Vehicle control method, vehicle control system, and recording medium

Abstract

The present disclosure relates to a vehicle control method, a vehicle control system, and a recording medium. The vehicle control method controls a vehicle that supports automatic valet parking in a parking lot. The vehicle control method includes a positioning process, which is a process of estimating the vehicle position of the vehicle in the parking lot based on the detection results obtained by the sensor mounted on the vehicle. The vehicle control method also includes a process of accumulating the estimated error of the vehicle position estimated by the positioning process. The vehicle control method also includes a first margin setting process, which is a process of setting the margin distance to a correction value longer than a default value, and the correction value reflects the variance of the distribution of the estimated error. The vehicle control method also includes a vehicle stop process, which is a process of controlling the vehicle to stop at a margin distance before a designated stop position in the parking lot based on the vehicle position estimated by the positioning process.

Description

Technical Field

[0001] The present disclosure relates to a technology for controlling a vehicle that supports automatic valet parking (AVP: Automated Valet Parking) in a parking lot. Background Art

[0002] A known method is to use map information and on-board sensors to estimate a vehicle's own position through "positioning processing (localization)". Specifically, map information includes the positions of landmarks such as signs. The vehicle uses on-board sensors to calculate the approximate vehicle position and identify surrounding landmarks. The vehicle position is estimated with high accuracy by combining the approximate vehicle position, the positions of landmarks shown in the map information, and the relative positions of the landmarks identified by the on-board sensors.

[0003] Japanese Patent Laid-Open No. 2020-038359 discloses a map system for generating or updating map data including coordinate information of a plurality of landmarks for positioning processing.

[0004] Japanese Patent Application Laid-Open No. 2015-041348 discloses a parking guidance device, which is mounted on a vehicle and performs automatic parking of the vehicle. In automatic parking, various guide signs drawn on the road surface are used. For example, a stop position mark indicates a stop space for the vehicle for the driver to get on the vehicle. The parking guidance device uses an on-vehicle camera to recognize the stop position mark, guides the vehicle to a stop frame included in the stop position mark, and stops the vehicle.

[0005] Japanese Patent Laid-Open No. 2008-287572 discloses a vehicle driving assistance system that stops a vehicle at a stop line in front of an intersection. The on-board device detects the stop line from an image captured by a camera, and assists the driving of the vehicle by stopping at the detected stop line. In order to improve the detection accuracy of the stop line, the on-board device obtains information related to the distance to the stop line from an optical beacon and a road device arranged in front of the stop line. The on-board device obtains the measurement error of a rangefinder that measures the travel distance based on the distance to the stop line obtained at different positions. Furthermore, the on-board device calculates the error range of the travel distance when the vehicle travels to the stop line based on the measurement error of the rangefinder. Then, the on-board device determines the search area for searching the stop line in the captured image based on the error range of the travel distance.

[0006] An automatic valet parking service in a parking lot is known. A vehicle supporting automatic valet parking automatically drives at least in the parking lot. During the automatic driving, the vehicle performs positioning processing to estimate its own position using parking lot map information and on-board sensors. Then, the vehicle performs vehicle driving control based on the estimated vehicle position.

[0007] When the vehicle needs to stop during automatic driving in a parking lot, the automatic valet parking system instructs the vehicle to stop before a designated stop position (stop node). In response to the stop instruction, the vehicle performs vehicle driving control (brake control) in such a way that the vehicle stops before the designated stop position. However, if the positioning processing accuracy is not high, the vehicle may not stop before the designated stop position but may exceed the designated stop position. Summary of the invention

[0008] An object of the present disclosure is to provide a technology capable of suppressing a vehicle from exceeding a designated stop position when stopping a vehicle supporting automatic valet parking in a parking lot.

[0009] A first aspect of the present disclosure is associated with a vehicle control method for controlling a vehicle that supports automatic valet parking in a parking lot.

[0010] The vehicle control method includes: a positioning process for estimating a vehicle position of a vehicle in a parking lot based on a detection result obtained by a sensor mounted on the vehicle; a process for accumulating an estimated error of the vehicle position estimated by the positioning process; a first margin setting process for setting a margin distance to be longer than a default value by a correction value, the correction value reflecting the variance of a distribution of the estimated error; and a vehicle stop process for controlling the vehicle to stop at a margin distance before a designated stop position in the parking lot based on the vehicle position estimated by the positioning process.

[0011] A second aspect of the present disclosure is associated with a vehicle control system that controls a vehicle that supports automatic valet parking in a parking lot.

[0012] The vehicle control system comprises: one or more processors for performing positioning processing, which is a process of estimating the vehicle position of the vehicle in the parking lot based on the detection results obtained by the sensors mounted on the vehicle; and one or more storage devices for storing positioning error information, which indicates the distribution of the estimated error of the vehicle position estimated by the positioning processing.

[0013] The one or more processors also perform: a first margin setting process of setting the margin distance to a correction value longer than a default value, the correction value reflecting the variance of the distribution of the estimated error; and a vehicle stop process of controlling the vehicle to stop at a margin distance before a designated stop position in the parking lot based on the vehicle position estimated by the positioning process.

[0014] A third aspect of the present disclosure relates to a recording medium recording a vehicle control program for controlling a vehicle supporting automatic valet parking in a parking lot.

[0015] The vehicle control program is executed by a computer, thereby causing the computer to execute: a positioning process for estimating a vehicle position of a vehicle in a parking lot based on a detection result obtained by a sensor mounted on the vehicle; a process for accumulating an estimated error of the vehicle position estimated by the positioning process; a first margin setting process for setting a margin distance to be longer than a default value by a correction value, the correction value reflecting a variance of a distribution of the estimated error; and a vehicle stop process for controlling the vehicle to stop at a margin distance before a designated stop position in the parking lot based on the vehicle position estimated by the positioning process.

[0016] According to the present disclosure, in a vehicle stop process, the vehicle is controlled to stop at a margin distance before a designated stop position. The vehicle stop process is performed based on the vehicle position estimated by a positioning process. A first margin setting process sets the margin distance to be a correction value longer than a default value. The correction value is a parameter that reflects the variance of the distribution of the estimated error of the vehicle position estimated by the positioning process. By using the margin distance set by the first margin setting process, even if the accuracy of the positioning process is assumed to be reduced, the vehicle can be prevented from exceeding the designated stop position. That is, the vehicle can be stopped more reliably in front of the designated stop position. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Hereinafter, features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described with reference to the accompanying drawings, wherein like reference numerals denote like elements, and wherein:

[0018] Figure 1 This is a conceptual diagram for explaining the outline of the automatic valet parking system according to the embodiment of the present disclosure.

[0019] Figure 2 It is a block diagram showing a configuration example of an AVP vehicle according to an embodiment of the present disclosure.

[0020] Figure 3 It is a conceptual diagram for explaining the positioning process performed by the AVP vehicle according to the embodiment of the present disclosure.

[0021] Figure 4 It is a conceptual diagram for explaining the positioning process performed by the AVP vehicle according to the embodiment of the present disclosure.

[0022] Figure 5 This is a conceptual diagram for explaining positioning error.

[0023] Figure 6 This is a conceptual diagram for explaining an outline of a vehicle stop process according to an embodiment of the present disclosure.

[0024] Figure 7 This is a conceptual diagram for explaining an example of setting a remaining distance in a vehicle stop process according to an embodiment of the present disclosure.

[0025] Figure 8 This is a conceptual diagram for explaining another example of setting the remaining distance in the vehicle stop process according to the embodiment of the present disclosure.

[0026] Fig. 9 This is a conceptual diagram for explaining another example of setting the remaining distance in the vehicle stop process according to the embodiment of the present disclosure.

[0027] Fig.10 This is a conceptual diagram for explaining another example of setting the remaining distance in the vehicle stop process according to the embodiment of the present disclosure.

[0028] Fig.11 It is a schematic diagram showing a configuration example of an automatic valet parking system according to an embodiment of the present disclosure.

[0029] Fig.12 This is a block diagram showing a configuration example of a control center according to an embodiment of the present disclosure.

[0030] Fig.13 Detailed Description of the Invention It is a flowchart showing a vehicle stop process according to an embodiment of the present disclosure.

[0031] Fig.14 1 is a flowchart showing a first example of the remaining amount setting process (step S300 ) according to the embodiment of the present disclosure.

[0032] Fig.15 1 is a flowchart showing a second example of the remaining amount setting process (step S300 ) according to the embodiment of the present disclosure.

[0033] Fig.16 3 is a flowchart showing a third example of the remaining amount setting process (step S300 ) according to the embodiment of the present disclosure.

[0034] Fig.17 4 is a flowchart showing a fourth example of the remaining amount setting process (step S300 ) according to the embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] Embodiments of the present disclosure will be described with reference to the drawings.

[0036] 1. Overview of Automated Valet Parking System

[0037] Figure 1 This is a conceptual diagram for explaining the outline of the automatic valet parking system 100 according to the present embodiment. The automatic valet parking system 100 manages an automatic valet parking (AVP: Automated Valet Parking) service in the parking lot 1 .

[0038] Hereinafter, a vehicle that supports automatic valet parking will be referred to as an "AVP vehicle 10". The AVP vehicle 10 can communicate with the automatic valet parking system 100. In addition, the AVP vehicle 10 can travel at least in the parking lot 1 without relying on the driving operation performed by the driver. The AVP vehicle 10 can be an automatic driving vehicle. It should be noted that in the case of realizing automatic valet parking by using an autonomous driving robot to tow a vehicle that does not have an autonomous driving function, the combination of the vehicle and the autonomous driving robot is equivalent to the "AVP vehicle 10", and the control of the AVP vehicle 10 refers to the control of the autonomous driving robot.

[0039] The parking lot 1 is used at least by the AVP vehicle 10. The parking lot 1 may also be used by general vehicles other than the AVP vehicle 10.

[0040] The parking lot 1 includes a boarding and alighting area 2, an aisle 3, and a plurality of parking frames (parking spaces) 4. In the boarding and alighting area 2, an AVP vehicle 10 entering or exiting the parking lot 1 stops. In the boarding and alighting area 2, a passenger gets off or gets on the AVP vehicle 10. The aisle 3 is an area for vehicles such as the AVP vehicle 10 and general vehicles to travel. The parking frame 4 is a space for vehicles such as the AVP vehicle 10 and general vehicles to park. For example, the parking frame 4 is separated by a dividing line.

[0041] Furthermore, a plurality of marks 5 (landmarks) are arranged in the parking lot 1. The marks 5 are used to guide the AVP vehicle 10 in the parking lot 1. As the marks 5, a sign, a pole, etc. can be exemplified.

[0042] Furthermore, a stop node 6 is arranged in the parking lot 1. The stop node 6 indicates a "designated stop position" where the AVP vehicle 10 should stop. For example, the stop node 6 is set at a position adjacent to the boarding and alighting area 2. As another example, the stop node 6 is set at a position adjacent to an intersection in the parking lot 1. The stop node 6 can be represented by a stop line.

[0043] Hereinafter, an example of a flow when a certain user X utilizes the automatic valet parking service will be described. It is assumed that the member information of the user X is registered in advance in the automatic valet parking system 100 .

[0044] First, user X makes a reservation for automatic valet parking. For example, user X operates the terminal device 200 to input the ID information of user X, the desired parking lot 1, the desired use day, the desired use time (desired entry time and desired exit time), etc. The terminal device 200 transmits the reservation information including the input information to the automatic valet parking system 100. The automatic valet parking system 100 performs reservation processing based on the reservation information, and transmits a reservation completion notification to the terminal device 200. In addition, the automatic valet parking system 100 transmits authentication information corresponding to the reservation information to the terminal device 200. The terminal device 200 receives the authentication information and stores the received authentication information.

[0045] The AVP vehicle 10 enters (checks in) the parking lot 1 as follows.

[0046] The AVP vehicle 10 carrying the user X arrives at the boarding and alighting area 2 of the parking lot 1 and stops. In the boarding and alighting area 2, the user X (and other passengers if there are any) get off the AVP vehicle 10. Then, the user X requests the parking of the AVP vehicle 10 using the authentication information held in the terminal device 200. In response to the parking request, the automatic valet parking system 100 authenticates the user X. When the authentication is completed, the operation authority of the AVP vehicle 10 is transferred from the user X to the automatic valet parking system 100. The automatic valet parking system 100 performs the parking process related to the AVP vehicle 10.

[0047] During the parking process, the automatic valet parking system 100 communicates with the AVP vehicle 10 and starts the AVP vehicle 10 (ignition ON).

[0048] In addition, the automatic valet parking system 100 allocates an empty parking frame 4 to the AVP vehicle 10 with reference to the utilization status of the parking lot 1. Then, the automatic valet parking system 100 communicates with the AVP vehicle 10 and provides the AVP vehicle 10 with a parking entry instruction. The parking entry instruction includes information on the allocated parking frame 4 and map information on the parking lot 1. The automatic valet parking system 100 may also specify a driving route from the boarding and alighting area 2 to the allocated parking frame 4. In this case, the parking entry instruction includes information on the designated driving route.

[0049] In response to the parking instruction, the AVP vehicle 10 starts vehicle driving control. Specifically, the AVP vehicle 10 automatically drives from the boarding and alighting area 2 to the assigned parking frame 4 in the aisle 3, and automatically parks in the assigned parking frame 4. At this time, the AVP vehicle 10 can also drive along the driving route specified by the automatic valet parking system 100. The automatic valet parking system 100 can also communicate with the AVP vehicle 10 and remotely control the automatic driving of the AVP vehicle 10.

[0050] When parking is completed, the AVP vehicle 10 notifies the automatic valet parking system 100 of the completion of parking. Alternatively, the automatic valet parking system 100 may also use an infrastructure sensor installed in the parking lot 1 to sense that the parking of the AVP vehicle 10 is completed. After parking is completed, the automatic valet parking system 100 communicates with the AVP vehicle 10 to stop the operation of the AVP vehicle 10 (ignition OFF). The automatic valet parking system 100 associates the information of the parking frame 4 of the AVP vehicle 10 with the user X and maintains it.

[0051] The AVP vehicle 10 is unloaded (checked out) from the parking lot 1 as follows.

[0052] User X uses the terminal device 200 to request the AVP vehicle 10 to leave the garage. The leave request includes authentication information, information of the boarding and alighting area 2 specified by user X, etc. In response to the leave request, the automatic valet parking system 100 authenticates user X and performs a leave process related to the AVP vehicle 10.

[0053] In the exit process, the automatic valet parking system 100 communicates with the AVP vehicle 10 and starts the AVP vehicle 10 (ignition is turned on).

[0054] In addition, the automatic valet parking system 100 communicates with the AVP vehicle 10 and provides the AVP vehicle 10 with a departure instruction. The departure instruction includes information about the boarding and alighting area 2 specified by the user X and map information about the parking lot 1. The automatic valet parking system 100 may also specify a driving route from the parking frame 4 to the specified boarding and alighting area 2. In this case, the departure instruction includes information about the specified driving route.

[0055] In response to the exit instruction, the AVP vehicle 10 starts vehicle driving control. Specifically, the AVP vehicle 10 automatically drives from the parking frame 4 to the designated boarding and alighting area 2 in the aisle 3. At this time, the AVP vehicle 10 can also drive along the driving route designated by the automatic valet parking system 100. The automatic valet parking system 100 can also communicate with the AVP vehicle 10 to remotely control the automatic driving of the AVP vehicle 10.

[0056] The AVP vehicle 10 arrives at the boarding and alighting area 2 designated by the user X and stops. The operating authority of the AVP vehicle 10 is transferred from the automatic valet parking system 100 to the user X. The user X (and other passengers if any) board the AVP vehicle 10. The AVP vehicle 10 departs toward the next destination.

[0057] 2.AVP Vehicle

[0058] 2-1. Example of configuration

[0059] Figure 2 1 is a block diagram showing a configuration example of the AVP vehicle 10 according to the present embodiment. The AVP vehicle 10 includes a vehicle state sensor 20 , an identification sensor 30 , a communication device 40 , a travel device 50 , and a control device 60 .

[0060] The vehicle state sensor 20 detects the state of the AVP vehicle 10. Examples of the vehicle state sensor 20 include a vehicle speed sensor (wheel speed sensor), a steering angle sensor, a yaw rate sensor, and a lateral acceleration sensor.

[0061] The recognition sensor 30 recognizes (detects) the situation around the AVP vehicle 10. Examples of the recognition sensor 30 include a camera, a laser radar (LIDAR: Laser Imaging Detection and Ranging), and a radar.

[0062] The communication device 40 communicates with the outside of the AVP vehicle 10. For example, the communication device 40 communicates with the automatic valet parking system 100.

[0063] The travel device 50 includes a steering device, a drive device, and a brake device. The steering device steers the wheels of the AVP vehicle 10. For example, the steering device includes a power steering (EPS: Electric Power Steering) 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 brake device generates a braking force.

[0064] The control device 60 is a computer that controls the AVP vehicle 10. Specifically, the control device 60 includes one or more processors 70 (hereinafter referred to as simply the processor 70) and one or more storage devices 80 (hereinafter referred to as simply the storage device 80). The processor 70 performs various processes. The storage device 80 stores various information. Examples of the storage device 80 include a volatile memory, a non-volatile memory, a HDD (Hard Disk Drive), an SSD (Solid State Drive), and the like. The control device 60 (processor 70) executes a "vehicle control program" which is a computer program, thereby implementing various processes performed by the control device 60. The vehicle control program is stored in the storage device 80 or recorded in a computer-readable recording medium.

[0065] 2-2. Information acquisition processing

[0066] The processor 70 (control device 60 ) executes information acquisition processing for acquiring various information. The various information includes parking lot map information MAP, vehicle status information STA, surrounding situation information SUR, etc. The acquired information is stored in the storage device 80 .

[0067] The parking lot map information MAP is map information of the parking lot 1. Specifically, the parking lot map information MAP indicates the arrangement of the boarding and alighting areas 2, aisles 3, parking frames 4, marks 5, stop nodes 6 (designated stop positions), etc. in the parking lot 1. The parking lot map information MAP is provided by the automatic valet parking system 100. The processor 70 acquires the parking lot map information MAP from the automatic valet parking system 100 via the communication device 40.

[0068] The vehicle state information STA is information indicating the state of the AVP vehicle 10, and indicates the detection result obtained by the vehicle state sensor 20. Examples of the state of the AVP vehicle 10 include vehicle speed (wheel speed), steering angle (wheel steering angle), yaw rate, lateral acceleration, etc. The processor 70 obtains the vehicle state information STA from the vehicle state sensor 20.

[0069] The surrounding condition information SUR is information indicating the condition of the surroundings of the AVP vehicle 10, and indicates the recognition result obtained by the recognition sensor 30. For example, the surrounding condition information SUR includes image information captured by a camera. As another example, the surrounding condition information SUR includes measurement information indicating the measurement result obtained by a laser radar or a radar. In addition, the surrounding condition information SUR includes object information related to objects around the AVP vehicle 10. As the objects around the AVP vehicle 10, examples include aisles 3, parking frames 4, signs 5, white lines, other vehicles, etc. The object information includes the relative position and relative speed of the object relative to the AVP vehicle 10 (recognition sensor 30). Objects around the AVP vehicle 10 can be identified based on at least one of the image information and the measurement information, and the relative position and relative speed of the identified objects can also be calculated. In this way, the processor 70 acquires the surrounding condition information SUR based on the recognition result obtained by the recognition sensor 30.

[0070] 2-3. Positioning processing

[0071] The processor 70 (control device 60) performs a "positioning process (self-position estimation process, Localization)" to estimate the position of the AVP vehicle 10 in the parking lot 1. Hereinafter, the position of the AVP vehicle 10 in the parking lot 1 is referred to as the "vehicle position P". In the positioning process, the above-mentioned vehicle state sensor 20 and identification sensor 30 are used.

[0072] Figure 3: is a conceptual diagram for explaining the positioning process. The vehicle coordinate system is a relative coordinate system that is fixed to the AVP vehicle 10 and changes as the AVP vehicle 10 moves. The position of the origin O of the vehicle coordinate system in the AVP vehicle 10 is arbitrary. The vehicle position P can also be said to be the position of the origin O in the parking lot 1. The calibration parameter E represents the installation position and installation direction of the recognition sensor 30 in the vehicle coordinate system. That is, the calibration parameter E represents the translation and rotation movement amount of the recognition sensor 30 relative to the origin O.

[0073] The positioning process estimates the vehicle position P at a certain period. For convenience, the time t of the estimation object is k and the vehicle position P k They are respectively called “object time t k ” and “target vehicle position P k In order to estimate the target time t k The target vehicle position P k , using the previous moment t k－1 The vehicle position P k－1 For convenience, the previous moment t k－1 and the vehicle position P k－1 are respectively called “reference time t k－1 ” and “reference vehicle position P k－1 ”.

[0074] Vehicle movement amount T k－1，k is the reference time t k－1 With the object time t k The amount of movement of the AVP vehicle 10 between k－1，k It can be calculated based on the vehicle state information STA obtained by the vehicle state sensor 20. For example, the vehicle movement amount T k－1，k Based on the reference time t k－1 With the object time t k It is calculated based on the history of wheel speed and steering angle detected during the period.

[0075] Furthermore, by using the recognition sensor 30, one or more markers 5 existing around the AVP vehicle 10 can be recognized. j . Parking lot 1, sign 5 j The absolute position F j is known and obtained from the parking lot map information MAP. k Sign 5 j Relative position R with respect to the recognition sensor 30 j，k Obtained from the surrounding situation information SUR.

[0076] Figure 4 It is used to describe the object time t kThe "internal estimated vehicle position PI k ” and “External estimated vehicle position PO j，k Conceptual diagram of the internal estimated vehicle position PI k is based on the reference time t k－1 The reference vehicle position P k－1 and vehicle movement T k－1，k The calculated (estimated) target vehicle position P k On the other hand, the external estimated vehicle position PO j，k According to the sign 5 j The absolute position F j 、Mark 5 j Relative position R with respect to the recognition sensor 30 j，k and the target vehicle position P calculated (estimated) by the calibration parameter E k .

[0077] Internal estimated vehicle position PI k and one or more signs 5 j Related external estimated vehicle position PO j，k Therefore, by using the internal estimated vehicle position PI k and one or more signs 5 j Related external estimated vehicle position PO j，k Combination to determine the object time t k The object vehicle position P k For example, by using and and one or more signs 5 j Related external estimated vehicle position PO j，k The best matching method corrects the internal estimated vehicle position PI k To determine the target vehicle position P k In other words, by using the internal estimated vehicle position PI k and the external estimated vehicle position PO j，k The evaluation function is used to convert the object vehicle position P k Optimization: It should be noted that the evaluation function and optimization method are well known and are not particularly limited in this embodiment.

[0078] The estimated target vehicle position P k is used as the next time t k+1 The vehicle position P of the AVP vehicle 10 is continuously estimated by repeating the above process.

[0079] The positioning information LOC indicates the vehicle position P estimated by the positioning process. As described above, the processor 70 performs the positioning process based on the parking lot map information MAP, the vehicle state information STA, and the surrounding situation information SUR, and acquires the positioning information LOC. The positioning information LOC is stored in the storage device 80.

[0080] Figure 5 This is a conceptual diagram for explaining positioning error. k AVP vehicle 10 with mark 5 j The distance between them, considering the estimated distance DE j，k and measuring distance DM j，k These two types. Estimated distance DE j，k is the vehicle position P estimated by positioning processing k And mark 5 j The absolute position F j On the other hand, the distance DM is measured j，k is the distance obtained by the recognition result obtained by the recognition sensor 30, according to the mark 5 j The relative position R j，k and calibration parameter E. For the measured distance DM j，k , assuming that the identification sensor 30 and the marker 5 j The measurement error σ corresponding to the type j As the measurement error σ j , you can also set it to be independent of mark 5 j A constant value of Figure 5 As an example, the estimated distances DE corresponding to the two marks 51 and 52 (j=1, 2) are shown. j，k and measuring distance DM j，k Positioning error ε k represents the vehicle position P estimated by positioning processing k The estimated error of the positioning error ε k For example, it is represented by the following equation (1).

[0081] [Formula 1]

[0082]

[0083] The processor 70 continuously performs positioning processing and accumulates the positioning error ε k The positioning error information ERR indicates the accumulated positioning error ε k The positioning error information ERR is stored in the storage device 80. The processor 70 updates the positioning error information ERR while performing positioning processing.

[0084] 2-4. Communication processing

[0085] The processor 70 (control device 60) communicates with the automatic valet parking system 100 via the communication device 40. For example, the processor 70 receives parking lot map information MAP, parking entry guidance information, parking exit guidance information, etc. from the automatic valet parking system 100. In addition, the processor 70 may also periodically send vehicle status information STA and positioning information LOC to the automatic valet parking system 100. The processor 70 may also send positioning error information ERR to the automatic valet parking system 100.

[0086] 2-5. Vehicle driving control

[0087] The processor 70 (control device 60) performs "vehicle driving control" that controls the driving of the AVP vehicle 10 independently of the driver's driving operation. The vehicle driving control includes steering control, acceleration control, and braking control. The processor 70 performs vehicle driving control by controlling the driving device 50. Specifically, the processor 70 performs steering control by controlling the steering device. In addition, the processor 70 performs acceleration control by controlling the drive device. In addition, the processor 70 performs braking control by controlling the braking device.

[0088] In the parking lot 1, the processor 70 performs vehicle travel control so that the AVP vehicle 10 autonomously travels. Specifically, the processor 70 grasps the map of the parking lot 1 and the vehicle position P in the parking lot 1 based on the parking lot map information MAP and the positioning information LOC. Then, the processor 70 performs vehicle travel control in such a manner that the AVP vehicle 10 automatically travels toward the destination.

[0089] For example, in the case of the above-mentioned warehousing process, the departure point is the boarding and alighting area 2, and the destination is the assigned parking frame 4. The processor 70 performs vehicle driving control in a manner that the AVP vehicle 10 automatically drives from the boarding and alighting area 2 to the assigned parking frame 4, and automatically parks in the assigned parking frame 4. The position of the assigned parking frame 4 is obtained from the parking lot map information MAP. The processor 70 may also perform vehicle driving control in a manner that the AVP vehicle 10 drives along the driving route specified by the automatic valet parking system 100. When the AVP vehicle 10 is parked in the parking frame 4, the parking conditions of the parking frame 4 and the surroundings may be grasped by referring to the surrounding condition information SUR. The vehicle driving control may also be performed in a manner that avoids collision with other vehicles or structures by referring to the surrounding condition information SUR.

[0090] In the case of the unloading process, the departure point is the assigned parking frame 4 and the destination is the boarding and alighting area 2. The processor 70 performs vehicle travel control so that the AVP vehicle 10 automatically travels from the parking frame 4 to the boarding and alighting area 2 and automatically stops in the boarding and alighting area 2.

[0091] 3. Vehicle stop processing

[0092] Figure 6 1 is a conceptual diagram for explaining the outline of the "vehicle stop process" of the present embodiment. When the AVP vehicle 10 needs to be stopped during automatic driving in the parking lot 1, the vehicle stop process is performed. Specifically, the automatic valet parking system 100 specifies a stop node 6 and instructs the AVP vehicle 10 to stop in front of the specified stop node 6 (specified stop position).

[0093] The processor 70 of the AVP vehicle 10 receives a stop instruction from the automatic valet parking system 100 via the communication device 40. In response to the stop instruction, the processor 70 performs vehicle travel control (brake control) so as to stop in front of the designated stop node 6. The position of the stop node 6 in the parking lot 1 is obtained from the parking lot map information MAP. The processor 70 performs brake control based on the vehicle position P estimated by the positioning process.

[0094] More specifically, the processor 70 performs braking control so that the AVP vehicle 10 stops at the "remaining distance M" before the designated stop node 6. To this end, the processor 70 sets the position at the remaining distance M before the stop node 6 as the target stop position PT. Then, the processor 70 performs braking control so that the AVP vehicle 10 stops at the target stop position PT based on the vehicle position P estimated by the positioning process. The target deceleration in the braking control can be predetermined.

[0095] Here, the possibility of a decrease in the accuracy of the positioning process is considered. For example, when an abnormality occurs in at least one of the vehicle state sensor 20 and the recognition sensor 30, the accuracy of the positioning process may decrease. As another example, the recognition accuracy of the mark 5 may decrease depending on the time period and weather, which may lead to a decrease in the accuracy of the positioning process. When the accuracy of the positioning process decreases, the accuracy of the vehicle position P used in the vehicle driving control will also decrease. When the accuracy of the vehicle position P decreases, the actual stop position where the AVP vehicle 10 actually stops will deviate from the target stop position PT.

[0096] Consider the case where a default value α that is set to be independent of the accuracy of the positioning process is used as the above-mentioned margin distance M. The default value α may be a fixed value or a variable value. When the default value α is used, if the accuracy of the positioning process is low, the AVP vehicle 10 may not stop in front of the stop node 6 but may pass the stop node 6. This means that the reliability of the vehicle stop process is reduced, so it is not preferred.

[0097] In order to more reliably stop the AVP vehicle 10 before the stop node 6, it is preferable to set the margin distance M in consideration of the possibility of a decrease in the accuracy of the positioning process. Hereinafter, the margin distance M set in consideration of the possibility of a decrease in the accuracy of the positioning process is referred to as a "first margin distance M1".

[0098] The first margin distance M1 is represented by the sum of the default value α and the correction value β, that is, "α+β". That is, the first margin distance M1 is longer than the default value α by the correction value β. The default value α is a value greater than 0. The correction value β reflects the positioning error ε mentioned above. k The variance σL of the distribution of is the parameter. For example, the correction value β is the positioning error ε k The variance of the distribution of σL itself. As another example, the correction value β can also be a real multiple of the variance σL. The positioning error ε k The variance σL of the distribution is obtained from the positioning error information ERR. By using such a first margin distance M1 to perform vehicle stopping processing, even if the accuracy of the positioning processing is assumed to be reduced, the AVP vehicle 10 can be prevented from exceeding the stop node 6. That is, the AVP vehicle 10 can be stopped in front of the stop node 6 more reliably.

[0099] For comparison, the margin distance M set without considering the possibility of a decrease in the accuracy of the positioning process is hereinafter referred to as a “second margin distance M2.” The second margin distance M2 is a default value α.

[0100] The first margin distance M1 and the second margin distance M2 may be selected (used separately) according to the situation. For example, when the illumination of the environment around the AVP vehicle 10 is low, the recognition accuracy of the mark 5 may decrease, and the accuracy of the positioning processing may decrease. Therefore, when the illumination is less than a specified value, the first margin distance M1 may be selected.

[0101] As another example, consider the following scenario: if the AVP vehicle 10 does not stop and passes the stop node 6, there is a possibility that a collision with a pedestrian or another vehicle may occur. For convenience, such a stop node 6 is referred to as the "first stop node 6-1". In the case of the first stop node 6-1, it is particularly desirable to stop the AVP vehicle 10 before the first stop node 6-1. Therefore, as the margin distance M for the first stop node 6-1, the first margin distance M1 is preferably selected.

[0102] Figure 7 This is a conceptual diagram for explaining an example of the first stop node 6 - 1 . Figure 7The first stop node 6-1 shown is a stop node 6 that exists before the boarding and alighting area 2. In other words, the first stop node 6-1 is a stop node adjacent to the boarding and alighting area 2. In the boarding and alighting area 2, a user gets off the AVP vehicle 10 or a user gets on the AVP vehicle 10. In order to avoid collision with the user, the margin distance M is preferably set to the first margin distance M1.

[0103] Figure 8 It is a conceptual diagram for explaining another example of the first stop node 6 - 1 . Figure 8 The first stop node 6-1 shown is a stop node 6 existing before the parking frame 4. For example, the first stop node 6-1 is a stop node 6 existing before the parking frame 4 assigned to another vehicle Q that is leaving the garage or is scheduled to leave the garage in the near future. In order to avoid collision with another vehicle Q, the margin distance M is preferably set to the first margin distance M1.

[0104] Fig. 9 This is a conceptual diagram for explaining another example of the first stop node 6 - 1 . Fig. 9 The first stop node 6-1 shown is a stop node 6 existing before the intersection 7. In other words, the first stop node 6-1 is a stop node adjacent to the intersection 7. Another vehicle Q may exist at the intersection 7. To avoid collision with another vehicle Q, the margin distance M is preferably set to the first margin distance M1.

[0105] Fig.10 This is a conceptual diagram for explaining another example of the first stop node 6 - 1 . Fig.10 The first stop node 6-1 shown is a stop node 6 adjacent to a dangerous area 8. The dangerous area 8 is an area where pedestrians, operators, and the like may exist. In order to avoid collision with people, the margin distance M is preferably set to a first margin distance M1.

[0106] As described above, in the vehicle stop processing of the present embodiment, the vehicle driving control is performed in such a manner that the AVP vehicle 10 stops at the margin distance M before the stop node 6. The vehicle driving control is performed based on the vehicle position P estimated by the positioning processing. The first margin distance M1 is represented by the sum (α+β) of the default value α and the correction value β. The correction value β is a value that reflects the estimation error (positioning error ε) of the vehicle position P estimated by the positioning processing. k ) is a parameter of the variance σL of the distribution of . By using such a first margin distance M1 as the margin distance M, even if the accuracy of the positioning processing is assumed to be reduced, the AVP vehicle 10 can be prevented from exceeding the stop node 6. That is, the AVP vehicle 10 can be stopped in front of the stop node 6 more reliably.

[0107] 4. Example of the configuration of an automatic valet parking system

[0108] Fig.11 1 is a schematic diagram showing a configuration example of an automatic valet parking system 100 according to the present embodiment. The automatic valet parking system 100 includes a vehicle management center 110, a control center 120, and an infrastructure sensor 130. The control center 120 is provided in each parking lot 1. Therefore, there are the same number of control centers 120 as the parking lots 1. The vehicle management center 110 manages all the control centers 120. The infrastructure sensor 130 is provided in the parking lot 1 and recognizes the conditions in the parking lot 1. For example, the infrastructure sensor 130 includes an infrastructure camera that captures the conditions in the parking lot 1.

[0109] Fig.12 1 is a block diagram showing a configuration example of the control center 120 according to the present embodiment. The control center 120 includes an information processing device 121 (control device). The information processing device 121 manages the AVP vehicles 10 in the parking lot 1 and controls the AVP vehicles 10 in the parking lot 1 .

[0110] The information processing device 121 includes one or more processors 122 (hereinafter referred to as processors 122) and one or more storage devices 123 (hereinafter referred to as storage devices 123). The processor 122 performs various processes. The storage device 123 stores various information. Examples of the storage device 123 include volatile memory, nonvolatile memory, HDD, SSD, etc. The information processing device 121 (processor 122) executes a control program as a computer program, thereby realizing various processes performed by the information processing device 121. The control program is stored in the storage device 123 or recorded in a computer-readable recording medium.

[0111] The information processing device 121 further includes a communication device 124. The communication device 124 communicates with the vehicle management center 110, the AVP vehicle 10, and the infrastructure sensor 130 via a communication network.

[0112] The parking lot map information MAP is prepared in advance and stored in the storage device 123 of the control center 120 . The processor 122 communicates with the AVP vehicle 10 via the communication device 124 , and provides the AVP vehicle 10 with the parking lot map information MAP.

[0113] Positioning error information ERR indicates positioning error ε k For example, the processor 122 receives the positioning error information ERR from the AVP vehicle 10 via the communication device 124. As another example, the processor 122 may also receive the positioning information LOC and the surrounding condition information SUR from the AVP vehicle 10 via the communication device 124. In this case, the processor 122 performs positioning error ε based on the received positioning information LOC and the surrounding condition information SUR.k The positioning error information ERR is generated and updated by calculating and accumulating the positioning error information ERR. The positioning error information ERR is stored in the storage device 123.

[0114] The default remaining amount information DEF indicates a default value α of the remaining amount distance M. The default remaining amount information DEF is prepared in advance and stored in the storage device 123 . The processor 122 transmits the default remaining amount information DEF to the AVP vehicle 10 via the communication device 124 . In this case, the default remaining amount information DEF is also stored in the storage device 80 of the AVP vehicle 10 .

[0115] 5. Processing flow of vehicle stop processing

[0116] Fig.13 : is a flowchart showing the vehicle stop processing according to the present embodiment.

[0117] In step S100, the processor 122 of the control center 120 determines whether a stop instruction is required for the AVP vehicle 10 traveling in the parking lot 1. For example, the processor 122 receives the positioning information LOC from the AVP vehicle 10 via the communication device 124. Then, the processor 122 determines whether the AVP vehicle 10 has entered a range at a certain distance from the stop node 6 on the driving route based on the positioning information LOC and the parking lot map information MAP. In the case where the AVP vehicle 10 has entered a range at a certain distance from the stop node 6 on the driving route, the processor 122 determines that a stop instruction is required for the AVP vehicle 10. In the case where a stop instruction is required (step S100; yes), the process proceeds to step S200.

[0118] In step S200, the processor 122 transmits a stop instruction to the AVP vehicle 10 via the communication device 124. The stop instruction specifies at least the stop node 6 and instructs the AVP vehicle 10 to stop before the specified stop node 6 (specified stop position).

[0119] In step S300, a "remaining distance M" is set by performing a "remaining distance setting process". Various examples of remaining distance setting processes will be described later. It should be noted that the remaining distance setting process may be performed by the processor 70 of the AVP vehicle 10 or by the processor 122 of the control center 120. In the case where the remaining distance setting process is performed by the processor 122 of the control center 120, the remaining distance setting process may be performed before step S200. In this case, the stop instruction may also include the remaining distance M set by the processor 122.

[0120] In step S400, the processor 70 of the AVP vehicle 10 performs vehicle travel control (brake control) so as to stop the vehicle before the designated stop node 6. More specifically, the processor 70 performs brake control so as to stop the vehicle at a remaining distance M before the stop node 6. To this end, the processor 70 sets the position at the remaining distance M before the stop node 6 as the target stop position PT (refer to Figure 6 ). The position of the stop node 6 in the parking lot 1 is obtained from the parking lot map information MAP. The remaining distance M is set in the above-mentioned step S300. The processor 70 performs braking control in such a manner that the AVP vehicle 10 stops at the target stop position PT based on the vehicle position P estimated by the positioning process. The target deceleration in the braking control can be predetermined.

[0121] Various examples of the remaining amount setting process (step S300 ) will be described below.

[0122] 5-1. First example

[0123] Fig.14 3 is a flowchart showing a first example of the remaining amount setting process (step S300 ). In the first example, the remaining amount setting process includes a “selection process” of selecting, as the remaining amount distance M, either the first remaining amount distance M1 or the second remaining amount distance M2 .

[0124] More specifically, in step S310, the processor 70 of the AVP vehicle 10 or the processor 122 of the control center 120 (hereinafter referred to as "processor") determines whether the stop node 6 is the first stop node 6-1. The first stop node 6-1 is a stop node 6 associated with the following scenario: if the AVP vehicle 10 does not stop and passes the stop node 6, there is a possibility of collision with pedestrians or other vehicles. For example, the first stop node 6-1 includes Figure 7 to Figure 10 The processor may determine whether the stop node 6 is the first stop node 6-1 based on the parking lot map information MAP.

[0125] When the designated stop node 6 is the first stop node 6-1 (step S310; yes), the process proceeds to step S330. In step S330, the processor selects the first margin distance M1 and sets the first margin distance M1 as the margin distance M. The first margin distance M1 is the sum (α+β) of the default value α and the correction value β. That is, the first margin distance M1 is longer than the default value α by the correction value β. The default value α is obtained from the default margin information DEF. The correction value β reflects the positioning error ε. k The variance σL of the distribution of is the parameter. For example, the correction value β is the positioning error ε kThe variance of the distribution of σL itself. As another example, the correction value β can also be a real multiple of the variance σL. The positioning error ε k The variance σL of the distribution is obtained from the positioning error information ERR.

[0126] On the other hand, when the designated stop node 6 is not the first stop node 6 - 1 (step S310 ; No), the process proceeds to step S340 . In step S340 , the processor selects the second margin distance M2 and sets the second margin distance M2 as the margin distance M. The second margin distance M2 is a default value α.

[0127] Hereinafter, the process of step S330 is referred to as "first margin setting process". On the other hand, the process of step S340 is referred to as "second margin setting process". When the designated stop node 6 is the first stop node 6-1 (step S310; yes), the processor selects the first margin setting process to set the margin distance M. On the other hand, when the designated stop node 6 is not the first stop node 6-1 (step S310; no), the processor selects the second margin setting process to set the margin distance M.

[0128] 5-2. Second example

[0129] Fig.15 3 is a flowchart showing a second example of the remaining amount setting process (step S300 ). In the second example, the remaining amount setting process also includes a “selection process” of selecting, as the remaining amount distance M, either the first remaining amount distance M1 or the second remaining amount distance M2 .

[0130] In more detail, in step S320, the processor determines whether the illumination of the environment around the AVP vehicle 10 is less than a specified value. For example, the identification sensor 30 of the AVP vehicle 10 includes an illumination sensor. As another example, the infrastructure sensor 130 provided in the parking lot 1 includes an illumination sensor. As yet another example, the illumination is calculated based on an image captured by a camera included in the identification sensor 30 of the AVP vehicle 10. As yet another example, the illumination is calculated based on an image captured by an infrastructure camera included in the infrastructure sensor 130.

[0131] When the illuminance is less than the predetermined value (step S320; Yes), the process proceeds to step S330 (first margin setting process). The first margin distance M1 is set as the margin distance M by the first margin setting process.

[0132] On the other hand, when the illuminance is equal to or greater than the predetermined value (step S320 ; No), the process proceeds to step S340 (second margin setting process). The second margin distance M2 is set as the margin distance M by the second margin setting process.

[0133] 5-3. The third example

[0134] Fig.16 2 is a flowchart showing a third example of the remaining amount setting process (step S300). The third example is a combination of the first example and the second example.

[0135] When the designated stop node 6 is the first stop node 6-1 (step S310; Yes), the process proceeds to step S330 (first margin setting process). In other cases (step S310; No), the process proceeds to step S320. When the illuminance is less than the prescribed value (step S320; Yes), the process proceeds to step S330 (first margin setting process). On the other hand, when the illuminance is greater than the prescribed value (step S320; No), the process proceeds to step S340 (second margin setting process).

[0136] 5-4. Fourth Example

[0137] Fig.17 2 is a flowchart showing a fourth example of the remaining amount setting process (step S300). In the fourth example, step S330 (first remaining amount setting process) is always performed, and the first remaining amount distance M1 is set as the remaining amount distance M.

[0138] 6. Vehicle control system

[0139] The control device 60 of the AVP vehicle 10 (see Figure 2 ) and the information processing device 121 of the control center 120 (refer to Fig.12 ) constitutes a "vehicle control system" for controlling the AVP vehicle 10. The vehicle control system includes one or more processors (70, 122) and one or more storage devices (80, 123). The one or more processors perform positioning processing to estimate the vehicle position P of the AVP vehicle 10. Positioning error information ERR is stored in the one or more storage devices. In addition, the one or more processors perform positioning processing to estimate the vehicle position P of the AVP vehicle 10. Figure 13 to Figure 17 The process shown in .

Claims

1. A vehicle control method for controlling a vehicle supporting automatic valet parking in a parking lot, the vehicle control method comprising: Positioning processing for estimating a vehicle position of the vehicle in the parking lot based on a detection result obtained by a sensor mounted on the vehicle; a process of accumulating an estimation error of the vehicle position estimated by the positioning process; a first margin setting process of setting the margin distance to be longer than a default value by a correction value, the correction value reflecting a variance of a distribution of the estimated error; as well as The vehicle stopping process controls the vehicle to stop at the remaining distance before the designated stopping position in the parking lot based on the vehicle position estimated by the positioning process.

2. The vehicle control method according to claim 1, further comprising: A second margin setting process, setting the default value as the margin distance; as well as A selection process selects one of the first margin setting process and the second margin setting process to set the margin distance.

3. The vehicle control method according to claim 2, wherein: The selection process includes the following processes: selecting the first margin setting process to set the margin distance when the designated stop position is a first designated stop position; and If the designated stop position is not the first designated stop position, the second margin setting process is selected to set the margin distance.

4. The vehicle control method according to claim 2, wherein: The selection process includes the following processes: In a case where the designated stop position is a first designated stop position, selecting the first margin setting process to set the margin distance; When the designated stop position is not the first designated stop position and the illumination of the environment around the vehicle is less than a predetermined value, selecting the first margin setting process to set the margin distance; as well as When the designated stop position is not the first designated stop position and the illuminance is equal to or greater than the predetermined value, the second margin setting process is selected to set the margin distance.

5. The vehicle control method according to claim 3 or 4, wherein: The first designated stop position includes at least one of the designated stop position adjacent to a boarding and alighting area, the designated stop position in front of a parking frame, and the designated stop position adjacent to an intersection.

6. The vehicle control method according to claim 2, wherein: The selection process includes the following processes: When the illumination of the environment around the vehicle is less than a predetermined value, selecting the first margin setting process to set the margin distance; and When the illuminance is equal to or greater than the predetermined value, the second margin setting process is selected to set the margin distance.

7. A vehicle control system for controlling a vehicle supporting automatic valet parking in a parking lot, the vehicle control system comprising: one or more processors for performing positioning processing, the positioning processing being processing for estimating a vehicle position of the vehicle in the parking lot based on a detection result obtained by a sensor mounted on the vehicle; as well as one or more storage devices storing positioning error information indicating a distribution of an estimated error of the vehicle position estimated by the positioning process, The one or more processors also perform: a first margin setting process of setting the margin distance to be longer than a default value by a correction value, the correction value reflecting a variance of a distribution of the estimated error; as well as The vehicle stopping process controls the vehicle to stop at the remaining distance before the designated stopping position in the parking lot based on the vehicle position estimated by the positioning process.

8. A recording medium having a vehicle control program recorded thereon, the vehicle control program controlling a vehicle supporting automatic valet parking in a parking lot, The vehicle control program is executed by a computer, thereby causing the computer to execute: Positioning processing for estimating a vehicle position of the vehicle in the parking lot based on a detection result obtained by a sensor mounted on the vehicle; a process of accumulating an estimation error of the vehicle position estimated by the positioning process; a first margin setting process of setting the margin distance to be longer than a default value by a correction value, the correction value reflecting a variance of a distribution of the estimated error; as well as The vehicle stopping process controls the vehicle to stop at the remaining distance before the designated stopping position in the parking lot based on the vehicle position estimated by the positioning process.

Images