Method, apparatus and electronic device for positioning initialization

Abstract

The application discloses a positioning initialization method, device and electronic equipment. The positioning initialization method comprises the following steps: in response to the first number of searched stars being greater than a preset number threshold, screening a plurality of coarse matching maps from a plurality of initial maps based on a current positioning signal and a positioning signal corresponding to each of the plurality of initial maps and acquired in advance; in response to the second number of searched stars being not greater than the number threshold, screening a plurality of to-be-precise matching maps from the plurality of coarse matching maps based on a second position and the position of each point on a map line of an underground parking lot in the plurality of coarse matching maps and acquired in advance; and determining a target precise matching map and an initial pose of the vehicle in the target precise matching map based on the current environment feature of the vehicle and the plurality of to-be-precise matching maps. The application can realize autonomous parking, and the driving path of the vehicle from the entrance of the underground parking lot to the starting point of the map does not have to be completely consistent with the driving path in the mapping stage.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a method, apparatus and electronic device for positioning initialization. Background Technology

[0002] Home-zone Parking Assist (HPA), also known as Home-zone Parking Pilot (HPP), is a technology that can park a car in a designated parking space in certain scenarios, replacing the user's intervention. Memory parking is a key technology in intelligent driving assistance and a popular research area. The HPA parking process mainly consists of two stages: route learning and mapping (hereinafter referred to as the mapping stage) and route playback autonomous parking (hereinafter referred to as the parking stage). In the mapping stage, the vehicle learns, records, and stores the starting point location, parking location, parking path, and environmental feature map information during the journey using its own sensors. In the parking stage, when the vehicle is near the mapping starting point again, it performs repositioning (also called localization initialization). After the user activates the HPA function, the vehicle automatically drives to the designated parking space according to the learned path and completes the parking.

[0003] Location initialization is a crucial step in memory-based parking technology. Only after location initialization is completed, and the vehicle's pose on the map is known (describing the position and orientation of an object in a specified coordinate system; position: location in space, orientation: facing direction in space), can autonomous parking begin. The accuracy of location initialization also affects the accuracy of location during subsequent parking processes, and is even a key factor determining whether the vehicle can ultimately be successfully parked in the designated parking space.

[0004] In related technologies, memory parking positioning initialization is achieved through path matching. The path matching method involves recording the vehicle's path from the entrance of the underground parking lot (HPA function is mostly for underground parking lot scenarios) to the mapping starting point, such as how many times the vehicle turns left and right. During the autonomous parking phase, if the vehicle's driving path is basically consistent with the path recorded during the learning and mapping phase, it is considered to have reached the vicinity of the mapping starting point. The current real-time map is then compared with the map saved during the learning and mapping phase to construct an optimization target, calculate the current vehicle pose, and complete the positioning initialization.

[0005] However, this path matching method requires that the driving path of a vehicle entering the underground parking lot and arriving at the map starting point (i.e., the mapping starting point) during the autonomous parking phase must be completely consistent with the driving path during the mapping phase; otherwise, path matching will fail. To improve the robustness of this method, it is sometimes required that only one route can be learned within an underground parking lot. Summary of the Invention

[0006] The purpose of this application is to provide a method, apparatus, and electronic device for positioning initialization, in order to solve the problem in the prior art that the driving path of a vehicle entering the underground parking lot entrance and arriving at the map starting point during the autonomous parking stage must be completely consistent with the driving path during the mapping stage, otherwise path matching will fail.

[0007] In a first aspect, embodiments of this application provide a method for positioning initialization, including:

[0008] Obtain the first satellite count and current positioning signal when the vehicle reaches its first location;

[0009] In response to the first number of satellite searches exceeding a preset threshold, multiple coarse matching maps are selected from the multiple initial maps based on the current positioning signal and the pre-acquired positioning signals corresponding to the multiple initial maps respectively; wherein the difference between the positioning signal corresponding to the coarse matching map and the current positioning signal is less than a preset first signal threshold.

[0010] Obtain the second satellite count when the vehicle reaches the second location;

[0011] In response to the second number of satellite searches not exceeding a quantity threshold, based on the second location and the locations of each point on the underground parking lot map route obtained in advance from multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps; among them, there are point locations on the map route of the maps to be finely matched where the distance to the second location is less than a preset first distance threshold.

[0012] Obtain the current environmental characteristics of the vehicle;

[0013] Based on the vehicle's current environmental characteristics and multiple maps to be precisely matched, the target precise matching map and the vehicle's initial pose on the target precise matching map are determined.

[0014] In some embodiments, it also includes:

[0015] In response to the second number of satellite searches being greater than a quantity threshold, based on the current positioning signal and the pre-acquired positioning signals corresponding to multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps; wherein, the difference between the positioning signal corresponding to the map to be finely matched and the current positioning signal is less than a preset second signal threshold; the first signal threshold is greater than the second signal threshold.

[0016] In some embodiments, based on the vehicle's current environmental characteristics and multiple maps to be precisely matched, determining the target precise matching map and the vehicle's initial pose on the target precise matching map includes:

[0017] Based on the vehicle's current environmental characteristics and multiple maps to be matched, the vehicle's pose in the multiple maps to be matched is determined, and multiple matching scores are output.

[0018] If the matching score is greater than the preset threshold score, the map to be finely matched and the vehicle's pose on the map to be finely matched, which correspond to the matching score greater than the threshold score, are respectively determined as the target finely matched map and the initial pose of the vehicle on the target finely matched map.

[0019] In some embodiments, based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps respectively, multiple coarse matching maps are selected from the multiple initial maps, including:

[0020] For each of the multiple positioning signals corresponding to the multiple initial maps, the following operations are performed: calculate the longitude difference between the longitude of the current positioning signal and the longitude of each positioning signal, and calculate the latitude difference between the latitude of the current positioning signal and the latitude of each positioning signal.

[0021] Based on the longitude and latitude differences, determine whether the initial map is a coarse-match map;

[0022] If, among the positioning signals corresponding to the initial map, there exists a positioning signal with a longitude difference less than a preset longitude threshold and a latitude difference less than a preset latitude threshold, then the initial map is determined to be a coarse matching map.

[0023] The first signal threshold includes: longitude threshold and latitude threshold.

[0024] In some embodiments, the method further includes: acquiring an initial map and a positioning signal corresponding to the initial map, including:

[0025] From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location;

[0026] If the number of satellites searched in the third search exceeds the quantity threshold, the positioning signal is recorded and saved.

[0027] In response to the mapping command, a map of the above-ground parking lot is created using the vehicle's current location as the starting point for mapping on the ground, and the positioning signals of each point on the above-ground map route are saved. The initial map includes: a map of the above-ground parking lot; the positioning signals corresponding to the initial map are the positioning signals along the route traversed by the vehicle from the start of vehicle startup to the end of mapping.

[0028] In some embodiments, the method further includes: acquiring an initial map, positioning signals corresponding to the initial map, and the locations of points along the map route of the underground parking lot, including:

[0029] From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location;

[0030] If the number of satellites searched in the third search exceeds the quantity threshold, the positioning signal is recorded and saved.

[0031] During vehicle movement, if the number of satellites detected in the third phase does not exceed the threshold, the vehicle's current location will be used as the entrance to the underground parking lot, and the recording of the location signal will stop.

[0032] When a vehicle enters from the underground parking lot entrance, in response to the mapping command, a map of the underground parking lot is created using the vehicle's current position as the starting point for underground mapping, and the poses of each point on the underground parking lot map route relative to the starting point of underground mapping are saved; wherein, the initial map includes: the map of the underground parking lot; the positioning signal corresponding to the initial map is: the positioning signal on the route traversed by the vehicle from the start of vehicle startup until the third satellite search number is not greater than the quantity threshold.

[0033] In some embodiments, each point on the underground parking lot map route includes the starting point for underground mapping.

[0034] Based on the second location and the locations of points along the underground parking lot map route obtained from multiple pre-acquired coarse-matching maps, multiple maps to be finely matched are selected from the multiple coarse-matching maps, including:

[0035] Obtain the poses of the underground mapping starting points relative to the entrance of the underground parking lot from multiple coarsely matched maps;

[0036] Based on the second location and the location of the underground mapping starting point relative to the entrance of the underground parking lot from multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps.

[0037] In some embodiments, obtaining the poses of the underground mapping starting points of multiple coarse-matched maps relative to the entrance of an underground parking lot includes:

[0038] Obtain the trajectory of the vehicle from the entrance of the underground parking lot to the starting point of the underground mapping;

[0039] Based on the trajectory before the starting point, the pose of the underground mapping starting point relative to the entrance of the underground parking lot is obtained.

[0040] In some embodiments, based on the second location and the locations of points on the underground parking lot map route obtained in advance from multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps, including:

[0041] The poses of the underground mapping starting points of multiple coarse matching maps relative to the underground parking lot entrance are obtained; and the poses of each point on the underground parking lot map line relative to the underground mapping starting point are obtained. Through position transformation, the positions of each point on the underground parking lot map line relative to the underground parking lot entrance are obtained.

[0042] Based on the second location and the position of each point on the underground parking lot map line relative to the underground parking lot entrance, multiple maps to be finely matched are selected from multiple coarse matching maps.

[0043] Secondly, embodiments of this application provide a positioning initialization apparatus, comprising:

[0044] The first acquisition module is configured to acquire the first satellite search count and the current positioning signal when the vehicle travels to the first position;

[0045] The coarse matching module is configured to, in response to a first number of satellite searches exceeding a preset threshold, select multiple coarse matching maps from multiple initial maps based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps respectively; wherein, among the points on the map lines of the map to be finely matched, there are points whose distance to the second position is less than a preset first distance threshold.

[0046] The second acquisition module is configured to acquire the second satellite search count when the vehicle travels to the second position;

[0047] The first fine matching module is configured to, in response to the second number of satellite searches not exceeding a quantity threshold, filter out multiple maps to be finely matched from multiple coarse matching maps based on the second location and the locations of each point on the map route of the underground parking lot obtained in advance; wherein, the distance between the second location and the locations of each point on the map route of the map to be finely matched is less than a preset first distance threshold.

[0048] The third acquisition module is configured to acquire the current environmental characteristics of the vehicle.

[0049] The pose determination module is configured to determine the target fine-matching map and the vehicle's initial pose on the target fine-matching map based on the vehicle's current environmental characteristics and multiple maps to be fine-matched.

[0050] The positioning initialization method provided in this application, during vehicle movement, firstly determines whether the vehicle is above ground or underground using satellite search counts. Based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps, multiple coarse matching maps are selected from the multiple initial maps to determine whether the vehicle has reached the vicinity of a parking lot along a learned route. Then, satellite search counts are used again to determine whether the vehicle is above ground or underground, and based on a second satellite search count not exceeding a certain threshold, it is determined that the vehicle has reached an underground parking lot. Based on a second location and the positions of points on the underground parking lot map route from the pre-acquired multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps to determine whether the vehicle is near the underground parking lot map route. Map feature matching only begins when the vehicle reaches the vicinity of the underground parking lot map route, thus allowing positioning initialization to be performed at any location near the underground parking lot map route, rather than being limited to the mapping start point, improving the user experience.

[0051] In this application, during the parking phase, it is only necessary to compare the distances between the vehicle's second position and various points on the map route. When there is a point on the map route whose distance to the second position is less than a preset first distance threshold, it proves that the vehicle has reached the vicinity of the map route, and then positioning initialization can be performed. It can be seen that the positioning initialization method of this invention does not depend on the mapping starting point and does not concern itself with the vehicle's driving trajectory near the map route. The vehicle can complete positioning initialization at any position near the map route. Therefore, the driving path of the vehicle from the entrance of the underground parking lot to the mapping starting point does not need to be completely consistent with the driving path during the mapping phase, and it is not necessary to find the mapping starting point; as long as the vehicle reaches the vicinity of the map route, positioning initialization can be performed. For example, if the vehicle has circled the underground parking lot, and the driving path is different from the previously learned path, positioning initialization can still be performed as long as the vehicle reaches the vicinity of the map route. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating a positioning initialization method according to an embodiment of this application;

[0053] Figure 2 This is a block diagram illustrating a positioning initialization method according to an embodiment of this application.

[0054] Figure 3 This is a flowchart illustrating the coarse matching process during the underground parking stage in an embodiment of this application.

[0055] Figure 4 This is a schematic diagram of the fine matching process during the underground parking stage according to an embodiment of this application;

[0056] Figure 5a This is a schematic diagram of the mapping stage of an above-ground parking lot according to an embodiment of this application;

[0057] Figure 5b This is a schematic diagram of the mapping stage of an underground parking lot according to an embodiment of this application;

[0058] Figure 6 This is a schematic diagram of a positioning initialization device according to an embodiment of this application. Detailed Implementation

[0059] Various embodiments and features of this application are described herein with reference to the accompanying drawings.

[0060] It should be understood that various modifications can be made to the embodiments described herein. Therefore, the above description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this application will be apparent to those skilled in the art.

[0061] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0062] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0063] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application.

[0064] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0065] Specific embodiments of this application are described thereafter with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in a variety of substantially any suitable detailed structures.

[0066] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.

[0067] See Figure 2 HPA (Home-zone Parking Assist) consists of two phases: mapping and parking. The localization initialization process will also be explained in terms of these two phases (mapping and parking).

[0068] During the mapping phase, information related to the initial map is saved, and the generated initial map (i.e., such as...) Figure 2 The land in the middle Figure 1 ,land Figure 2 ,land Figure 3 (map n) is added to the initial map list. Optionally, the initial map is a map that has been pre-learned and stored based on the user's selection. For example, it could be the parking lot of the user's residential community, the parking lot of their office, or the parking lot of a shopping mall or supermarket they frequently visit.

[0069] During the parking phase, the first stage (coarse matching) uses positioning signals to determine if the vehicle has arrived near a parking lot with a previously learned route. It then filters out successfully matched initial maps from the initial map list and designates these as the coarse-matched map (i.e., ...). Figure 2 Maps i, j, ..., k in the dataset are added to the coarse matching list.

[0070] During the parking phase, the second phase (fine matching) determines whether the vehicle has reached the vicinity of a certain map route based on the second location and the locations of various points on the underground parking lot map route obtained from multiple pre-acquired coarse matching maps. If so, the target fine matching map and the vehicle's initial pose in the target fine matching map are obtained through the vehicle's current environmental characteristics (i.e., such as...). Figure 2 The map x and the current vehicle's pose in map x, ..., map y and the current vehicle's pose in map y) are added to the fine-matching list to complete the localization initialization.

[0071] like Figure 2 As shown, (i,j,...,k)∈[1,2,3,...,n]; (x,y)∈[i,j,...,k].

[0072] The positioning initialization method provided in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0073] See Figure 1 As shown, Figure 1 This is a flowchart illustrating a positioning initialization method according to an embodiment of this application. The embodiment of this application provides a positioning initialization method, including:

[0074] S10, acquire the first satellite count and current positioning signal when the vehicle reaches the first location;

[0075] For example, the positioning signal can be a GPS signal, a BeiDou satellite signal, etc. The satellite search count refers to the number of satellites the device has detected. Generally, this value is greater than 0 outdoors, indicating a valid GPS signal detection. However, once underground or in a tunnel, the signal is blocked, and no satellites can be detected, resulting in a satellite search count of 0. For ease of description, the embodiments of this application will be described below using GPS signals as an example.

[0076] S20, in response to the first number of satellite searches being greater than a preset number threshold, based on the current positioning signal and the pre-acquired positioning signals corresponding to the multiple initial maps respectively, multiple coarse matching maps are selected from the multiple initial maps; wherein, the difference between the positioning signal corresponding to the coarse matching map and the current positioning signal is less than a preset first signal threshold.

[0077] In some embodiments, among the positioning signals corresponding to the coarse matching map, there is at least one positioning signal whose difference from the current positioning signal is less than a preset first signal threshold. Optionally, there are multiple positioning signals corresponding to the initial map. When there is at least one positioning signal whose difference from the current positioning signal is less than the preset first signal threshold, the initial map corresponding to that positioning signal is selected as the coarse matching map.

[0078] For example, the preset quantity threshold is 0. The positioning signal corresponding to the initial map refers to the positioning signal saved during the mapping stage. This positioning signal can be multiple positioning signals, such as 3000 GPS signals, 5000 GPS signals, etc., and the number of GPS signals stored varies depending on the storage space.

[0079] Optionally, the positioning signals corresponding to the initial map can be stored as soon as the vehicle starts. Once a certain number are stored, considering the size of the storage space, a first-in-first-out (FIFO) method can be used to store them, ensuring that a fixed number of positioning signals are always stored in the storage space.

[0080] S30, obtain the second satellite search count when the vehicle reaches the second position;

[0081] For example, satellite search counts are acquired in real time from the moment the vehicle starts until it is in motion.

[0082] S40a, in response to the second number of satellite searches not being greater than a quantity threshold, based on the second location and the locations of each point on the underground parking lot map route of the multiple pre-acquired coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps; wherein, among the locations of each point on the map route of the maps to be finely matched, there are point locations whose distance from the second location is less than a preset first distance threshold.

[0083] For example, if the second number of satellite searches is not greater than a preset threshold, it can be determined that the vehicle has entered the underground parking lot. Optionally, it can be determined whether there are points on each map line that are close to the second location (the distance between them is less than a first distance threshold) based on the location of the second location and the locations of points on the underground parking lot map lines of multiple coarse matching maps. If so, it proves that the vehicle is near the map line.

[0084] S50, acquires the vehicle's current environmental characteristics;

[0085] For example, the current environmental features may include point cloud features, semantic features, etc.

[0086] S60 determines the target fine-matching map and the vehicle's initial pose on the target fine-matching map based on the vehicle's current environmental characteristics and multiple maps to be finely matched.

[0087] For example, the initial map includes pre-acquired environmental features, so the coarse matching map and the fine matching map filtered from the initial map also include environmental features, such as point cloud features and semantic features.

[0088] In this application, based on the current environmental characteristics of the vehicle and multiple maps to be precisely matched, a target precise matching map can be selected from the multiple maps to be precisely matched, and the initial pose of the vehicle on the target precise matching map can be determined to complete the positioning initialization of vehicle memory parking.

[0089] The positioning initialization method provided in this application, during vehicle movement, firstly determines whether the vehicle is above ground or underground using satellite search counts. Based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps, multiple coarse matching maps are selected from the multiple initial maps to determine whether the vehicle has reached the vicinity of a parking lot along a learned route. Then, satellite search counts are used again to determine whether the vehicle is above ground or underground, and based on a second satellite search count not exceeding a certain threshold, it is determined that the vehicle has reached an underground parking lot. Based on a second location and the positions of points on the underground parking lot map route from the pre-acquired multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps to determine whether the vehicle is near the underground parking lot map route. Map feature matching only begins when the vehicle reaches the vicinity of the underground parking lot map route, thus allowing positioning initialization to be performed at any location near the underground parking lot map route, rather than being limited to the mapping start point, improving the user experience.

[0090] In this application, during the parking phase, it is only necessary to compare the distances between the vehicle's second position and various points on the map route. When there is a point on the map route whose distance to the second position is less than a preset first distance threshold, it proves that the vehicle has reached the vicinity of the map route, and then positioning initialization can be performed. It can be seen that the positioning initialization method of this invention does not depend on the mapping starting point and does not concern itself with the vehicle's driving trajectory near the map route. The vehicle can complete positioning initialization at any position near the map route. Therefore, the driving path of the vehicle from the entrance of the underground parking lot to the mapping starting point does not need to be completely consistent with the driving path during the mapping phase, and it is not necessary to find the mapping starting point; as long as the vehicle reaches the vicinity of the map route, positioning initialization can be performed. For example, if the vehicle has circled the underground parking lot, and the driving path is different from the previously learned path, positioning initialization can still be performed as long as the vehicle reaches the vicinity of the map route.

[0091] In some embodiments, see continue to see Figure 1 As shown, when the second number of satellite searches exceeds the threshold, it indicates that the vehicle is on the ground. At this point, it is necessary to complete the memory parking process in the ground parking lot. The positioning initialization method also includes:

[0092] S40b, in response to the second number of satellite searches being greater than the quantity threshold, based on the current positioning signal and the pre-acquired positioning signals corresponding to the multiple coarse matching maps respectively, multiple maps to be finely matched are selected from the multiple coarse matching maps; wherein, the difference between the positioning signal corresponding to the map to be finely matched and the current positioning signal is less than a preset second signal threshold; the first signal threshold is greater than the second signal threshold.

[0093] This application embodiment addresses the situation of above-ground parking lots. It utilizes the current positioning signal and pre-acquired positioning signals corresponding to multiple coarse-matching maps, and further refines the matching by setting a first signal threshold greater than a second signal threshold. This results in the selection of maps to be refined having points closer to the second location. In other words, by using the current positioning signal and the pre-acquired positioning signals corresponding to multiple coarse-matching maps, it determines whether the vehicle is infinitely close to or on the map route. If so, subsequent positioning initialization is performed. By retaining the positioning signal corresponding to the map during the mapping stage, this application ensures that memory parking is not limited to underground parking scenarios but is also applicable to above-ground parking scenarios.

[0094] In some embodiments, based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps respectively, multiple coarse matching maps are selected from the multiple initial maps, including:

[0095] For each of the multiple positioning signals corresponding to the multiple initial maps, the following operations are performed: calculate the longitude difference between the longitude of the current positioning signal and the longitude of each positioning signal, and calculate the latitude difference between the latitude of the current positioning signal and the latitude of each positioning signal.

[0096] Based on the longitude and latitude differences, determine whether the initial map is a coarse-match map;

[0097] If, among the positioning signals corresponding to the initial map, there exists a positioning signal with a longitude difference less than a preset longitude threshold and a latitude difference less than a preset latitude threshold, then the initial map is determined to be a coarse matching map; wherein, the first signal threshold includes: a longitude threshold and a latitude threshold.

[0098] For example, the positioning signal corresponding to an initial map can be multiple positioning signals, such as multiple GPS signals. The number of GPS signals can be determined based on a preset storage space size, calculation accuracy, or user settings. For example, an initial map can correspond to 5000 GPS signals. The acquired current GPS signal needs to be compared with the longitude and latitude of each of the 5000 GPS signals to obtain 5000 longitude and latitude differences. If at least one (e.g., as long as one) longitude difference and the latitude difference meets a preset condition, then the coarse matching is considered successful, and the initial map is determined to be a coarse matching map.

[0099] The process of calculating the difference between the current GPS signal and another GPS signal is as follows, where the first GPS signal is denoted as the first GPS signal.

[0100] Specifically, the longitude difference is obtained by taking the difference between the longitude of the current GPS signal and the longitude of the first GPS signal;

[0101] The latitude difference is obtained by taking the difference between the latitude of the current GPS signal and the latitude of the first GPS signal.

[0102] When the longitude difference is less than a preset longitude threshold and the latitude difference is less than a preset latitude threshold, the initial map corresponding to the first GPS signal is determined as a coarse matching map.

[0103] The following uses parking in an underground parking scenario as an example to illustrate the positioning initialization method provided in this application.

[0104] In one specific embodiment, see Figure 3 , Figure 3 This is a flowchart illustrating the coarse matching process for an underground parking scenario according to an embodiment of this application. First, a coarse matching list is set up to store the coarse matching maps that have been successfully matched. The coarse matching list is initially empty.

[0105] After the vehicle starts, it checks if the first satellite search count (current GPS satellite search count) is greater than 0 (a preset threshold). If the first satellite search count is greater than 0, the coarse matching list is cleared and recalculated. For example, if the first satellite search count is greater than 0, it calculates the difference between the vehicle's current positioning signal and the positioning signals corresponding to multiple initial maps. If any difference is less than the first signal threshold, the coarse matching is successful, and the coarse matching map is added to the coarse matching list. The next moment, it checks again if the first satellite search count is greater than 0. If the first satellite search count is still greater than 0, the coarse matching list is cleared and recalculated. This process is repeated until the first satellite search count equals 0, at which point the saved coarse matching list is retained.

[0106] See Figure 4 As shown, Figure 4 This is a flowchart illustrating the fine-matching process for an underground parking scenario according to an embodiment of this application. First, a fine-matching list is set up to store the target fine-matching maps that have been successfully matched and the initial pose of the current vehicle in the target fine-matching map. The fine-matching list is initially empty.

[0107] While the vehicle is moving, if the first satellite search count is 0, check if the coarse matching list is empty. If it is, the positioning initialization fails and the process exits. Otherwise, check if the second satellite search count (current GPS satellite search count) is greater than 0. If the second satellite search count (current GPS satellite search count) is 0, the vehicle is determined to be underground. It iterates through all coarse matching maps in the coarse matching map list and calculates the distance between the vehicle's current position and each point on the map lines of the multiple coarse matching maps. If there is a point on the map lines of the multiple coarse matching maps whose distance to the vehicle's current position is less than a first distance threshold, the coarse matching map corresponding to that point is selected as a map to be finely matched.

[0108] Based on the vehicle's current environmental characteristics and multiple maps to be finely matched, a target fine-matching map and the vehicle's initial pose in the target fine-matching map are determined, and the target fine-matching map and the vehicle's initial pose in the target fine-matching map are added to the fine-matching list.

[0109] In some embodiments, based on the vehicle's current environmental characteristics and multiple maps to be precisely matched, determining the target precise matching map and the vehicle's initial pose on the target precise matching map includes:

[0110] Based on the vehicle's current environmental characteristics and multiple maps to be matched, the vehicle's pose in the multiple maps to be matched is determined, and multiple matching scores are output.

[0111] If the matching score is greater than the preset threshold score, the map to be finely matched and the vehicle's pose on the map to be finely matched, which correspond to the matching score greater than the threshold score, are respectively determined as the target finely matched map and the initial pose of the vehicle on the target finely matched map.

[0112] For example, the current environmental features of the vehicle can be detected by sensors, and then map feature matching can be performed based on the current environmental features of the vehicle and multiple maps to be matched to calculate the current pose of the vehicle in the maps to be matched.

[0113] The types of maps to be matched can include point cloud maps, semantic maps, etc. Different map feature matching methods can be selected depending on the type of map to be matched.

[0114] In one embodiment, the map to be precisely matched in this application is a point cloud map. The corresponding point cloud map feature matching algorithm may include ICP (Iterative Closest Point) and NDT (Normal Distribution Transform), which are commonly used point cloud map feature matching algorithms. The point cloud map feature matching algorithm can output the current vehicle's pose in the map to be precisely matched and provide a matching score. Based on the matching score and a threshold score, it is determined whether the map feature matching is successful.

[0115] If the matching score is greater than the preset threshold score, the map feature matching is determined to be successful. The matching pose is recorded and used as the initial pose for localization. The matching poses of the map to be finely matched and the vehicle in the map to be finely matched (matching poses) corresponding to the matching scores greater than the threshold score are determined as the initial poses of the target finely matched map and the vehicle in the target finely matched map (matching poses), respectively. The target finely matched map and the vehicle in the target finely matched map are added to the fine matching list.

[0116] The fine-match list is the result of the localization initialization, outputting the target fine-match map and the vehicle's initial pose in the target fine-match map.

[0117] For example, the number of target fine-matching maps is greater than or equal to 1, that is, the fine-matching list can have at least one target fine-matching map and the initial pose of the vehicle in the corresponding target fine-matching map.

[0118] If the number of target precise matching maps in the precise matching list is greater than 1, the user can choose which target precise matching map to use for parking.

[0119] The following describes the mapping process for above-ground parking lots and the mapping process for underground parking lots respectively:

[0120] (a) The process of drawing up the above-ground parking lot;

[0121] In some embodiments, acquiring an initial map and a positioning signal corresponding to the initial map includes:

[0122] From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location;

[0123] If the number of satellites searched in the third search exceeds the quantity threshold, the positioning signal is recorded and saved.

[0124] In response to the mapping command, a map of the above-ground parking lot is created using the vehicle's current location as the starting point for mapping on the ground, and the positioning signals of each point on the above-ground map route are saved. The initial map includes: a map of the above-ground parking lot; the positioning signals corresponding to the initial map are the positioning signals along the route traversed by the vehicle from the start of vehicle startup to the end of mapping.

[0125] In some embodiments, during the mapping phase of the above-ground parking lot, the location signal (GPS signal) is continuously recorded and saved to a preset storage space from the moment the vehicle starts. Considering factors such as storage space size, positioning efficiency, or human settings, the location signal in the preset storage space can be updated in a first-in-first-out manner.

[0126] like Figure 5a As shown, the vehicle continuously records GPS signals from the moment it starts until it receives a mapping command. The map of the above-ground parking lot is created using the vehicle's current location as the starting point. The GPS signals recorded before the mapping start point can be called the "pre-start GPS signal," and the GPS signals recorded between the mapping start and end point can be called the "map route GPS signal." Therefore, the positioning signal corresponding to the initial above-ground map is the sum of the "pre-start GPS signal" and the "map route GPS signal." In other words, for creating an above-ground parking lot, GPS positioning signals must be recorded continuously from the moment the vehicle starts until the mapping is completed. In summary, each above-ground parking lot map has a corresponding "pre-start GPS signal" plus a "map route GPS signal." Optionally, in addition to the above information, the created map may also include environmental features, such as point cloud features and semantic features.

[0127] For above-ground parking lots, the embodiments of this application can realize the location initialization based on any point on the map route between the mapping start point and the mapping end point.

[0128] (II) The process of drawing up the underground parking lot;

[0129] In some embodiments, obtaining an initial map, a positioning signal corresponding to the initial map, and the locations of points along the map route of the underground parking lot includes:

[0130] From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location;

[0131] If the third number of satellite searches is greater than the number threshold, then the positioning signal is recorded and saved;

[0132] During vehicle movement, if the number of satellites detected in the third phase does not exceed the threshold, the vehicle's current location will be used as the entrance to the underground parking lot, and the recording of the location signal will stop.

[0133] When a vehicle enters from the underground parking lot entrance, in response to the mapping command, the map of the underground parking lot is created with the vehicle's current position as the starting point for underground mapping, and the pose of each point on the underground parking lot map line relative to the starting point of underground mapping is saved. The initial map includes: the map of the underground parking lot; the positioning signal corresponding to the initial map is: the positioning signal on the line traversed by the vehicle from the start of vehicle startup until the third satellite search number is not greater than the quantity threshold.

[0134] In some embodiments, such as Figure 5b As shown, after the vehicle starts, when the third satellite count is greater than 0, it indicates that the vehicle is currently on the ground. The system continuously saves and records the vehicle's positioning signal from startup until the third satellite count is greater than 0. The vehicle's location is then taken as the entrance to the underground parking lot when the third satellite count equals 0, and positioning signal recording stops. The positioning signal before the vehicle reaches the underground parking lot entrance (before the third satellite count becomes 0) is referred to as the "GPS signal before the parking lot entrance" (see [link]). Figure 5b ).

[0135] When a vehicle enters from the underground parking lot entrance, in response to a mapping command, a map of the underground parking lot is created using the vehicle's current position as the starting point for underground mapping, and the poses of each point on the underground parking lot map route relative to the starting point are saved. Optionally, the trajectory of the vehicle from the underground parking lot entrance to the starting point of the underground mapping is also saved, that is, the poses of each point on the trajectory before the starting point are saved relative to the underground parking lot entrance.

[0136] In summary, each underground parking lot map contains the "GPS signal before the parking lot entrance," the trajectory before the starting point, and the poses of each point on the underground parking lot map route relative to the starting point of the underground mapping, all corresponding to the underground parking lot map. Optionally, in addition to the above information, the created map may also include environmental features such as point cloud features and semantic features.

[0137] In underground parking scenarios, since the locations of points on the underground parking map routes of multiple coarse-matched maps include the locations of points including the map's starting point, the positioning initialization method provided in this application can complete positioning initialization based on either the map's starting point or any point on the map. The following explains these two cases:

[0138] In the first scenario, the positioning initialization is completed based on the mapping starting point.

[0139] In some embodiments, each point on the underground parking lot map route includes the starting point for underground mapping;

[0140] Based on the second location and the locations of points along the underground parking lot map route obtained from multiple pre-acquired coarse-matching maps, multiple maps to be finely matched are selected from the multiple coarse-matching maps, including:

[0141] Obtain the poses of the underground mapping starting points relative to the entrance of the underground parking lot from multiple coarsely matched maps;

[0142] Based on the second location and the location of the underground mapping starting point relative to the entrance of the underground parking lot from multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps.

[0143] In some embodiments, obtaining the poses of the underground mapping starting points of multiple coarse-matched maps relative to the entrance of an underground parking lot includes:

[0144] Obtain the trajectory of the vehicle from the entrance of the underground parking lot to the starting point of the underground mapping;

[0145] Based on the trajectory before the starting point, the pose of the underground mapping starting point relative to the entrance of the underground parking lot is obtained.

[0146] In some embodiments, since the trajectory of a vehicle from the entrance of the underground parking lot to the starting point of the underground mapping is saved during the mapping process, the pose of the starting point of the underground mapping relative to the entrance of the underground parking lot can be obtained.

[0147] During the actual parking phase, the real-time position of the vehicle during its journey can be obtained, specifically the second position relative to the entrance of the underground parking lot. Both the mapping starting point and the second position are relative to the entrance of the underground parking lot. By comparing the distance between the mapping starting point and the second position to see if it is less than a preset first distance threshold, it is determined whether the coarse matching map corresponding to the mapping starting point can be filtered into a map for fine matching. After filtering out the maps for fine matching in the above manner, the target fine matching map and the initial pose of the vehicle on the target fine matching map are then determined.

[0148] The above method is applicable when a vehicle is very close to the mapping starting point as soon as it enters the parking lot. The mapping starting point can be found quickly. Since only the distance between the second position and the mapping starting point needs to be calculated, the amount of calculation is small, the initial positioning time is short, the efficiency is high, and the user experience is good.

[0149] In the second scenario, positioning initialization is completed based on any point on the map route.

[0150] In some embodiments, based on the second location and the locations of points on the underground parking lot map route obtained in advance from multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps, including:

[0151] The poses of the underground mapping starting points of multiple coarse matching maps relative to the underground parking lot entrance are obtained; and the poses of each point on the underground parking lot map line relative to the underground mapping starting point are obtained. Through position transformation, the positions of each point on the underground parking lot map line relative to the underground parking lot entrance are obtained.

[0152] Based on the second location and the position of each point on the underground parking lot map line relative to the underground parking lot entrance, multiple maps to be finely matched are selected from multiple coarse matching maps.

[0153] In one embodiment, the distance between the vehicle's second position and all points P on the map route from the mapping start point to the mapping end point is calculated. It is then determined whether the vehicle is located on the map route or is infinitely close to the map route. If so, the map corresponding to the point P that is close to the second position is taken as the map to be finely matched.

[0154] The above method applies when the vehicle's trajectory after entering the underground parking lot differs from the trajectory during mapping, making it impossible to quickly find the mapping starting point. In this case, any point on the map line whose distance from the second location is less than a preset first distance threshold can be found, and then positioning initialization can be completed based on that point. That is, the positioning initialization method provided in this application does not rely on the driving trajectory during the parking phase being the same as the driving trajectory during the mapping phase, nor does it rely on finding the mapping starting point. As long as the vehicle is close to any point on the map line (i.e., there exists a point on the map line whose distance from the second location is less than the preset first distance threshold), positioning initialization can be completed based on that point. This expands the scenarios for parking applications and improves the user experience.

[0155] Based on the same inventive concept, embodiments of this application provide a positioning initialization device, see [link to relevant documentation]. Figure 6 As shown, Figure 6 This is a schematic diagram of a positioning initialization device according to an embodiment of this application. The positioning initialization device includes:

[0156] The first acquisition module 10 is configured to acquire the first satellite search count and the current positioning signal when the vehicle travels to the first position;

[0157] The coarse matching module 20 is configured to, in response to a first number of satellite searches exceeding a preset threshold, select multiple coarse matching maps from multiple initial maps based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps respectively; wherein the difference between the positioning signal corresponding to the coarse matching map and the current positioning signal is less than a preset first signal threshold.

[0158] The second acquisition module 30 is configured to acquire the second satellite search count when the vehicle travels to the second position;

[0159] The first fine matching module 40a is configured to, in response to the second number of satellite searches not being greater than a quantity threshold, filter out multiple maps to be finely matched from multiple coarse matching maps based on the second location and the locations of each point on the underground parking lot map line of multiple pre-acquired coarse matching maps; wherein, among the locations of each point on the map line of the maps to be finely matched, there are point locations whose distance from the second location is less than a preset first distance threshold.

[0160] The third acquisition module 50 is configured to acquire the current environmental characteristics of the vehicle.

[0161] The pose determination module 60 is configured to determine the target fine-matching map and the vehicle's initial pose on the target fine-matching map based on the vehicle's current environmental characteristics and multiple maps to be fine-matched.

[0162] In some embodiments, the positioning initialization apparatus further includes:

[0163] The second fine matching module 40b is configured to, in response to a second number of satellite searches exceeding a quantity threshold, select multiple maps to be finely matched from multiple coarse matching maps based on the current positioning signal and pre-acquired positioning signals corresponding to multiple coarse matching maps respectively; wherein the difference between the positioning signal corresponding to the map to be finely matched and the current positioning signal is less than a preset second signal threshold; and the first signal threshold is greater than the second signal threshold.

[0164] In some embodiments, the pose determination module 60 is specifically configured as follows:

[0165] Based on the vehicle's current environmental characteristics and multiple maps to be matched, the vehicle's pose in the multiple maps to be matched is determined, and multiple matching scores are output.

[0166] If the matching score is greater than the preset threshold score, the map to be finely matched and the vehicle's pose on the map to be finely matched, which correspond to the matching score greater than the threshold score, are respectively determined as the target finely matched map and the initial pose of the vehicle on the target finely matched map.

[0167] In some embodiments, the coarse matching module 20 is specifically configured as follows:

[0168] For each of the multiple positioning signals corresponding to the multiple initial maps, the following operations are performed: calculate the longitude difference between the longitude of the current positioning signal and the longitude of each positioning signal, and calculate the latitude difference between the latitude of the current positioning signal and the latitude of each positioning signal.

[0169] Based on the longitude and latitude differences, determine whether the initial map is a coarse-match map;

[0170] If, among the positioning signals corresponding to the initial map, there exists a positioning signal with a longitude difference less than a preset longitude threshold and a latitude difference less than a preset latitude threshold, then the initial map is determined to be a coarse matching map.

[0171] The first signal threshold includes: longitude threshold and latitude threshold.

[0172] In some embodiments, the positioning initialization apparatus further includes a first mapping module, which is configured to:

[0173] From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location;

[0174] If the number of satellites searched in the third search exceeds the quantity threshold, the positioning signal is recorded and saved.

[0175] In response to the mapping command, a map of the above-ground parking lot is created using the vehicle's current location as the starting point for mapping on the ground, and the positioning signals of each point on the above-ground map route are saved. The initial map includes: a map of the above-ground parking lot; the positioning signals corresponding to the initial map are the positioning signals along the route traversed by the vehicle from the start of vehicle startup to the end of mapping.

[0176] In some embodiments, the positioning initialization apparatus further includes a second mapping module, which is configured to:

[0177] From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location;

[0178] If the number of satellites searched in the third search exceeds the quantity threshold, the positioning signal is recorded and saved.

[0179] During vehicle movement, if the number of satellites detected in the third phase does not exceed the threshold, the vehicle's current location will be used as the entrance to the underground parking lot, and the recording of the location signal will stop.

[0180] When a vehicle enters from the underground parking lot entrance, in response to the mapping command, a map of the underground parking lot is created using the vehicle's current position as the starting point for underground mapping, and the poses of each point on the underground parking lot map route relative to the starting point of underground mapping are saved; wherein, the initial map includes: the map of the underground parking lot; the positioning signal corresponding to the initial map is: the positioning signal on the route traversed by the vehicle from the start of vehicle startup until the third satellite search number is not greater than the quantity threshold.

[0181] In some embodiments, the first fine matching module 40a is specifically configured as follows:

[0182] Based on the second location and the locations of points along the underground parking lot map route obtained from multiple pre-acquired coarse-matching maps, multiple maps to be finely matched are selected from the multiple coarse-matching maps, including:

[0183] Obtain the poses of the underground mapping starting points relative to the entrance of the underground parking lot from multiple coarsely matched maps;

[0184] Based on the second location and the location of the underground mapping starting point relative to the entrance of the underground parking lot from the multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps.

[0185] In some embodiments, the first fine matching module 40a is further configured as follows:

[0186] Obtain the trajectory of the vehicle from the entrance of the underground parking lot to the starting point of the underground mapping;

[0187] Based on the trajectory before the starting point, the pose of the underground mapping starting point relative to the entrance of the underground parking lot is obtained.

[0188] In some embodiments, the first fine matching module 40a is further configured as follows:

[0189] The poses of the underground mapping starting points of multiple coarse matching maps relative to the underground parking lot entrance are obtained; and the poses of each point on the underground parking lot map line relative to the underground mapping starting point are obtained. Through position transformation, the positions of each point on the underground parking lot map line relative to the underground parking lot entrance are obtained.

[0190] Based on the second location and the position of each point on the underground parking lot map line relative to the underground parking lot entrance, multiple maps to be finely matched are selected from multiple coarse matching maps.

[0191] The method described in this application has the following beneficial effects:

[0192] 1. This embodiment first uses GPS signals to determine whether the vehicle has arrived near a parking lot along a learned route. If it's a surface parking lot, GPS signals are used to determine if the vehicle is approaching or on the map route; if it's an underground parking lot, the vehicle's current location is compared with the locations of points on the map route to determine if the vehicle is approaching or on the map route. Map feature matching only begins when the vehicle arrives at or is near the map route, significantly reducing computational load and improving positioning speed and accuracy.

[0193] 2. In the underground parking scenario, since it is only necessary to determine the vehicle's current position and the positions of various points on the map route, a point infinitely close to the vehicle's current position can be found on the map route, and positioning initialization can be completed based on that point. Therefore, the trajectory of the vehicle entering the underground parking lot does not need to be exactly the same as that during mapping, and it is not necessary to find the mapping starting point; it is sufficient for the vehicle to reach any point near the map route. Currently, many memory parking positioning initializations can only be completed at the mapping starting point, but the embodiments of this application can also realize positioning initialization from any point on the map route from the mapping starting point to the mapping ending point.

[0194] 3. Since the driving trajectory during the parking phase of this application does not need to be consistent with the trajectory during mapping, nor does it need to find the mapping starting point, multiple different routes can be learned in the same parking lot in stages, or a second route can be learned without leaving the parking lot after learning one route, making it more flexible and practical.

[0195] 4. By storing the location signal (GPS signal) corresponding to the map during the mapping stage, it can be applied to both above-ground and underground parking scenarios, and is not limited to underground parking lots.

[0196] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method as described in any of the above embodiments.

[0197] This application also provides a storage medium carrying one or more computer programs, which, when executed by a processor, implement the steps of the method described above.

[0198] This application also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method described above.

[0199] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for positioning initialization, characterized in that, For parking vehicles, the method includes: Obtain the first satellite count and current positioning signal when the vehicle reaches its first location; In response to the first number of satellite searches exceeding a preset threshold, based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps, multiple coarse matching maps are selected from the multiple initial maps; wherein, the difference between the positioning signal corresponding to the coarse matching map and the current positioning signal is less than a preset first signal threshold; the initial maps include maps of above-ground parking lots and maps of underground parking lots; Obtain the second satellite search count when the vehicle reaches the second position; In response to the second number of satellite searches not being greater than the quantity threshold, based on the second location and the locations of each point on the underground parking lot map route obtained from multiple pre-acquired coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps; wherein, among the locations of each point on the map route of the maps to be finely matched, there are points whose distance from the second location is less than a preset first distance threshold, and when the second number of satellite searches is not greater than the quantity threshold, the vehicle reaches the underground parking lot; Obtain the current environmental characteristics of the vehicle; Based on the vehicle's current environmental characteristics and multiple maps to be precisely matched, a target precise matching map and the vehicle's initial pose on the target precise matching map are determined.

2. The method according to claim 1, characterized in that, Also includes: In response to the second number of satellite searches being greater than the quantity threshold, based on the current positioning signal and the pre-acquired positioning signals corresponding to the multiple coarse matching maps respectively, multiple maps to be finely matched are selected from the multiple coarse matching maps; wherein, the difference between the positioning signal corresponding to the map to be finely matched and the current positioning signal is less than a preset second signal threshold; the first signal threshold is greater than the second signal threshold.

3. The method according to claim 1, characterized in that, Based on the vehicle's current environmental characteristics and multiple maps to be precisely matched, the target precise matching map and the vehicle's initial pose on the target precise matching map are determined, including: Based on the current environmental characteristics of the vehicle and multiple maps to be matched, the vehicle's pose in the multiple maps to be matched is determined, and multiple matching scores are output. If the matching score is greater than a preset threshold score, then the map to be finely matched corresponding to the matching score greater than the threshold score and the vehicle's pose on the map to be finely matched are respectively determined as the target finely matched map and the vehicle's initial pose on the target finely matched map.

4. The method according to claim 1, characterized in that, Based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps, multiple coarse matching maps are selected from the multiple initial maps, including: For each of the multiple positioning signals corresponding to the multiple initial maps, the following steps are performed: calculate the longitude difference between the longitude of the current positioning signal and the longitude of each positioning signal, and calculate the latitude difference between the latitude of the current positioning signal and the latitude of each positioning signal. Based on the longitude difference and the latitude difference, determine whether the initial map is a coarse matching map; If, among the positioning signals corresponding to the initial map, there exists a positioning signal with a longitude difference less than a preset longitude threshold and a latitude difference less than a preset latitude threshold, then the initial map is determined to be a coarse matching map. The first signal threshold includes the longitude threshold and the latitude threshold.

5. The method according to claim 2, characterized in that, Acquiring an initial map and the positioning signal corresponding to the initial map includes: From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location; If the third number of satellite searches is greater than the number threshold, then the positioning signal is recorded and saved; In response to the mapping command, a map of the above-ground parking lot is created using the current position of the vehicle as the starting point for mapping on the ground, and the positioning signals of each point on the above-ground map route are saved; wherein, the initial map includes: the map of the above-ground parking lot; the positioning signals corresponding to the initial map are the positioning signals on the route traversed by the vehicle from the start of vehicle startup to the end of mapping.

6. The method according to claim 1, characterized in that, Acquiring the initial map, the positioning signal corresponding to the initial map, and the positions of each point on the map route of the underground parking lot, including: From the moment the vehicle starts, acquire the third satellite count and positioning signal of the vehicle at its current location; If the third number of satellite searches is greater than the number threshold, then the positioning signal is recorded and saved; During the vehicle's movement, if the third number of satellite searches is not greater than the number threshold, then the vehicle's current position is taken as the entrance to the underground parking lot, and the recording of the positioning signal is stopped. When a vehicle enters from the underground parking lot entrance, in response to a mapping command, a map of the underground parking lot is created using the vehicle's current position as the starting point for underground mapping, and the poses of each point on the underground parking lot map route relative to the starting point are saved; wherein, the initial map includes: the map of the underground parking lot; the positioning signal corresponding to the initial map is: the positioning signal on the route traversed by the vehicle from the start of vehicle startup until the third satellite search number is not greater than the number threshold.

7. The method according to claim 6, characterized in that, Each point on the underground parking lot map route includes the starting point of the underground mapping; Based on the second location and the locations of points along the underground parking lot map route obtained from multiple pre-acquired coarse-matching maps, multiple maps to be finely matched are selected from the multiple coarse-matching maps, including: Obtain the poses of the underground mapping starting points relative to the entrance of the underground parking lot from multiple coarsely matched maps; Based on the second location and the location of the underground mapping starting point relative to the entrance of the underground parking lot from the multiple coarse matching maps, multiple maps to be finely matched are selected from the multiple coarse matching maps.

8. The method according to claim 7, characterized in that, Obtain the poses of the underground mapping starting points relative to the entrance of the underground parking lot from multiple coarsely matched maps, including: Obtain the vehicle's trajectory from the entrance of the underground parking lot to the starting point of the underground mapping; Based on the trajectory before the starting point, the pose of the underground mapping starting point relative to the entrance of the underground parking lot is obtained.

9. The method according to claim 6, characterized in that, Based on the second location and the locations of points along the underground parking lot map route obtained from multiple pre-acquired coarse-matching maps, multiple maps to be finely matched are selected from the multiple coarse-matching maps, including: The poses of the underground mapping starting points of multiple coarse matching maps relative to the entrance of the underground parking lot are obtained; and the poses of each point on the underground parking lot map line relative to the underground mapping starting points are obtained. Through position transformation, the positions of each point on the underground parking lot map line relative to the entrance of the underground parking lot are obtained. Based on the second location and the location of each point on the underground parking lot map line relative to the entrance of the underground parking lot, multiple maps to be finely matched are selected from the multiple coarse matching maps.

10. A positioning initialization device, characterized in that, For vehicle parking, the device includes: The first acquisition module is configured to acquire the first satellite search count and the current positioning signal when the vehicle travels to the first position; The coarse matching module is configured to, in response to the first number of satellite searches being greater than a preset threshold, filter out multiple coarse matching maps from the multiple initial maps based on the current positioning signal and pre-acquired positioning signals corresponding to multiple initial maps respectively; wherein, the difference between the positioning signal corresponding to the coarse matching map and the current positioning signal is less than a preset first signal threshold; the initial maps include maps of above-ground parking lots and maps of underground parking lots. The second acquisition module is configured to acquire the second number of satellite searches when the vehicle travels to the second position; The first fine-matching module is configured to, in response to the second satellite search count not being greater than the quantity threshold, filter out multiple maps to be fine-matched from the multiple coarse-matching maps based on the second location and the locations of each point on the underground parking lot map route of multiple pre-acquired coarse-matching maps; wherein, among the locations of each point on the map route of the maps to be fine-matched, there are points whose distance from the second location is less than a preset first distance threshold; when the second satellite search count is not greater than the quantity threshold, the vehicle reaches the underground parking lot; The third acquisition module is configured to acquire the current environmental characteristics of the vehicle. The pose determination module is configured to determine a target fine-matching map and the initial pose of the vehicle on the target fine-matching map based on the current environmental characteristics of the vehicle and multiple maps to be finely matched.

Images