereinafter simply referred to as storage devices 120). The processor 110 executes various processes. For example, the processor 110 includes a CPU (Central Processing Unit). The storage device 120 stores various information 200. Examples of the storage device 120 include a volatile memory, a non-volatile memory, a HDD (Hard Disk Drive), and an SSD (Solid State Drive). The control device 100 may be distributed between the in-vehicle system 10V and the management system 10M (see FIG. 2).

The vehicle control program PROG is a computer program for controlling the vehicle 1. The processor 110 executes the vehicle control program PROG, thereby realizing various processes by the control device 100. The vehicle control program PROG is stored in the storage device 120. Alternatively, the vehicle control program PROG may be recorded on a computer-readable recording medium.

8 is a block diagram showing an example of various information 200 stored in the storage device 120. The various information 200 includes map information 210, surrounding situation information 220, vehicle state information 230, localization information 240, target route information 250, and the like.

The map information 210 includes at least map information of a specified area AR. The map information 210 of the specified area AR indicates the positions of roads and structures within the specified area AR. The map information 210 also indicates the positions of each marker M installed in the specified area AR. Furthermore, the map information 210 may indicate a first position P1 (see FIG. 6) where the unevenness of the road surface exceeds a threshold value.

The surrounding situation information 220 indicates the situation around the vehicle 1. The surrounding situation information 220 indicates the result of recognition by the recognition sensor 21. For example, the surrounding situation information 220 includes images captured by the camera 22. The surrounding situation information 220 also includes object information related to objects around the vehicle 1. Examples of objects include pedestrians, other vehicles, white lines, markers M, structures, etc. The object information indicates the relative position and relative speed of the object with respect to the vehicle 1. For example, by analyzing the image obtained by the camera 22, the object can be identified and the relative position of the object can be calculated. For example, the control device 100 uses image recognition AI obtained by machine learning to identify objects in the image. Also, it is possible to identify objects and obtain the relative position and relative speed of the object based on point cloud information obtained by the lidar.

The vehicle state information 230 is information indicating the state of the vehicle 1, and indicates the detection results by the vehicle state sensor 23. Examples of the state of the vehicle 1 include the vehicle speed (wheel speed), steering angle, yaw rate, lateral acceleration, etc. The state of the vehicle 1 further includes the stroke amount of each suspension 30 detected by the sensor 24.

Vehicle status information 230 may further indicate whether automated valet parking is being performed for vehicle 1.

Localization information 240 indicates the location of vehicle 1 obtained by localization.

The target route information 250 indicates the target route TP of the vehicle 1 in the specified area AR. For example, the target route information 250 indicates the target route TP from the entrance area in the parking lot PL to the target parking space (see FIG. 2).

5-3. Suspension Control The control device 100 performs the suspension control described above in Section 3. That is, the control device 100 sets the damping force of the suspension 30 of each wheel 2 higher than the default, as necessary.

According to the first example described above (see section 3-1), the damping force of the suspension 30 is set higher within the specified area AR compared to outside the specified area AR. The control device 100 determines whether the vehicle 1 is within the specified area AR based on the map information 210 and the localization information 240 (vehicle position). If the vehicle 1 is outside the specified area AR, the control device 100 sets the damping force of the suspension 30 to the default. On the other hand, if the vehicle 1 is within the specified area AR, the control device 100 sets the damping force of the suspension 30 to a higher value than the default.

According to the second example described above (see section 3-2), when automatic valet parking is being performed, the damping force of the suspension 30 is set higher compared to when automatic valet parking is not being performed. The control device 100 determines whether or not automatic valet parking of the vehicle 1 is being performed based on the vehicle state information 230. When automatic valet parking of the vehicle 1 is not being performed, the control device 100 sets the damping force of the suspension 30 to a default. On the other hand, when automatic valet parking of the vehicle 1 is being performed, the control device 100 sets the damping force of the suspension 30 higher than the default.

According to the third example described above (see Section 3-3), when the vehicle 1 passes through the first position P1, the damping force of the suspension 30 is set higher than at positions other than the first position P1. The first position P1 is a position where the unevenness of the road surface exceeds a threshold value.

For example, the first position P1 is preregistered in the map information 210 of the specified area AR. In this case, the control device 100 can obtain information on the first position P1 from the map information 210 of the specified area AR.

As another example, the control device 100 may communicate with an infrastructure camera installed within the specified area AR via the communication device 40 and acquire images captured by the infrastructure camera. The control device 100 can recognize the first position P1 based on the acquired images.

As yet another example, the control device 100 may recognize the first position P1 based on surrounding situation information 220 that indicates the situation around the vehicle 1.

Other than the first position P1, the control device 100 sets the damping force of the suspension 30 to the default. On the other hand, when the vehicle 1 passes through the first position P1, the control device 100 sets the damping force of the suspension 30 to a value higher than the default.

5-4. Localization The control device 100 performs "localization" to estimate the position of the vehicle 1 in the specified area AR. More specifically, the control device 100 calculates the amount of movement of the vehicle 1 based on the vehicle state information 230 (vehicle speed, steering angle), and estimates the position of the vehicle 1 based on the amount of movement. The control device 100 also acquires the positions of the markers M in the specified area AR from the map information 210, and acquires the relative positions of the markers M with respect to the vehicle 1 from the surrounding situation information 220. The control device 100 then corrects the estimated position of the vehicle 1 based on the estimated position of the vehicle 1, the relative positional relationship between the vehicle 1 and the markers M, and the installation positions of the markers M in the specified area AR. By repeatedly performing position estimation and position correction, a highly accurate vehicle position can be continuously obtained.

As described in section 2 above, the control device 100 may perform localization taking into account the stroke amount of the suspension 30 of the vehicle 1. The stroke amount of each suspension 30 is obtained from the vehicle state information 230. The control device 100 corrects the image captured by the camera 22 based on the stroke amount of each suspension 30. More specifically, the control device 100 corrects the position of the marker M recognized from the image based on the stroke amount of each suspension 30 so that it is the position in the reference state. The control device 100 then performs localization based on the position of the marker M after the correction, i.e., the position of the marker M corresponding to the reference state.

5-5. Vehicle Driving Control The control device 100 performs vehicle driving control for controlling the driving of the vehicle 1. The control device 100 performs vehicle driving control by controlling the driving devices 50 (steering device, drive device, braking device).

In particular, the control device 100 executes driving control processing so that the vehicle 1 follows the target route TP. To achieve this, the control device 100 calculates the deviation (e.g., lateral deviation, yaw angle deviation) between the vehicle 1 and the target route TP. The position of the vehicle 1 is obtained from the localization information 240. The target route TP is obtained from the target route information 250. Then, the control device 100 controls the driving of the vehicle 1 so that the deviation between the vehicle 1 and the target route TP is reduced.

Vehicle driving control also enables automatic valet parking of vehicle 1 in parking lot PL (see Figure 2).

REFERENCE SIGNS LIST 1 Vehicle 2 Wheel 10 Vehicle control system 10M Management system 10V On-board system 20 On-board sensor 21 Recognition sensor 22 Camera 23 Vehicle state sensor 24 Sensor 30 Suspension 40 Communication device 50 Travel device 100 Control device 110 Processor 120 Storage device 200 Various information 210 Map information 220 Surrounding situation information 230 Vehicle state information 240 Localization information 250 Target route information AR Predetermined area TP Target route

Claims (6)

A vehicle position estimation method for estimating a position of a vehicle in a predetermined area in which a marker is arranged, comprising:
Recognizing positions of markers around the vehicle using a camera mounted on the vehicle;
correcting the position of the recognized marker based on a stroke amount of a suspension of the vehicle;
estimating the position of the vehicle in the predetermined area based on the corrected positions of the markers;
setting a damping force of the suspension higher within the predetermined area compared to a damping force outside the predetermined area. 2. A vehicle position estimation method according to claim 1, comprising:
In the predetermined area, automatic valet parking of the vehicle is performed based on the position of the vehicle. 3. A vehicle position estimation method according to claim 2, further comprising:
The vehicle position estimation method, wherein the automatic valet parking is performed in a state where no occupants are present in the vehicle. A vehicle position estimation method for estimating a position of a vehicle having a function of performing automatic valet parking, comprising:
Recognizing positions of markers around the vehicle using a camera mounted on the vehicle;
correcting the position of the recognized marker based on a stroke amount of a suspension of the vehicle;
estimating a position of the vehicle based on the corrected positions of the markers;
setting a damping force of the suspension higher while the automatic valet parking is being performed compared to when the automatic valet parking is not being performed. 5. A vehicle position estimation method according to claim 4, further comprising:
The vehicle position estimation method, wherein the automatic valet parking is performed in a state where no occupants are present in the vehicle. A vehicle position estimation method for estimating a position of a vehicle, comprising:
Recognizing positions of markers around the vehicle using a camera mounted on the vehicle;
correcting the position of the recognized marker based on a stroke amount of a suspension of the vehicle;
estimating the position of the vehicle based on the corrected positions of the markers;
Obtaining information on a first position where a degree of unevenness of a road surface exceeds a threshold;
setting a damping force of the suspension higher when the vehicle passes through the first position as compared to positions other than the first position.

Images