A method, apparatus, device, and medium for parking autonomous vehicles in underground garages.

Abstract

This invention discloses a method, apparatus, device, and medium for parking autonomous vehicles in underground garages. The method includes: recording the starting position information of the vehicle when it enters the garage; determining the real-time position information of the vehicle relative to the starting position information while it is driving in the garage; determining, based on the real-time position information, whether the vehicle has entered the geofence corresponding to the starting point of a preset memory parking map, wherein the preset memory parking map is a trained parking map corresponding to the current garage, and the geofence is a geographical area with the starting point of the preset memory parking map as the center and a preset length as the radius; if it is determined that the vehicle has entered the geofence, the preset memory parking map corresponding to the starting point is used as the current target parking map, and the vehicle is driven according to the current target parking map to perform parking. By adopting the above technical solution, the problem of low success rate of memory parking is improved.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and more specifically, to a method, apparatus, device, and medium for parking autonomous vehicles in underground garages. Background Technology

[0002] Automated memory parking is an advanced car parking technology that allows a vehicle to automatically complete the entire parking process without human intervention.

[0003] Memory parking technology typically includes a learning phase and an application phase. During the learning phase, the driver manually operates the vehicle to park the first time. During this process, the vehicle collects information about the surrounding environment, including the size of the parking space and any obstacles around the vehicle. In the application phase, the driver can activate the memory parking function, and the vehicle automatically senses the surrounding environment and uses the previously learned information to perform the parking operation. However, when two locations in a parking garage have very similar environments—for example, in a multi-level parking garage where the surrounding environment information of location A on the first basement level and location B on the second basement level is the same—the autonomous vehicle may encounter location positioning errors. For instance, it might mistake location B on the second basement level for location A on the first basement level, or vice versa. Such location positioning errors can lead to a low success rate for subsequent memory parking operations. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and medium for parking autonomous vehicles in underground garages to solve the problem of low success rate of memory parking caused by location errors of autonomous vehicles.

[0005] The specific technical solution is as follows:

[0006] In a first aspect, embodiments of the present invention provide a method for parking an autonomous vehicle in an underground garage, comprising:

[0007] Record the starting position information of the vehicle when it enters the underground parking garage;

[0008] When the vehicle is currently moving in the underground parking garage, determine the real-time position information of the vehicle relative to the starting point position information;

[0009] Based on real-time location information, it is determined whether the current vehicle has entered the geofence corresponding to the starting point of the preset memory parking map. The preset memory parking map is a parking map that has been trained and corresponds to the current garage. The starting point of different preset memory parking maps is different. The geofence is a geographical area with the starting point of the preset memory parking map as the center and a preset length as the radius.

[0010] If it is determined that the current vehicle has entered the geofence, the preset memory parking map corresponding to the map starting point will be used as the current target parking map, and the vehicle will drive according to the current target parking map to park.

[0011] As can be seen from the above scheme, by determining the real-time position information of the vehicle relative to the starting point of entering the underground parking garage, and based on the positional relationship between this real-time position information and the starting point of the preset memory parking map, if it is determined that the current autonomous vehicle has entered the geofence of the starting point of a preset memory parking map, then the preset memory parking map corresponding to the starting point is used as the target parking map, and parking is performed according to the target parking map. This setting solves the problem of two locations being mispositioned due to the same surrounding environmental information in the underground parking garage scenario, and also solves the problem that the autonomous vehicle cannot determine the current target parking map when there are multiple preset memory parking maps in the underground parking garage, effectively improving the success rate of autonomous vehicles parking in underground parking garages.

[0012] Optionally, record the starting position information of the vehicle when it enters the underground parking garage, including:

[0013] Based on the vehicle pitch angle and / or the visual image texture features collected by the sensors currently configured in the vehicle, identify whether the current vehicle has entered the underground parking garage;

[0014] Once the vehicle is identified as entering the underground parking garage, its starting position information is recorded.

[0015] Optionally, determine the real-time location information of the current vehicle relative to the starting point location information, including:

[0016] The real-time position information of the current vehicle relative to the starting point position information is determined by any one of the following sensor data, or by fusing the data from various sensors, wherein each sensor includes an inertial measurement unit (IMU), an image sensor, radar, and an odometer.

[0017] Optionally, the method provided in this embodiment of the invention further includes:

[0018] While the vehicle is moving through the underground parking garage, save the real-time position information of the vehicle relative to the starting point.

[0019] When the current vehicle is turned off and powered down, the current vehicle's location information at the time of power failure is obtained as historical location information;

[0020] When the vehicle is powered on again, the current location information is calculated based on the historical location information and used as the real-time location information after power-on.

[0021] As can be seen from the above technical solution, under the premise that the current autonomous vehicle does not leave the underground parking garage, it can still determine whether it has entered the geofence of the starting point of a preset memory parking map based on the vehicle's real-time location information, without having to drive out of the underground parking garage to obtain a GPS signal to continue using the memory parking function, which effectively improves the efficiency of automatic parking in underground parking garage scenarios.

[0022] Optionally, the method provided in this embodiment of the invention further includes:

[0023] If a user-triggered memory parking map creation command is received and the current parking garage does not have a preset memory parking map, a memory parking map will be created based on the current vehicle's driving trajectory in manual driving mode.

[0024] Secondly, embodiments of the present invention also provide a parking device for an autonomous vehicle in an underground garage, comprising:

[0025] The starting position recording module is configured to record the starting position information of the current vehicle when it enters the underground parking garage;

[0026] The real-time location determination module is configured to determine the real-time location information of the current vehicle relative to the starting point when the vehicle is driving in the underground parking garage.

[0027] The map starting point determination module is configured to determine whether the current vehicle has entered the geofence corresponding to the map starting point of the preset memory parking map based on real-time location information. The preset memory parking map is a parking map that has been trained and corresponds to the current garage. The starting point of different preset memory parking maps is different. The geofence is a geographical area with the map starting point of the preset memory parking map as the center and a preset length as the radius.

[0028] The target parking map determination module is configured to, if it determines that the current vehicle has entered the geofence, use the preset memory parking map corresponding to the map starting point as the current target parking map, and drive according to the current target parking map to park.

[0029] Optionally, the starting position recording module is configured as follows:

[0030] Based on the vehicle pitch angle and / or the visual image texture features collected by the sensors currently configured in the vehicle, identify whether the current vehicle has entered the underground parking garage;

[0031] Once the vehicle is identified as entering the underground parking garage, its starting position information is recorded.

[0032] Optional, a real-time location determination module is configured as follows:

[0033] The real-time position information of the current vehicle relative to the starting point position information is determined by any one of the following sensor data, or by fusing the data from various sensors, wherein each sensor includes an inertial measurement unit (IMU), an image sensor, radar, and an odometer.

[0034] Optionally, the apparatus provided in this embodiment of the invention further includes:

[0035] The real-time location information keeping module is configured to save the real-time location information of the current vehicle relative to the starting position while the vehicle is driving in the underground parking garage.

[0036] The historical location information acquisition module is configured to acquire the current vehicle's location information at the time of power-off as historical location information when the current vehicle is turned off.

[0037] The power-on location information determination module is configured to calculate the current location information based on historical location information when the current vehicle is powered on again, and use this as the real-time location information after power-on.

[0038] Optionally, the apparatus provided in this embodiment of the invention further includes:

[0039] The mapping module is configured to create a memory parking map based on the current vehicle's driving trajectory in manual driving mode if it receives a user-triggered memory parking map creation command when it detects that no preset memory parking map exists in the current parking garage.

[0040] Thirdly, embodiments of the present invention provide a computer device, the computer device comprising:

[0041] At least one processor is coupled to a memory, the memory storing a program or instructions that run on the processor, the program or instructions which, when executed by the processor, implement the parking method for an autonomous vehicle as provided in any embodiment of the present invention.

[0042] Fourthly, embodiments of the present invention provide a readable storage medium having a program or instructions stored thereon, which, when executed by a processor, implement the parking method for an autonomous vehicle in an underground garage as provided in any embodiment of the present invention.

[0043] Fifthly, embodiments of the present invention provide a vehicle that includes an underground parking device for an autonomous vehicle provided in any embodiment of the present invention, or includes a computer device provided in any embodiment of the present invention.

[0044] In a sixth aspect, embodiments of the present invention provide a computer program including program instructions that, when executed by a computer, implement the parking method for an autonomous vehicle in an underground garage as provided in any embodiment of the present invention. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1a This is a flowchart of a parking method for an autonomous vehicle in an underground garage provided in Embodiment 1 of the present invention;

[0047] Figure 1b This is a schematic diagram of a basement scene provided in Embodiment 1 of the present invention;

[0048] Figure 2 This is a structural block diagram of an underground parking device for an autonomous vehicle provided in Embodiment 2 of the present invention;

[0049] Figure 3 This is a structural block diagram of a computer device provided in Embodiment 3 of the present invention;

[0050] Figure 4 This is a schematic diagram of a vehicle provided in Embodiment 4 of the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] It should be noted that the terms "comprising" and "having," and any variations thereof, in the embodiments and drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0053] This invention discloses a method, apparatus, device, and medium for parking autonomous vehicles in underground garages. These are described in detail below. Example 1

[0054] Figure 1a This is a flowchart illustrating a parking method for an autonomous vehicle in an underground parking garage according to Embodiment 1 of the present invention. This method can be applied to in-vehicle terminals such as onboard computers and industrial personal computers (IPCs), and can also be applied to servers; the present invention does not limit its application to these applications. The method provided in this embodiment can be applied to underground parking garage scenarios. The method provided in this embodiment can be executed by an underground parking garage device for an autonomous vehicle, which can be implemented through software and / or hardware. Figure 1a As shown, the method provided in this embodiment specifically includes:

[0055] S110. Record the starting position information of the current vehicle when it enters the underground parking garage.

[0056] In this embodiment, recording the starting position information when the current vehicle enters the underground parking garage requires first detecting whether the current vehicle has entered the underground parking garage.

[0057] As an optional implementation, image texture features can be obtained from image information collected by the vehicle's onboard sensors, and then the vehicle can be determined to have entered the underground parking garage based on these image texture features.

[0058] As an alternative implementation, since there is usually a sloping section when driving from the surface road into the underground parking garage, the vehicle's attitude and the horizontal plane will have a certain angle when driving on the sloping section, i.e., the vehicle's pitch angle relative to the horizontal plane. The size of the current vehicle pitch angle relative to the horizontal plane can determine whether the vehicle has entered the underground parking garage. For example, if the pitch angle is greater than 4°, it means that the vehicle is driving on the sloping section leading to the underground parking garage, and that the vehicle has entered the underground parking garage. The vehicle pitch angle can be determined by an IMU (Inertial Measurement Unit), or it can be obtained by fusing data from multiple onboard sensors (such as IMU, GPS, and image sensors).

[0059] As another alternative implementation, it is also possible to identify whether a vehicle has entered the underground parking garage by combining the vehicle's pitch angle with the visual image texture features collected by the sensors currently configured on the vehicle.

[0060] Once it is determined that the vehicle is entering the underground parking garage, the starting position information of the vehicle when it enters the garage can be recorded. For example, the starting position can be set to coordinate starting point 0.

[0061] S120. When the current vehicle is driving in the underground parking garage, determine the real-time position information of the current vehicle relative to the starting point position information.

[0062] In this embodiment, the real-time position information of the current vehicle relative to the starting point can be determined by any of the following sensor data, or by fusing multiple sensor data. The sensor data includes: IMU data, image data, radar point cloud data, and odometer data, etc. This embodiment does not specifically limit the type and number of sensors.

[0063] S130. Based on the real-time location information, determine whether the current vehicle has entered the geofence corresponding to the map starting point of the preset memory parking map. If yes, proceed to step S140; otherwise, return to step S130.

[0064] The preset memory parking map is a pre-trained parking map corresponding to the current garage. During the training process, the driver needs to manually activate the autonomous vehicle's memory parking function. The autonomous vehicle will record the parking route based on the driver's driving route and create the parking map. The parking map creation is complete when the driver deactivates the parking function. Different preset memory parking maps have different starting points. Each starting point has a corresponding geofence, which is a geographical area centered on the starting point of the preset memory parking map and with a preset length (e.g., two meters) as its radius.

[0065] Specifically, Figure 1b This is a schematic diagram of a parking garage scene provided in Embodiment 1 of the present invention, as shown below. Figure 1b As shown, point A is the starting point for the driver to activate the memory parking function, and memory parking route 1 corresponds to the preset memory parking map 1. Point B is another starting point for the driver to activate the memory parking function, and memory parking route 2 corresponds to the preset memory parking map 1. Figure 2 .like Figure 1b As shown, the endpoint for the driver to end the memory parking function can be the target parking space p, or it can be a location near the target parking space q; either of these is acceptable.

[0066] In this embodiment, during the autonomous vehicle's operation in the underground parking garage, the positional relationship between its real-time location information and the starting point of a preset memory parking map allows it to select a preset memory parking map corresponding to its current location, enabling automatic parking according to the preset memory parking map. For example, if the autonomous vehicle's current real-time location is... Figure 1b Within the geofence of point A shown, select the preset memory parking map 1. The current real-time location is... Figure 1b Within the geofence area shown at point B, select the preset memory parking spot. Figure 2This avoids the problem of incorrect positioning of points A and B caused by the same surrounding environmental information in the underground parking scenario.

[0067] Furthermore, in related technologies, autonomous vehicles can currently only select one preset memory parking map as the target parking map to guide the vehicle's automatic parking. If a parking lot has multiple memory parking maps, for example... Figure 1b The two maps shown cannot determine which map to use as the target parking map for an autonomous vehicle. In this embodiment, different preset parking maps correspond to different starting points. By determining the real-time position information of the current vehicle relative to the starting point of the parking garage, and based on the positional relationship between the real-time position information and the starting point of the preset map, the target parking map corresponding to that position can be selected according to the current real-time position information. This solves the problem in related technologies where, when multiple preset memory parking maps exist in a parking garage, the autonomous vehicle cannot determine the target parking map to select.

[0068] S140. Use the preset memory parking map corresponding to the starting point of the map as the current target parking map, and drive according to the target parking map to park.

[0069] Those skilled in the art will understand that if it is determined that the current vehicle has entered the geofence of the starting point of the preset memory parking map, then compared with the method of matching the preset memory parking map starting from the starting position of the underground parking garage, the technical solution provided in this embodiment can reduce the threshold for judging the number of matching points between the current real-time location and the preset memory parking map, and increase the positioning success rate.

[0070] Those skilled in the art will also understand that, while driving according to the target parking map, the autonomous vehicle needs to consider the perceived environmental scene. If braking factors occur, such as the presence of an obstacle ahead, it needs to brake and stop. Alternatively, if a vehicle is parked in the target parking space corresponding to the target parking map, even if the destination of the memorized parking route on the target parking map is the target parking space, the current autonomous vehicle needs to brake and stop before entering the target parking space. After the current autonomous vehicle brakes and stops, it can reselect a suitable target parking space based on the perceived current environmental information, or the driver can activate manual driving mode to park. If the destination of the parking route on the target parking map is a location in front of the target parking space, for example... Figure 1bThe endpoint of the preset parking map 1 shown is a certain position before entering parking space q. The current autonomous vehicle performs the first stage of parking action according to the target parking map, that is, the autonomous vehicle drives to the front of the target parking space and completes the first stage of parking action at the endpoint of the target parking map. In the second stage of parking action, the autonomous vehicle can determine the target parking space based on the perceived current environmental information, or it can also determine the target parking space using manual driving mode in the second stage.

[0071] Furthermore, in related technologies, if the autonomous vehicle is already in the underground parking garage when memory parking is required, it cannot obtain its own GPS signal in the garage. Therefore, after the autonomous vehicle is turned off and powered on again, it must first be driven out of the garage to obtain its own GPS coordinates before being driven back into the garage; otherwise, the memory parking function cannot be used directly inside the garage. In this embodiment, the real-time position information of the autonomous vehicle relative to its starting point can be saved during its movement in the garage. When the vehicle is turned off and powered on, its position information at the time of power-off can be obtained as historical position information. After the vehicle is turned off and powered on, it will continue to move forward for a distance due to inertia. When the vehicle is powered on again, its current position information can be calculated based on the historical position information, serving as its real-time position information after power-on. The advantage of this setup is that when the vehicle is turned off and powered off, once the vehicle is powered on again, it can still determine whether it has entered the geofence of a preset memory parking map starting point based on the vehicle's real-time location information, provided that the autonomous vehicle does not leave the underground parking garage. This eliminates the need to drive out of the garage again to obtain a GPS signal, effectively improving the efficiency of automatic parking in underground parking garage scenarios.

[0072] Specifically, the current location information can be inferred from historical location information by using sensor data such as IMU data, image data, radar point cloud data, or odometer data, or by combining multiple sensors. This embodiment does not specifically limit this method.

[0073] The technical solution provided in this embodiment determines the real-time location information of the vehicle relative to the starting point of entering the underground parking garage. Based on the positional relationship between this real-time location information and the starting point of a preset memory parking map, if it is determined that the current autonomous vehicle has entered the geofence of the starting point of a preset memory parking map, the preset memory parking map corresponding to that starting point is used as the target parking map, and parking is performed according to the target parking map. This setting solves the problem of two locations being mislocated due to the same surrounding environmental information in the underground parking garage scenario. It also solves the problem that the autonomous vehicle cannot determine the current target parking map when there are multiple preset memory parking maps in the underground parking garage, effectively improving the success rate of autonomous vehicles parking in underground parking garages.

[0074] Furthermore, if a preset memory parking map is detected as not existing in the current parking garage, and a memory parking map creation command triggered by the user is received, a memory parking map will be created based on the current vehicle's driving trajectory in manual driving mode. Example 2

[0075] Figure 2 This is a structural block diagram of an underground parking device for an autonomous vehicle provided in Embodiment 2 of the present invention, as shown below. Figure 2 As shown, the device includes: a starting position recording module 210, a real-time position determination module 220, a map starting point determination module 230, and a target parking map determination module 240, wherein,

[0076] The starting position recording module 210 is configured to record the starting position information of the current vehicle when it enters the underground parking garage;

[0077] The real-time location determination module 220 is configured to determine the real-time location information of the current vehicle relative to the starting point when the current vehicle is driving in the underground parking garage.

[0078] The map starting point determination module 230 is configured to determine whether the current vehicle has entered the geofence corresponding to the map starting point of the preset memory parking map based on the real-time location information. The preset memory parking map is a parking map that has been trained and corresponds to the current garage. The starting point of different preset memory parking maps is different. The geofence is a geographical area with the map starting point of the preset memory parking map as the center and a preset length as the radius.

[0079] The target parking map determination module 240 is configured to, if it is determined that the current vehicle has entered the geofence, use the preset memory parking map corresponding to the map starting point as the current target parking map, and drive according to the current target parking map to park.

[0080] Optionally, the starting position recording module 210 is configured as follows:

[0081] Based on the vehicle pitch angle and / or the visual image texture features collected by the sensors currently configured in the vehicle, identify whether the current vehicle has entered the underground parking garage;

[0082] Once the vehicle is identified as entering the underground parking garage, its starting position information is recorded.

[0083] Optionally, the real-time position determination module 220 is specifically configured as follows:

[0084] The real-time position information of the current vehicle relative to the starting point position information is determined by any one of the following sensor data, or by fusing the data from various sensors, wherein each sensor includes an inertial measurement unit (IMU), an image sensor, radar, and an odometer.

[0085] Optionally, the apparatus provided in this embodiment of the invention further includes:

[0086] The real-time location information keeping module is configured to save the real-time location information of the current vehicle relative to the starting position while the vehicle is driving in the underground parking garage.

[0087] The historical location information acquisition module is configured to acquire the current vehicle's location information at the time of power-off as historical location information when the current vehicle is turned off.

[0088] The power-on location information determination module is configured to calculate the current location information based on historical location information when the current vehicle is powered on again, and use this as the real-time location information after power-on.

[0089] Optionally, the apparatus provided in this embodiment of the invention further includes:

[0090] The mapping module is configured to create a memory parking map based on the current vehicle's driving trajectory in manual driving mode if it receives a user-triggered memory parking map creation command when it detects that no preset memory parking map exists in the current parking garage.

[0091] The parking device for autonomous vehicles provided in this embodiment of the invention can execute the parking method for autonomous vehicles provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method. Technical details not described in detail in the above embodiments can be found in the parking method for autonomous vehicles provided in any embodiment of the invention. Example 3

[0092] Figure 3 This is a structural block diagram of a computer device provided in Embodiment 3 of the present invention, as shown below. Figure 3 As shown, the computer device includes:

[0093] At least one processor ( Figure 3 A processor 520 is shown in the image.

[0094] The processor 520 is coupled to the memory 510, which stores a program or instructions that run on the processor 520. When the program or instructions are executed by the processor 520, they implement the parking method for an autonomous vehicle in an underground garage provided in any embodiment of the present invention.

[0095] Based on the above embodiments, another embodiment of the present invention provides a vehicle that includes the apparatus as described in any of the above embodiments, or includes the computer equipment as described above. Example 4

[0096] Figure 4 This is a schematic diagram of a vehicle provided in Embodiment 4 of the present invention. Figure 4 As shown, the vehicle includes a speed sensor 61, an ECU (Electronic Control Unit) 62, a GPS (Global Positioning System) positioning device 63, and a T-Box (Telematics Box) 64. The speed sensor 61 measures the vehicle speed and uses this speed as an empirical speed for model training; the GPS positioning device 63 obtains the vehicle's current geographical location; the T-Box 64 can act as a gateway to communicate with the server; and the ECU 62 can execute the aforementioned parking method for the autonomous vehicle in underground parking garages.

[0097] In addition, the vehicle may also include: a V2X (Vehicle-to-Everything) module 65, a radar 66, and a camera 67. The V2X module 65 is used to communicate with other vehicles, roadside equipment, etc.; the radar 66 or camera 67 is used to perceive road environment information in front and / or other directions to obtain raw point cloud data; the radar 66 and / or camera 67 can be configured at the front and / or rear of the vehicle.

[0098] Based on the above method embodiments, another embodiment of the present invention provides a readable storage medium having a program or instructions stored thereon, which, when executed by a processor, causes the processor to implement the parking method for an autonomous vehicle in a parking garage as described in any of the above embodiments.

[0099] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0100] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for parking an autonomous vehicle in an underground garage, characterized in that, include: Record the starting position information of the vehicle when it enters the underground parking garage; When the current vehicle is driving in the underground parking garage, determine the real-time position information of the current vehicle relative to the starting point position information; Based on the real-time location information, it is determined whether the current vehicle has entered the geofence corresponding to the map starting point of the preset memory parking map. The preset memory parking map is a parking map that has been trained and corresponds to the current garage. The starting point of different preset memory parking maps is different. The geofence is a geographical area with the map starting point of the preset memory parking map as the center and a preset length as the radius. If it is determined that the current vehicle has entered the geofence, the preset memory parking map corresponding to the map starting point is used as the current target parking map, and the vehicle drives according to the current target parking map to park.

2. The method according to claim 1, characterized in that, The information recording the starting position of the vehicle when it enters the underground parking garage includes: Based on the vehicle pitch angle and / or the visual image texture features collected by the sensors currently configured in the vehicle, identify whether the current vehicle has entered the underground parking garage; Once the vehicle is identified as entering the underground parking garage, its starting position information is recorded.

3. The method according to claim 1, characterized in that, Determining the real-time position information of the current vehicle relative to the starting point position information includes: The real-time position information of the current vehicle relative to the starting point position information is determined by any one of the following sensor data, or by fusing the data from various sensors, wherein each sensor includes an inertial measurement unit (IMU), an image sensor, radar, and an odometer.

4. The method according to claim 1, characterized in that, The method further includes: During the process of the current vehicle driving in the underground parking garage, the real-time position information of the current vehicle relative to the starting position is saved; When the current vehicle is turned off and powered down, the location information of the current vehicle at the time of power-down will be used as historical location information; When the current vehicle is powered on again, the current location information is calculated based on the historical location information, which is used as the real-time location information after power-on.

5. The method according to claim 1, characterized in that, The method further includes: If a user-triggered memory parking map creation command is received and the current parking garage does not have a preset memory parking map, a memory parking map will be created based on the current vehicle's driving trajectory in manual driving mode.

6. A parking device for an autonomous vehicle in an underground garage, characterized in that, include: The starting position recording module is configured to record the starting position information of the current vehicle when it enters the underground parking garage; The real-time location determination module is configured to determine the real-time location information of the current vehicle relative to the starting point location information when the current vehicle is driving in the underground parking garage. The map starting point determination module is configured to determine, based on the real-time location information, whether the current vehicle has entered the geofence corresponding to the map starting point of the preset memory parking map. The preset memory parking map is a parking map that has been trained and corresponds to the current garage. The starting point of different preset memory parking maps is different. The geofence is a geographical area with the map starting point of the preset memory parking map as the center and a preset length as the radius. The target parking map determination module is configured to, if it is determined that the current vehicle has entered the geofence, use the preset memory parking map corresponding to the map starting point as the current target parking map, and drive according to the current target parking map to park.

7. The apparatus according to claim 6, characterized in that, The starting position recording module is configured as follows: Based on the vehicle pitch angle and / or the visual image texture features collected by the sensors currently configured in the vehicle, identify whether the current vehicle has entered the underground parking garage; Once the vehicle is identified as entering the underground parking garage, its starting position information is recorded.

8. The apparatus according to claim 6, characterized in that, The real-time location determination module is specifically configured as follows: The real-time position information of the current vehicle relative to the starting point position information is determined by any one of the following sensor data, or by fusing the data from various sensors, wherein each sensor includes an inertial measurement unit (IMU), an image sensor, radar, and an odometer.

9. The apparatus according to claim 6, characterized in that, The device further includes: The real-time location information holding module is configured to save the real-time location information of the current vehicle relative to the starting position during the process of the current vehicle driving in the underground parking garage. The historical location information acquisition module is configured to use the current vehicle's location information at the time of power-off as historical location information when the current vehicle is turned off and powered down. The power-on location information determination module is configured to calculate the current location information based on the historical location information when the current vehicle is powered on again, and use it as the real-time location information after power-on.

10. A computer device, characterized in that, It includes at least one processor coupled to a memory storing a program or instructions that run on the processor, the program or instructions being executed by the processor to implement the steps of the parking method for an autonomous vehicle as described in any one of claims 1 to 5.

11. A readable storage medium having a program or instructions stored thereon, characterized in that, When the program or instructions are executed by the processor, they implement the steps of the parking method for an autonomous vehicle as described in any one of claims 1 to 5.

Images