Car searching map generation method, car searching method, device, equipment and medium

Abstract

The present disclosure relates to a car searching map generation method, a car searching method, a device, an equipment and a medium. The method comprises: acquiring a plurality of first hotspots searched by a terminal at a plurality of sampling positions in a moving route, the terminal moving relative to a vehicle; clustering the plurality of first hotspots according to coordinate information of the plurality of sampling positions and intensity information of the plurality of first hotspots, to obtain a clustering map comprising a plurality of hotspot areas, wherein different hotspot areas comprise different hotspots, and the clustering map comprises position information of the vehicle. The present disclosure clusters a plurality of first hotspots searched by a terminal at a plurality of sampling positions in a moving route relative to a vehicle, to obtain a clustering map comprising a plurality of hotspot areas, without laying equipment in advance and collecting data manually, thereby saving equipment cost and labor cost and reducing car searching cost.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a method for generating a vehicle-finding map, a vehicle-finding method, an apparatus, a device, and a medium. Background Technology

[0002] In existing technologies, the data used for car-finding methods is obtained by deploying specific Bluetooth or Ultra Wide Band (UWB) devices at various locations in the parking lot, which is costly due to the high cost of deploying such devices in advance; or it is obtained through location-based services (LBS), which requires users to manually collect multiple data points, resulting in high labor costs.

[0003] Therefore, there is an urgent need for a method to generate car-finding maps to solve the above problems. Summary of the Invention

[0004] To solve the above-mentioned technical problems, or at least partially solve them, this disclosure provides a vehicle-finding map generation method, vehicle-finding method, apparatus, equipment, and medium to reduce vehicle-finding costs.

[0005] In a first aspect, embodiments of this disclosure provide a method for generating a vehicle-finding map, including:

[0006] The terminal acquires multiple first hotspots at multiple sampling locations along the moving route, the terminal being moved relative to the vehicle;

[0007] Based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots, the multiple first hotspots are clustered to obtain a cluster map including multiple hotspot regions, wherein different hotspot regions include different hotspots, and the cluster map includes the location information of the vehicle.

[0008] Secondly, embodiments of this disclosure provide a vehicle locating method, including:

[0009] Acquire multiple secondary hotspots around the terminal;

[0010] Based on the clustering maps of the multiple second hotspots and the first hotspot, the feature vectors of the second hotspots are determined. The clustering map of the first hotspot includes the hotspot center of the hotspot region and the multiple second hotspots.

[0011] Based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle location, information guiding the terminal to move toward the vehicle is generated.

[0012] Thirdly, embodiments of this disclosure provide a vehicle-finding map generation device, comprising:

[0013] The first acquisition module is used to acquire multiple first hotspots searched by the terminal at multiple sampling locations in the moving route, wherein the terminal is moving relative to the vehicle;

[0014] The clustering module is used to cluster the multiple first hotspots based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots to obtain a clustered map including multiple hotspot regions, wherein different hotspot regions include different hotspots, and the clustered map includes the location information of the vehicle.

[0015] Fourthly, embodiments of this disclosure provide a vehicle finding device, including:

[0016] The second acquisition module is used to acquire multiple secondary hotspots around the terminal;

[0017] The determination module is used to determine the feature vector of the second hotspot based on the clustering map of the plurality of second hotspots and the first hotspot, wherein the clustering map of the first hotspot includes the hotspot center of the hotspot region and the plurality of second hotspots;

[0018] The generation module is used to generate information guiding the terminal to move toward the vehicle based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle location.

[0019] Fifthly, embodiments of this disclosure provide an electronic device, including:

[0020] Memory;

[0021] Processor; and

[0022] Computer programs;

[0023] The computer program is stored in the memory and configured to be executed by the processor to implement the method as described in the first or second aspect.

[0024] In a sixth aspect, embodiments of this disclosure provide a non-volatile computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method described in the first or second aspect.

[0025] In a seventh aspect, embodiments of this disclosure also provide a computer program product comprising a computer program or instructions that, when executed by a processor, implement the method described in the first or second aspect.

[0026] The vehicle-finding map generation method, vehicle-finding method, device, equipment, and medium provided in this disclosure acquire multiple first hotspots found by a terminal at multiple sampling locations along a moving route, with the terminal moving relative to the vehicle; based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots, the multiple first hotspots are clustered to obtain a clustered map including multiple hotspot regions, wherein different hotspot regions include different hotspots, and the clustered map includes the vehicle's location information. Compared with the prior art of finding vehicles by deploying specific Bluetooth or UWB devices at various locations in a parking lot, this disclosure embodiment does not require the prior deployment of equipment, saving equipment costs. Compared with the prior art of finding vehicles through indoor positioning and navigation services, this disclosure embodiment does not require manual data collection, saving labor costs and reducing vehicle-finding costs. Attached Figure Description

[0027] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0028] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of a vehicle-finding map generation method provided in an embodiment of this disclosure;

[0030] Figure 2 A schematic diagram illustrating an application scenario provided by an embodiment of this disclosure;

[0031] Figure 3 A schematic diagram of the height change curve provided for an embodiment of this disclosure;

[0032] Figure 4 A schematic diagram illustrating the acquisition of a first hotspot as provided in an embodiment of this disclosure;

[0033] Figure 5 This is a flowchart of a vehicle-finding method provided in an embodiment of the present disclosure;

[0034] Figure 6 This is a flowchart of a vehicle-finding method provided in an embodiment of the present disclosure;

[0035] Figure 7 A flowchart illustrating the selection of Sigma points provided in this embodiment of the disclosure;

[0036] Figure 8 This is a schematic diagram of a car-finding scenario provided in an embodiment of this disclosure;

[0037] Figure 9 This is a schematic diagram of the vehicle-finding map generation device provided in an embodiment of the present disclosure;

[0038] Figure 10 This is a schematic diagram of the vehicle finding device provided in an embodiment of the present disclosure;

[0039] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0040] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0041] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0042] In existing technologies, the data used for vehicle location methods is obtained by deploying specific Bluetooth or Ultra Wide Band (UWB) devices at various locations in the parking lot, which incurs high costs due to the need for advance device deployment; or by obtaining data through location-based services (LBS), which requires users to manually collect multiple data points, resulting in high labor costs. To address this issue, this disclosure provides a method for generating vehicle location maps, which will be described below with reference to specific embodiments.

[0043] Figure 1 This is a flowchart illustrating a vehicle-finding map generation method provided in this embodiment. The method can be executed by a vehicle-finding map generation device, which can be implemented using software and / or hardware. This device can be configured in an electronic device, such as a server or terminal, where the terminal specifically includes a mobile phone, computer, or tablet computer. Furthermore, this method can be applied to various scenarios for vehicle-finding map generation, such as supermarkets, shopping malls, and office buildings. It is understood that the vehicle-finding map generation method provided in this embodiment can also be applied to other scenarios.

[0044] The following is about Figure 1 The method for generating a car-finding map is described below, and the specific steps involved are as follows:

[0045] S101. Obtain multiple first hotspots searched by the terminal at multiple sampling locations along the moving route, wherein the terminal is moving relative to the vehicle.

[0046] When the terminal moves relative to the vehicle, it acquires multiple sampling locations along the moving route and searches for multiple primary hotspots at each sampling location.

[0047] For example, such as Figure 2 As shown, after the user parks in the parking lot (i.e., vehicle 21 is in P gear and the engine is off), vehicle 21 transmits the user's exit information to terminal 20 via Flymelink (Flymelink is a proprietary protocol that establishes a wireless communication connection between vehicle 21 and terminal 20). Terminal 20 receives the parking notification from vehicle 21 and sets the position of vehicle 21 as O. When the user carries terminal 20 from O to E, for example, the movement route of terminal 20 can be OABCDE. The multiple sampling positions of this movement route can be sampling position O, sampling position A, sampling position B, sampling position C, sampling position D, and sampling position E. These sampling positions can be set according to distance or other methods, which are not limited in this embodiment. Multiple first hotspots are searched at sampling positions O, sampling position A, sampling position B, sampling position C, sampling position D, and sampling position E respectively. Each first hotspot includes a first hotspot name, first hotspot signal strength, first hotspot unique identifier, and first hotspot coordinates.

[0048] Optionally, obtaining multiple first hotspots found by the terminal at multiple sampling locations along the movement route includes: obtaining multiple original hotspots found by the terminal at multiple sampling locations along the movement route; filtering the multiple original hotspots at each sampling location according to a preset rule to obtain the multiple first hotspots.

[0049] Specifically, after the user parks in the parking lot (i.e., the vehicle 21 is in P gear and the engine is off), the vehicle 21 transmits the user's exit information to the terminal 20 via Flymelink. The terminal 20 receives the parking notification from the vehicle 21. When the terminal 20 moves relative to the vehicle 21, such as the movement route of the terminal 20 being OABCDE, the terminal 20 obtains multiple sampling locations in the movement route, such as sampling location O, sampling location A, sampling location B, sampling location C, sampling location D, and sampling location E. At each sampling location, multiple original hotspots are searched. These multiple original hotspots include terminal hotspots and fixed hotspots in the parking lot. According to preset rules, multiple original hotspots at each sampling location are filtered to obtain multiple first hotspots at each sampling location.

[0050] Optionally, obtaining multiple original hotspots searched by the terminal at multiple sampling locations along the movement route includes: in response to the vehicle parking, obtaining the vehicle's location information, using the vehicle's location as a first sampling location, the first sampling location belonging to the multiple sampling locations; obtaining the terminal's location information; and obtaining multiple original hotspots searched by the vehicle and the terminal at multiple sampling locations along the movement route based on the first sampling location and the terminal's location information.

[0051] Specifically, after the user parks in the parking lot (i.e., the vehicle 21 is in P gear and the engine is off), the vehicle 21 transmits the user's exit information to the terminal 20 via Flymelink. The terminal 20 receives the parking notification from the vehicle 21, obtains the location information of the vehicle 21, and uses the location of the vehicle 21 as the first sampling location O. This first sampling location O belongs to multiple sampling locations. The terminal 20 obtains the location information E. Based on the first sampling location (the location information O of the vehicle 21) and the location information E of the terminal 20, the terminal 20 obtains multiple original hotspots searched by the vehicle 21 and the terminal 20 at multiple sampling locations (such as sampling location O, sampling location A, sampling location B, sampling location C, sampling location D, and sampling location E) in the movement route OABCDE.

[0052] Optionally, obtaining the location information of the terminal includes: determining the terminal coordinates based on the vehicle's location information using pedestrian dead reckoning and / or hotspot positioning algorithms, and determining the floor where the terminal is located using a barometric pressure sensor.

[0053] Specifically, the terminal 20 determines its coordinates by pedestrian dead reckoning based on the location information of the vehicle 21, and / or the terminal 20 determines its coordinates by hotspot positioning algorithm based on the location information of the vehicle 21, and determines the floor where the terminal 20 is located by barometric pressure sensor.

[0054] Optionally, the pedestrian dead reckoning is calculated based on the current position, orientation angle, and stride length to determine the next position;

[0055] Specifically, Pedestrian Dead Reckoning (PDR) involves measuring and statistically analyzing a pedestrian's steps, stride length, and direction to deduce their walking trajectory and location.

[0056] Specifically, determining the terminal coordinates based on the vehicle's location information using pedestrian dead reckoning and / or hotspot positioning algorithms includes: acquiring the direction angle using a direction sensor; determining whether the user's position has changed using an acceleration sensor; and determining the terminal coordinates based on the change in the user's position, according to the vehicle coordinates, the direction angle, and the step length.

[0057] Specifically, terminal 20 is equipped with a direction sensor and an acceleration sensor. Terminal 20 obtains the direction angle through the direction sensor and determines whether the user's position has changed through the acceleration sensor. When the user's position changes, the coordinates of terminal 20 are determined based on the vehicle 21 coordinates, direction angle, step length, and number of steps. For example, for ease of calculation, the vehicle coordinate O can be set to (0,0,0). Based on the vehicle 21 coordinates, the direction angle at point O, and the user's step length, the user's first step position is determined. Similarly, based on the user's first step position coordinates, first step direction angle, and user step length, the user's next step position is determined. The user's next step position is then updated to the user's first step position. This process is repeated until the next step position matches the coordinates of terminal 20. It is understood that, generally, the user's step length is 0.45-0.5 times their height. The user's step length can be calculated based on their height, i.e., user height * 0.45 < user step length < user height * 0.5. For ease of calculation, the user's step length can also be set according to actual conditions, such as a user step length of 50cm.

[0058] Optionally, determining the floor where the terminal is located using a barometric pressure sensor includes: calculating the height difference between the vehicle and the terminal using the barometric pressure sensor, and determining the floor where the terminal is located based on the height difference and the floor where the vehicle is located.

[0059] Specifically, after the user parks in the parking lot (i.e., vehicle 21 is in P gear and the engine is off), vehicle 21 transmits the user's exit information to terminal 20 via Flymelink. Terminal 20 receives the parking notification from vehicle 21. The air pressure sensor on terminal 20 reads the air pressure of vehicle 21 when it is parked and uses this air pressure as the air pressure p1 of the floor where vehicle 21 is located. The air pressure sensor on terminal 20 reads the air pressure p2 of the floor where terminal 20 is located. The height difference between vehicle 21 and terminal 20 is calculated based on the air pressure difference between the floor where vehicle 21 is located and the floor where terminal 20 is located. Based on the height difference, the floor height, and the floor where vehicle 21 is located, the floor of terminal 20 is determined.

[0060] After the user parks in the parking lot (i.e., vehicle 21 is parked), terminal 20 receives a parking notification from vehicle 21 (terminal 20 detects that vehicle 21 is in P gear). The air pressure sensor interface on terminal 20 reads the air pressure of vehicle 21 when it is parked, and uses this air pressure as the air pressure p1 of the floor where vehicle 21 is located. The air pressure sensor on terminal 20 reads the air pressure p2 of the floor where terminal 20 is located. Through the native Android interface SensorManager.getAltitude(SensorManager.PRESSURE_STANDARD_ATMOSPHERE, p), the height of the floor where vehicle 21 is located and the height of the floor where terminal 20 is located are calculated, and the height change curve of terminal 20 is calculated. The height change curve is as follows. Figure 3 As shown, analyzing the height change curve, if the height difference of the height change curve within a preset time is greater than a preset distance, then the user is going up or down stairs, as... Figure 3 The time periods t1-t2, t3-t4, and t5-t6 shown represent the user going up and down stairs. It's understandable that the preset time and distance can be set based on the user's actual speed going up and down stairs or the speed of the elevator. This can be flexibly applied according to the actual situation. After the user finishes going up and down stairs, terminal 20 can calculate the floor difference between terminal 20 and vehicle 21 based on the height difference between terminal 20 and vehicle 21. The floor height can be calculated based on the actual situation or a preset value, which may depend on the preset space type and relevant building codes. The specific calculation method for the floor difference is: Floor difference = Height difference / Floor height. Generally (i.e., when the user is not in the elevator or on the stairs), the floor difference is an integer. If the floor difference is not an integer, it can be rounded to the nearest integer. The floor of terminal 20 is calculated based on the floor difference between terminal 20 and vehicle 21, and the floor of vehicle 21.

[0061] Optionally, the preset rules include: the hotspot signal strength at multiple sampling locations is continuously greater than a preset strength threshold, and / or the hotspot name is a preset model name, and / or the number of preset digits of the hotspot's unique identifier is a preset value.

[0062] Specifically, hotspots whose signal strength consistently exceeds a preset strength threshold at multiple sampling locations can be considered as following hotspot signals. For example, if a user is at sampling locations A, B, C, and D, the hotspot signal strength consistently exceeds the preset strength threshold. The preset strength threshold can be -50 or other relatively strong signal strengths, and can be set according to actual conditions; this embodiment does not limit this setting.

[0063] Specifically, the hotspot name is the preset model name, which can be the terminal model.

[0064] Specifically, the preset number of digits for the hotspot unique identifier BSSID is a preset value. For example, if the preset number of digits is the first two digits, the preset value is 92, and the hotspot unique identifier BSSID is: 92:f0:52:3c:61:6f.

[0065] Optionally, the plurality of original hotspots at each sampling location are filtered according to preset rules to obtain the plurality of first hotspots, including: removing hotspots whose hotspot signal strength is continuously greater than a preset strength threshold at multiple sampling locations, and / or removing hotspots whose hotspot names are preset model names, and / or removing hotspots whose unique hotspot identifiers have a preset number of digits.

[0066] Specifically, such as Figure 4 As shown, among multiple original hotspots 40, the terminal removes hotspots 41 whose hotspot signal strength is continuously greater than a preset strength threshold at multiple sampling locations, and / or removes hotspots 42 whose hotspot names are preset model names, and / or removes hotspots 43 whose hotspot unique identifiers have preset digits and preset values, thus obtaining multiple first hotspots 44.

[0067] S102. Based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hot spots, the multiple first hot spots are clustered to obtain a cluster map including multiple hot spot regions, wherein different hot spot regions include different hot spots, and the cluster map includes the location information of the vehicle.

[0068] The terminal clusters multiple primary hotspots based on the coordinate information of multiple sampling locations and the intensity information of multiple primary hotspots at each sampling location, resulting in a cluster map that includes multiple hotspot regions. Different hotspot regions include different hotspots. The cluster map includes vehicle location information, which can be the starting sampling location. The cluster map is then uploaded to a cloud database or a local database.

[0069] For example, the terminal acquires the coordinate information of sampling positions O, A, B, C, D, and E. The coordinate information of each sampling position can be obtained by pedestrian dead reckoning. Sampling position O can be the location of a vehicle. The terminal searches for multiple hotspots at each of the above sampling positions and determines the intensity information of the multiple first hotspots. Based on the coordinate information of the multiple sampling positions and the intensity information of the multiple first hotspots, the multiple first hotspots are clustered to obtain a clustered map including multiple hotspot regions. Different hotspot regions include different hotspots. It can be understood that a hotspot region can contain at least one sampling position.

[0070] This embodiment of the invention acquires multiple first hotspots found by the terminal at multiple sampling locations along the moving route, with the terminal moving relative to the vehicle. Based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots, the multiple first hotspots are clustered to obtain a clustered map including multiple hotspot regions. Different hotspot regions include different hotspots. The clustered map includes the vehicle's location information. Compared with the prior art of finding vehicles by deploying specific Bluetooth or UWB devices at various locations in the parking lot, this embodiment of the invention does not require the prior deployment of equipment, saving equipment costs. Compared with the prior art of finding vehicles through indoor positioning and navigation services, this embodiment of the invention does not require manual data collection, saving labor costs and reducing vehicle finding costs.

[0071] Based on the above embodiments, the multiple first hotspots are clustered according to the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots to obtain a clustered map including multiple hotspot regions. This includes: sorting the multiple sampling locations in chronological order to determine multiple first hotspots at each sampling location; for each sampling location, sorting the signal intensity of the multiple first hotspots in descending order, and determining the dimension of the target hotspot feature vector based on the signal intensity of the multiple first hotspots; constructing regional feature information based on the target hotspot feature vector to obtain a regional dataset of the first hotspots; and clustering the regional dataset of the first hotspots to obtain a clustered map including multiple hotspot regions.

[0072] The terminal sorts multiple sampling locations chronologically to identify multiple primary hotspots at each location. In other words, before the vehicle stops and is located, the sampling locations from the vehicle to the terminal are sorted chronologically to obtain the temporal order of each sampling location and its multiple primary hotspots. Following the temporal order of each sampling location, the signal strength of the multiple primary hotspots at each location is sorted in descending order. For example, if there are 10 primary hotspots at a sampling location, they are sorted from strongest to weakest signal strength. Based on the name and signal strength of the primary hotspot, the dimension of the target hotspot feature vector is determined. For instance, if a primary hotspot with a strength exceeding a certain threshold is selected from the 10 primary hotspots, the number of target hotspots to be extracted is determined, and a target hotspot feature vector is constructed. Regional feature information is constructed based on the target hotspot feature vector to obtain a regional dataset of primary hotspots. The regional dataset of primary hotspots is then clustered to obtain a clustered map including multiple hotspot regions.

[0073] Optionally, the method further includes: traversing the regional feature information to obtain first target region feature information corresponding to the first hotspot at the sampling location; extracting first hotspot features of the first hotspot at the sampling location through the first target region feature information; based on the value of the first hotspot feature, processing the first hotspot at the sampling location into a first hotspot feature vector according to the first target region feature information, and adding the first hotspot feature vector to the regional dataset corresponding to the first target region.

[0074] Specifically, the terminal traverses the regional feature information to obtain the first target region feature information corresponding to the first hotspot at the sampling location; it extracts the first hotspot feature of the first hotspot at the sampling location through the first target region feature information; when the first hotspot feature has a value, it processes the first hotspot at the sampling location into a first hotspot feature vector and the coordinates corresponding to the first hotspot feature vector according to the first target region feature information, and adds the first hotspot feature vector to the regional dataset corresponding to the first target region according to the coordinates corresponding to the first hotspot feature vector.

[0075] Optionally, the method further includes: based on the fact that all the first hotspot features are invalid values, sorting the signal strength of the first hotspots at the sampling locations corresponding to the first hotspot features in descending order, determining the dimension of the target hotspot feature vector to be extracted based on the number of first hotspots at the sampling locations; and updating the regional feature information constructed from the target hotspot feature vector to the regional dataset of the first hotspot.

[0076] Illegal values ​​are defined in the program. For example, if there are three primary hotspots in the constructed area feature information, and during the terminal's movement, there are multiple locations where none of the hotspots can be scanned, the hotspot may disappear due to factors such as wall obstruction, or it may disappear due to vehicle activity. In this case, the value of the hotspot in the primary hotspot feature constructed based on the area feature information is an illegal value of -999.

[0077] Specifically, when all the first hotspot features are invalid values, the signal strength of the first hotspot at the sampling location corresponding to the first hotspot feature is sorted in descending order. The dimension of the target hotspot feature vector is determined based on the number of first hotspots at the sampling location. The regional feature information constructed from the target hotspot feature vector is then updated to the regional dataset of the first hotspot.

[0078] For example, the terminal finds 30 first hotspots at sampling position (0, 0, 0). These 30 first hotspots are sorted according to their signal strength from strongest to weakest. When the feature vector dimension of the target hotspot is 5, the first 5 hotspots with the highest signal strength are extracted as target hotspots. For example, the target hotspots could be {1:10, 2:9, 3:11, 4:2, 5:1}. The feature vectors of these 5 target hotspots are used to construct regional feature information. When the terminal moves to sampling position (2, 0, 0), it receives 20 first hotspots. Since the sampling positions (0, 0, 0) and (2, 0, 0) are close, the 5 target hotspots extracted at sampling position (0, 0, 0) can also be partially found at sampling position (2, 0, 0). Therefore, based on the unique identifier of the 5 extracted target hotspots, the first 5 hotspots with the highest signal strength are selected as target hotspots at sampling position (2, 0, 0). Search for 5 target hotspots at sampling position (2, 0, 0). All 5 hotspots may have a signal, while some may have their fluctuations disappear due to obstruction from walls or other factors. When a hotspot's fluctuations disappear due to obstruction, its intensity is set to an invalid value of -999. Therefore, the target hotspots at sampling position (2, 0, 0) could be {1:10, 2:9, 3:11, 4:2, 5:-999}. The four hotspots with values ​​at sampling position (2, 0, 0) are processed into first hotspot feature vectors and their corresponding coordinates. These first hotspot feature vectors are then added to the region dataset constructed at sampling position (0, 0, 0) according to their coordinates. Therefore, sampling positions (0, 0, 0) and (2, 0, 0) can be considered the same region. When the terminal searches for hotspot signals at sampling position (5, 0, 0) based on the unique identifiers of the five target hotspots, the fluctuations of the five target hotspots at sampling position (5, 0, 0) all disappear because the sampling position (5, 0, 0) is far from the sampling position (0, 0, 0). In other words, the five target hotspots at sampling position (5, 0, 0) are all invalid values. Multiple first hotspots are searched at sampling position (5, 0, 0), and sorted according to the signal strength of the multiple first hotspots. After determining the dimension of the target hotspot feature vector, another regional feature information is constructed based on the target hotspot feature vector. Based on the regional feature information constructed by the terminal at multiple sampling positions, a regional dataset of the first hotspots is formed.

[0079] Optionally, clustering the regional dataset of the first hotspot to obtain a clustered map including multiple hotspot regions includes: determining the cluster points of the regional dataset of the first hotspot, calculating the K value, where the K value represents the number of hotspot centers in the hotspot region; selecting the center points of the clusters, and clustering the clusters using the K-nearest neighbor classification algorithm to obtain a clustered map including multiple hotspot regions.

[0080] K-nearest neighbor (KNN) is a classic classification algorithm that is parameter-free, performs well, and is simple.

[0081] Specifically, the terminal extracts the first and last sampling positions from the regional dataset of the first hotspot in chronological order, calculates the user's walking distance, and determines the cluster points of the regional dataset of the first hotspot according to a certain distance. For example, the certain distance can be 5 meters, until the distance between the last cluster point and the last sampling position is less than this certain distance. It is understood that the certain distance for different regional datasets can be the same or different, and can be set according to the actual situation. A K value is calculated, which represents the number of hotspot centers in the hotspot region. The order of the regional dataset is shuffled, and K cluster center points are randomly selected. The clusters are then clustered using the K-nearest neighbor classification algorithm to obtain a clustered map including multiple hotspot regions.

[0082] Optionally, the center point of a cluster is selected, and the clusters are clustered using the K-nearest neighbor classification algorithm to obtain a clustering map of the first hotspot. This includes: selecting the center point of the cluster; calculating the signal domain distance and the point distance between the center point of the cluster and the cluster points; calculating the Euclidean norm distance between the center point of the cluster and the cluster points based on the signal domain distance and the point distance; selecting the cluster point with the smallest Euclidean norm distance to add to the cluster and updating the cluster; calculating the average value of the cluster as the center point of the cluster, and repeatedly calculating the Euclidean norm distance between the center point of the cluster and the cluster points until the sum of the distances from the cluster points to the center points of the corresponding clusters is less than a preset threshold, thereby obtaining a clustering map of the first hotspot.

[0083] Specifically, K cluster center points are randomly selected, and the signal domain distance and point distance between each cluster center point and each cluster point are calculated. The signal domain distance refers to the signal difference. Based on the signal domain distance and point distance, a formula is constructed to calculate the Euclidean norm distance between the cluster center point and each cluster point. The specific formula is as follows:

[0084]

[0085] Among them, V i V represents the signal strength vector of the first hotspot received at point i. i =[v i,1 ,v i,2 ,.....],v i,1 This refers to the signal strength received at point i from the first hotspot signal transmission point, where P represents the cluster point.

[0086] Select the cluster point with the smallest Euclidean norm distance and add it to the cluster, then update the cluster. Calculate the cluster mean as the cluster center point. Repeat the calculation of the Euclidean norm distance between the cluster center point, the cluster center point, and the cluster point until the sum of the Euclidean norm distances from the cluster point to the corresponding cluster center point is less than a preset threshold. For example, this can be the sum of the Euclidean norm distances from two adjacent cluster points to the corresponding cluster center point being less than the preset threshold, which can be 5. Continue this process until all region datasets are clustered, resulting in the cluster map of the first hotspot. The calculation formula is as follows:

[0087]

[0088] Where E represents the sum of the Euclidean norm distances from cluster points to the center of the corresponding cluster, ||l-cc i ‖ represents the Euclidean norm distance from a cluster point to the cluster center, and cci represents the average value of cluster Ci.

[0089] This embodiment of the disclosure divides the first hotspot into regions and clusters the regional dataset of the first hotspot to obtain a clustered map including multiple hotspot regions, which facilitates users in finding their vehicles and improves their vehicle-finding efficiency.

[0090] In existing technologies, vehicle location typically employs several methods. One method involves partnering with shopping malls, which requires deploying specific Bluetooth or Ultra Wide Band (UWB) devices at various locations in the mall's parking lot. Location and navigation are determined by signal strength, a costly approach requiring both collaboration and pre-deployment of equipment. Another method utilizes Location-Based Services (LBS), where users manually input their location, collecting hotspot information from multiple locations and storing it in a database. Navigation is then based on this database data. However, this method requires frequent manual input and is susceptible to interference from mobile hotspots, potentially leading to inaccurate vehicle location calculations and low accuracy. To address these issues, this disclosure provides a vehicle location method, which will be described below with reference to specific embodiments.

[0091] Figure 5 This is a flowchart illustrating a vehicle-finding method provided in an embodiment of this disclosure. The method can be executed by a vehicle-finding device, which can be implemented using software and / or hardware. The vehicle-finding device can be configured in an electronic device, such as a server or terminal, where the terminal specifically includes a mobile phone, computer, or tablet computer. Furthermore, this method can be applied to vehicle-finding scenarios such as supermarkets, shopping malls, and office buildings. It is understood that the vehicle-finding method provided in this embodiment of the disclosure can also be applied to other scenarios.

[0092] The following is about Figure 5 The following describes the methods for finding your car, such as... Figure 5 As shown, the method includes the following steps:

[0093] S501: Obtain multiple secondary hotspots around the terminal.

[0094] The terminal collects multiple secondary hotspots around it. These secondary hotspots include: the name of the secondary hotspot, its signal strength, and its unique identifier. It is understood that when a user returns to their vehicle and retrieves these secondary hotspots, they will no longer acquire hotspots that fall under preset rules (i.e., hotspots whose signal strength continuously exceeds a preset strength threshold at multiple sampling locations, and / or whose names are preset model names, and / or whose unique identifiers have a preset number of digits), nor will they acquire the coordinates of the secondary hotspots.

[0095] S502. Based on the clustering map of the plurality of second hotspots and the first hotspot, determine the feature vector of the second hotspot. The clustering map of the first hotspot includes the hotspot center of the hotspot region and the plurality of second hotspots.

[0096] The terminal determines the feature vector of the second hotspot based on the cluster map of multiple second hotspots and the first hotspot. The cluster map of the first hotspot includes the hotspot center of the hotspot area and multiple second hotspots.

[0097] Optionally, based on the clustering map of the plurality of second hotspots and the first hotspot, the feature vector of the second hotspot is determined, including: matching the regional datasets of the plurality of second hotspots and the first hotspot, determining the second target region feature information of the terminal, and constructing the feature vector of the second hotspot.

[0098] Specifically, the terminal matches multiple second hotspot and first hotspot regional datasets to determine the second target region feature information of the terminal and constructs the second hotspot feature vector.

[0099] Optionally, matching the regional datasets of the plurality of second hotspots and the first hotspot to determine the second target region feature information of the terminal and constructing a second hotspot feature vector includes: extracting multiple second hotspot features from the plurality of second hotspots; determining the second target region corresponding to the plurality of second hotspots based on the values ​​of the multiple second hotspot features and according to the first hotspot region, and determining the second target region feature information of the terminal; and converting the second hotspot feature vector according to the second target region feature information of the terminal.

[0100] Specifically, the terminal extracts multiple second hotspot features from multiple second hotspots; based on the values ​​of the multiple second hotspot features, it determines the second target regions corresponding to the multiple second hotspots according to the first hotspot region, and determines the second target region feature information of the terminal; and it transforms the second hotspot feature vector according to the second target region feature information of the terminal.

[0101] Optionally, the method further includes: controlling the terminal interface to display an abnormality based on the fact that all of the plurality of second hotspot features are illegal values, so as to prompt the user to deviate from the planned trajectory.

[0102] Specifically, the terminal extracts multiple features of multiple secondary hotspots. When all features of multiple secondary hotspots are illegal values ​​of -999, it indicates that the user is not on the path for generating the clustering map. The terminal interface displays an abnormality and receives a NO_MATCH_AREA event to prompt the user that they are outside the range and have deviated from the planned trajectory.

[0103] S503. Based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle location, generate information to guide the terminal to move toward the vehicle.

[0104] Based on the feature vectors of multiple secondary hotspots, the hotspot center of the hotspot area, and the vehicle's location, the terminal generates information guiding the terminal to move toward the vehicle using hotspot positioning algorithms, pedestrian dead reckoning, and filtering algorithms.

[0105] Optionally, based on the feature vectors of the plurality of second hotspots, the hotspot center of the hotspot region, and the vehicle position, information guiding the terminal to move toward the vehicle is generated, including: obtaining the hotspot center of the second target region based on the second target region and the hotspot center of the hotspot region; calculating the distance between the feature vector of the second hotspot and the hotspot center of the second target region to determine the target cluster closest to the feature vector of the second hotspot; calculating the Euclidean distance between the feature vector of the second hotspot and the cluster points in the target cluster to obtain the coordinates of the K target cluster points closest to the Euclidean distance; calculating the terminal position based on the coordinates of the K target cluster points using a weighted sum and a preset formula; and generating information guiding the terminal to move toward the vehicle based on the terminal position, the hotspot center of the hotspot region, and the vehicle position.

[0106] Specifically, the terminal matches the second target area and the hotspot center of the hotspot area to obtain the hotspot center of the second target area; calculates the distance between the second hotspot feature vector and the hotspot center of the second target area to determine the target cluster closest to the second hotspot feature vector; calculates the Euclidean distance between the second hotspot feature vector and the cluster points in the target cluster to obtain the coordinates of the K target cluster points closest to the Euclidean distance; calculates the terminal position based on the coordinates (x, y) of the K target cluster points using a weighted sum and a preset formula; and generates information guiding the terminal to move towards the vehicle based on the terminal position, the hotspot center of the hotspot area, and the vehicle position.

[0107] Optionally, the preset formulas include: the unscented Kalman filter algorithm formula, the Kalman filter observation equation formula, the Kalman filter state equation formula, the proportional sampling formula, the variance formula, and the covariance formula.

[0108] Optionally, pedestrian dead reckoning specifically involves obtaining the orientation angle from the terminal's orientation sensor and determining whether the user's position has changed (i.e., whether the user has taken a step) using the terminal's acceleration sensor information. For ease of calculation, the user's stride length can be set to 50cm, and the average change curve can be calculated. For example, the terminal screen can be placed horizontally on the palm of the hand with the vertical axis coinciding with the Z-axis, the forward axis coinciding with the Y-axis, and the lateral axis coinciding with the X-axis. When the average value rises three times consecutively, it is counted as one peak. The time between the current peak and the previous peak is calculated. If the time exceeds a certain period, such as 0.25 seconds, the user is considered to have taken a step when the current peak is detected.

[0109] Optionally, based on the terminal location, the hotspot center of the hotspot area, and the vehicle location, information guiding the terminal to move towards the vehicle is generated. This includes: calculating Kalman gain, state estimation, and covariance using pedestrian dead reckoning, hotspot localization algorithm, and filtering algorithm based on the terminal location, the hotspot center of the hotspot area, and the vehicle location, to generate information guiding the terminal to move towards the vehicle. The Kalman gain is the accurate ratio of pedestrian dead reckoning and hotspot localization algorithm, the state estimation includes the next position, and the covariance is the error between the calculated coordinates and the actual coordinates.

[0110] Specifically, based on the terminal's location, the hotspot center of the hotspot area, and the vehicle's location, the terminal calculates Kalman gain, state estimation, and covariance using pedestrian dead reckoning, hotspot localization algorithms, and filtering algorithms. This generates information guiding the terminal to move toward the vehicle. The Kalman gain is the accurate ratio of pedestrian dead reckoning and hotspot localization algorithms, the state estimation includes the next position, and the covariance is the error between the calculated coordinates and the actual coordinates.

[0111] This embodiment of the disclosure obtains multiple second hotspots around the terminal; determines the feature vectors of the second hotspots based on the clustering map of the multiple second hotspots and the first hotspot, wherein the clustering map of the first hotspot includes the hotspot center of the hotspot area and multiple second hotspots; and generates information guiding the terminal to move to the vehicle based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle location. This solves the problems of poor parking lot network preventing GPS positioning and navigation, as well as unstable cross-floor hotspot signals and mobile hotspot interference leading to low navigation accuracy, thus improving vehicle search efficiency and accuracy.

[0112] In some embodiments, there are multiple sampling locations between the terminal location and the vehicle location. Figure 6 A flowchart of the vehicle locating method provided in the embodiments of this disclosure is shown below. Figure 6As shown, the method includes the following steps:

[0113] S601. Using the terminal location as the current sampling location, when listening to the user moving towards the next sampling location by pedestrian dead reckoning, determine the coordinates of the current sampling location, the direction angle corresponding to the current sampling location, and the step size, scan multiple second hotspots around the current sampling location, and determine the hotspot center of the second hotspot area.

[0114] The terminal uses its current location as the sampling location. By monitoring the user's movement towards the next sampling location using pedestrian dead reckoning, it determines the coordinates, orientation angle, and step size of the current sampling location. It then scans multiple secondary hotspots around the current sampling location to determine the center of each secondary hotspot area. In other words, the coordinates of the next sampling location are calculated using both pedestrian dead reckoning and hotspot localization algorithms; the next sampling location can also be the next step location.

[0115] S602. The initial mean and initial covariance of the sampling points are sampled by proportional sampling to obtain a preset number of first target sampling points and the weights corresponding to the first target sampling points. The state equation of the Kalman filter is used to perform nonlinear transmission on the current sampling position to calculate the predicted mean and predicted covariance of the current sampling position.

[0116] The terminal samples the initial mean and initial covariance of the sampling points using proportional sampling to obtain a preset number of first target sampling points and their corresponding weights. It then uses the state equation of a Kalman filter to perform nonlinear propagation on the current sampling position, calculating the predicted mean and predicted covariance of the current sampling position. It is understandable that the terminal can also sample the initial mean and initial covariance of the sampling points using proportionally corrected symmetric sampling.

[0117] Specifically, the terminal selects Sigma points using a proportionally corrected symmetric sampling strategy. The procedure for selecting the (2n+1) Sigma point set is as follows: Figure 7 As shown, within the theoretical framework of nonlinear Gaussian filtering, prediction calculations are performed using the UT transform based on the state equation to obtain the prediction mean and prediction covariance.

[0118] Unscented Kalman Filter (UKF) algorithm flow:

[0119] Assume the initial state estimate and the variance of the filter are:

[0120]

[0121] Time update: Assume state estimation at time k. And estimated variance P k|k By using a proportionally corrected symmetric sampling strategy, 2n+1 Sigma sampling points can be obtained. and corresponding weights and

[0122] Then, the sampling points are subjected to a nonlinear state function transfer to obtain:

[0123] One-step state prediction mean and prediction variance:

[0124]

[0125] (1) The initial state estimate x0 is calculated using the hotspot localization algorithm, and the initial variance P is set to

[0126] (2) The sampling strategy is proportional sampling:

[0127] Proportional sampling:

[0128] The mean and variance of the n-dimensional random variable x are respectively and P x Assuming L = 2n, then the total number of Sigma sampling points is 2n+1.

[0129] From the above conditions, we can conclude that:

[0130]

[0131] Corresponding mean weights and variance weights:

[0132]

[0133] Where k represents kappa, an indicator used for consistency testing and also for measuring the effectiveness of classification. The value of kappa directly reflects the higher-order moments (second order and above). The i-th row or i-th column of the matrix, where n is the dimension of the calculated state equation. The state equation is a nonlinear state function, and is as follows:

[0134]

[0135] Among them, W in the state equation k-1 It is the noise in the system state equation. This noise is random and does not participate in the calculation. The prior state estimate and prior variance (predicted mean) are calculated based on the above equation.

[0136] Calculations show that the dimension is 3, which means n is 3.

[0137] Consider a nonlinear system with Gaussian white noise:

[0138]

[0139] Where, x k Let z represent the state vector. k Let w represent the measurement vector, f(.) and h(.) be the system's nonlinear state function and measurement function, respectively. k and v k They have zero mean and covariance Q. k and R k Uncorrelated Gaussian white noise.

[0140] Calculate Q accordingly k This is the predicted covariance of the noise in the state equation. Generally, noise follows a normal distribution. Let Q... k for

[0141] S603. By sampling the predicted mean and predicted covariance of the sampling points through proportional sampling, a preset number of second target sampling points and the corresponding weights of the second target sampling points are obtained. The observation equation of the Kalman filter is used to pass the next sampling position, and the observation mean, observation variance, and observation covariance of the next sampling position are calculated.

[0142] The terminal samples the predicted mean and predicted covariance of the sampling points through proportional sampling to obtain a preset number of second target sampling points and the corresponding weights of the second target sampling points. The observation equation of the Kalman filter is then used to pass the data to the next sampling position to calculate the observation mean, observation variance, and observation covariance of the next sampling position.

[0143] Measurement update: obtained based on time updates and P k+1k The sampling strategy formula yields 2n+1 Sigma sampling points ζ. i ', and the corresponding weights and

[0144] After being passed through the nonlinear measurement function, we obtain:

[0145] The x and p obtained by proportional sampling are propagated through proportional sampling sigma point propagation, and then transferred according to the nonlinear measurement function (i.e., the measurement equation).

[0146] The measurement equation is:

[0147]

[0148] Among them, V k This is system observation noise, which is random and does not participate in the calculation.

[0149] One-step prediction of the mean, variance, and covariance of a measurement variable:

[0150]

[0151] Based on the measured value z at time k+1 k+1 The filter gain K can be calculated. k+1 Mean and variance of state observations at time k+1:

[0152]

[0153] Where R is the covariance of the observation noise, which generally follows a normal distribution, and is denoted as:

[0154]

[0155] Calculate the Kalman gain, observation mean, observation variance, and observation covariance.

[0156] S604. Based on the predicted mean of the current sampling position, the predicted covariance of the current sampling position, the observed mean of the next sampling position, the observed variance of the next sampling position, the observed covariance of the next sampling position, and the hotspot center of the second hotspot region, calculate the Kalman gain, state estimate, and covariance, determine the target coordinates of the next sampling position, and generate information to guide the terminal to move towards the vehicle.

[0157] The terminal calculates Kalman gain, state estimate, and covariance based on the predicted mean, predicted covariance, observed mean, observed variance, observed covariance, and the center of the second hotspot region at the current sampling location. This determines the target coordinates for the next sampling location and generates information guiding the terminal towards the vehicle. Here, covariance represents the error between the calculated coordinates and the actual coordinates, and Kalman gain refers to the bias ratio between the positioning results calculated by the pedestrian dead reckoning and hotspot localization algorithms.

[0158] For example, when the terminal is searching for a vehicle, the terminal's display interface is as follows: Figure 8 As shown, the display includes information such as "Searching for vehicle A," a location map showing the vehicle's position relative to the user, the distance between the vehicle and the user, the vehicle's direction relative to the user, and vehicle horn and light alerts. Users can clearly determine the vehicle's location, the direction the user is holding the terminal, the vehicle's direction, the distance between the user and the vehicle, and the vehicle's horn and light alerts based on the terminal's display, thus quickly locating the vehicle.

[0159] This embodiment calculates the next coordinates using a terminal based on pedestrian dead reckoning and hotspot localization algorithms. A filtering algorithm uses the two dimensions of the next coordinates obtained from pedestrian dead reckoning and hotspot localization algorithms to calculate Kalman gain, state estimation, and covariance, and then comprehensively calculates the next position coordinates. This not only eliminates hotspot signal fluctuations but also eliminates sensor errors in pedestrian dead reckoning, thus improving the accuracy of the parking method.

[0160] Figure 9 This is a schematic diagram of the vehicle-finding map generation device provided in this embodiment. The vehicle-finding map generation device can be a terminal as described in the above embodiment, or it can be a component or assembly within the terminal. The vehicle-finding map generation device provided in this embodiment can execute the processing flow provided in the vehicle-finding map generation method embodiment, such as... Figure 9 As shown, the vehicle location map generation device 90 includes: a first acquisition module 91 and a clustering module 92; wherein, the first acquisition module 91 is used to acquire multiple first hotspots searched by the terminal at multiple sampling locations in the moving route, the terminal moving relative to the vehicle; the clustering module 92 is used to cluster the multiple first hotspots according to the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots, to obtain a clustered map including multiple hotspot areas, wherein different hotspot areas include different hotspots, and the clustered map includes the location information of the vehicle.

[0161] Optionally, the first hotspot includes the name of the first hotspot, the signal strength of the first hotspot, the unique identifier of the first hotspot, and the coordinates of the first hotspot.

[0162] Optionally, the first acquisition module 91 is further configured to acquire multiple original hotspots found by the terminal at multiple sampling locations along the moving route; and filter the multiple original hotspots at each sampling location according to a preset rule to obtain the multiple first hotspots.

[0163] Optionally, the first acquisition module 91 is further configured to, in response to the vehicle parking, acquire the location information of the vehicle, use the location of the vehicle as a first sampling location, the first sampling location belonging to the plurality of sampling locations; acquire the location information of the terminal; and, based on the first sampling location and the location information of the terminal, acquire a plurality of original hotspots searched at the plurality of sampling locations along the movement route of the vehicle and the terminal.

[0164] Optionally, the first acquisition module 91 is further configured to determine the terminal coordinates based on the vehicle's location information by using pedestrian dead reckoning and / or hotspot positioning algorithms, and to determine the floor where the terminal is located by using a barometric pressure sensor.

[0165] Optionally, the pedestrian dead reckoning is calculated based on the current position, direction angle, and step length to determine the next position; the first acquisition module 91 is further configured to acquire the direction angle through a direction sensor; determine whether the user's position has changed through an acceleration sensor; and determine the terminal coordinates based on the change in the user's position, according to the vehicle coordinates, the direction angle, and the step length.

[0166] Optionally, the first acquisition module 91 is further configured to calculate the height difference between the vehicle and the terminal using a barometric pressure sensor, and determine the floor of the terminal based on the height difference and the floor where the vehicle is located.

[0167] Optionally, the preset rules include: the hotspot signal strength at multiple sampling locations is continuously greater than a preset strength threshold, and / or the hotspot name is a preset model name, and / or the number of preset digits of the hotspot's unique identifier is a preset value.

[0168] Optionally, the first acquisition module 91 is further configured to, from the plurality of original hotspots, remove hotspots whose hotspot signal strength at multiple sampling locations is continuously greater than a preset strength threshold, and / or remove hotspots whose hotspot names are preset model names, and / or remove hotspots whose hotspot unique identifiers have a preset number of digits, thereby obtaining the plurality of first hotspots.

[0169] Optionally, the clustering module 92 is further configured to sort multiple sampling locations in chronological order to determine multiple first hotspots at each sampling location; for each sampling location, sort the signal strengths of the multiple first hotspots in descending order, and determine the dimension of the target hotspot feature vector based on the signal strengths of the multiple first hotspots; construct regional feature information based on the target hotspot feature vector to obtain a regional dataset of the first hotspots; and cluster the regional dataset of the first hotspots to obtain a clustered map including multiple hotspot regions.

[0170] Optionally, the clustering module 92 is further configured to traverse the region feature information to obtain the first target region feature information corresponding to the first hotspot at the sampling location; extract the first hotspot feature of the first hotspot at the sampling location through the first target region feature information; based on the value of the first hotspot feature, process the first hotspot at the sampling location into a first hotspot feature vector according to the first target region feature information, and add the first hotspot feature vector to the region dataset corresponding to the first target region.

[0171] Optionally, the clustering module 92 is further configured to, based on the fact that all the first hotspot features are invalid values, sort the signal strength of the first hotspot at the sampling location corresponding to the first hotspot feature in descending order, determine the dimension of the target hotspot feature vector to be extracted according to the number of first hotspots at the sampling location, and update the regional feature information constructed by the target hotspot feature vector to the regional dataset of the first hotspot.

[0172] Optionally, the clustering module 92 is further configured to determine the cluster points of the regional dataset of the first hotspot, calculate the K value, where the K value represents the number of hotspot centers in the hotspot region; select the center point of the cluster, and cluster the clusters using the K nearest neighbor classification algorithm to obtain a cluster map including multiple hotspot regions.

[0173] Optionally, the clustering module 92 is further configured to select the center point of the cluster, calculate the signal domain distance and point distance between the center point of the cluster and the cluster points; calculate the Euclidean norm distance between the center point of the cluster and the cluster points based on the signal domain distance and point distance; select the cluster point with the smallest Euclidean norm distance to add to the cluster, and update the cluster; calculate the average value of the cluster as the center point of the cluster, and repeatedly calculate the Euclidean norm distance between the center point of the cluster and the cluster points until the sum of the Euclidean norm distances from the cluster points to the center points of the corresponding clusters is less than a preset threshold, thereby obtaining a clustering map of the first hotspot.

[0174] Figure 9 The vehicle-finding map generation device shown in the embodiment can be used to execute the technical solution of the above-described vehicle-finding map generation method embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0175] Figure 10 This is a schematic diagram of the vehicle-finding device provided in an embodiment of this disclosure. The vehicle-finding device can be a terminal as described in the above embodiments, or it can be a component or assembly within the terminal. The vehicle-finding device provided in this embodiment can execute the processing flow provided in the vehicle-finding method embodiments, such as... Figure 10 As shown, the vehicle-finding device 100 includes: a second acquisition module 101, used to acquire multiple second hotspots around the terminal; a determination module 102, used to determine the feature vectors of the second hotspots based on the clustering map of the multiple second hotspots and the first hotspot, wherein the clustering map of the first hotspot includes the hotspot center of the hotspot area and the multiple second hotspots; and a generation module 103, used to generate information guiding the terminal to move toward the vehicle based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle location.

[0176] Optionally, the determining module 102 is further configured to match the regional datasets of the plurality of second hotspots and the first hotspot, determine the second target region feature information of the terminal, and construct the second hotspot feature vector.

[0177] Optionally, the determining module 102 is further configured to extract multiple second hotspot features of the multiple second hotspots; based on the values ​​of the multiple second hotspot features, determine the second target region corresponding to the multiple second hotspots according to the first hotspot region, determine the second target region feature information of the terminal; and convert the second hotspot feature vector according to the second target region feature information of the terminal.

[0178] Optionally, the determining module 102 is also used to control the terminal interface to display an abnormality based on the fact that all of the multiple second hotspot features are illegal values, so as to prompt the user to deviate from the planned trajectory.

[0179] Optionally, the generation module 103 is further configured to: obtain the hotspot center of the second target region based on the second target region and the hotspot center of the hotspot region; calculate the distance between the second hotspot feature vector and the hotspot center of the second target region to determine the target cluster closest to the second hotspot feature vector; calculate the Euclidean distance between the second hotspot feature vector and the cluster points in the target cluster to obtain the coordinates of the K target cluster points closest to the Euclidean distance; calculate the terminal position based on the coordinates of the K target cluster points using a weighted sum and preset formula; and generate information guiding the terminal to move towards the vehicle based on the terminal position, the hotspot center of the hotspot region, and the vehicle position.

[0180] Optionally, the preset formulas include: the unscented Kalman filter algorithm formula, the Kalman filter observation equation formula, the Kalman filter state equation formula, the proportional sampling formula, the variance formula, and the covariance formula.

[0181] Optionally, the generation module 103 is further configured to calculate Kalman gain, state estimation, and covariance based on the terminal location, the hotspot center of the hotspot area, and the vehicle location, using pedestrian dead reckoning, hotspot positioning algorithm, and filtering algorithm, and generate information guiding the terminal to move toward the vehicle. The Kalman gain is the accurate ratio of pedestrian dead reckoning and hotspot positioning algorithm, the state estimation includes the next position, and the covariance is the error between the calculated coordinates and the actual coordinates.

[0182] Optionally, multiple sampling locations exist between the terminal location and the vehicle location; the generation module 103 is further configured to use the terminal location as the current sampling location, and when listening to the user moving towards the next sampling location through pedestrian dead reckoning, determine the coordinates of the current sampling location, the direction angle corresponding to the current sampling location, and the step size; scan multiple second hotspots around the current sampling location to determine the center of the second hotspot area; sample the initial mean and initial covariance of the sampling points through proportional sampling to obtain a preset number of first target sampling points and the weights corresponding to the first target sampling points; perform nonlinear propagation on the current sampling location through the state equation of Kalman filtering to calculate the predicted mean and predicted covariance of the current sampling location; and perform proportional sampling on the initial mean and initial covariance of the sampling points. The predicted mean and predicted covariance of the sampling points are sampled to obtain a preset number of second target sampling points and their corresponding weights. The observation equation of the Kalman filter is used to pass the observation to the next sampling position, and the observation mean, observation variance, and observation covariance of the next sampling position are calculated. Based on the predicted mean, prediction covariance, observation mean, observation variance, and observation covariance of the current sampling position, and the hotspot center of the second hotspot region, the Kalman gain, state estimate, and covariance are calculated to determine the target coordinates of the next sampling position and generate information to guide the terminal to move towards the vehicle.

[0183] Figure 10 The vehicle-finding device shown in the embodiment can be used to execute the technical solution of the above-described vehicle-finding method embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0184] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. The electronic device can be a terminal as described in the above embodiments. The electronic device provided in this disclosure can execute a vehicle-finding map generation method or a processing flow provided in an embodiment of a vehicle-finding method, such as… Figure 11 As shown, the electronic device 110 includes: a memory 111, a processor 112, a computer program, and a communication interface 113; wherein, the computer program is stored in the memory 111 and is configured to be executed by the processor 112 to perform the vehicle-finding map generation method or vehicle-finding method as described above.

[0185] In addition, this disclosure also provides a non-volatile computer-readable storage medium storing a computer program thereon, which is executed by a processor to implement the vehicle-finding map generation method or vehicle-finding method described in the above embodiments.

[0186] Furthermore, this disclosure also provides a computer program product, which includes a computer program or instructions that, when executed by a processor, implement the vehicle-finding map generation method or vehicle-finding method as described above.

[0187] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0188] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0189] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0190] The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to:

[0191] The terminal acquires multiple first hotspots at multiple sampling locations along the moving route, the terminal being moved relative to the vehicle;

[0192] Based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots, the multiple first hotspots are clustered to obtain a cluster map including multiple hotspot regions, wherein different hotspot regions include different hotspots, and the cluster map includes the location information of the vehicle.

[0193] In addition, the electronic device can also perform other steps in the vehicle-finding map generation method described above.

[0194] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0195] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0196] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0197] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0198] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0199] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0200] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a vehicle-finding map, characterized in that, The method is applied to a terminal and includes: The terminal acquires multiple first hotspots at multiple sampling locations along the moving route, the terminal being moved relative to the vehicle; Based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hot spots, the multiple first hot spots are clustered to obtain a cluster map including multiple hot spot regions, wherein different hot spot regions include different hot spots, and the cluster map includes the location information of the vehicle; The step of clustering the multiple first hotspots based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots to obtain a clustered map including multiple hotspot regions includes: Multiple sampling locations are sorted in chronological order to determine multiple primary hot spots at each sampling location; For each sampling location, the signal intensities of multiple first hotspots at the sampling location are sorted in descending order, and the feature vector dimension of the target hotspot is determined based on the signal intensities of the multiple first hotspots. Based on the target hotspot feature vector, regional feature information is constructed to obtain the regional dataset of the first hotspot; Cluster the regional dataset of the first hotspot to obtain a clustered map including multiple hotspot regions; The method further includes: Traverse the region feature information to obtain the first target region feature information corresponding to the first hotspot at the sampling location; The first hotspot feature of the first hotspot at the sampling location is extracted using the feature information of the first target region; Based on the value of the first hotspot feature, the first hotspot at the sampling location is processed into a first hotspot feature vector according to the feature information of the first target region, and the first hotspot feature vector is added to the regional dataset corresponding to the first target region. The method further includes: Since all the first hotspot features are invalid values, the signal strength of the first hotspot at the sampling location corresponding to the first hotspot feature is sorted in descending order, and the dimension of the target hotspot feature vector is determined according to the number of first hotspots at the sampling location. The regional feature information constructed from the target hotspot feature vector is then updated in the regional dataset of the first hotspot.

2. The method according to claim 1, characterized in that, The first hotspot includes the name of the first hotspot, the signal strength of the first hotspot, the unique identifier of the first hotspot, and the coordinates of the first hotspot.

3. The method according to claim 1, characterized in that, The system acquires multiple primary hotspots detected by the terminal at multiple sampling locations along its movement route, including: Acquire multiple raw hotspots found by the terminal at multiple sampling locations along the mobile route; The plurality of original hotspots at each sampling location are filtered according to preset rules to obtain the plurality of first hotspots.

4. The method according to claim 3, characterized in that, Obtain multiple raw hotspots found by the terminal at multiple sampling locations along the mobile route, including: In response to the vehicle parking, the vehicle's location information is obtained, and the vehicle's location is used as a first sampling location, which belongs to the plurality of sampling locations; Obtain the location information of the terminal; Based on the location information of the first sampling location and the terminal, multiple original hotspots are obtained from multiple sampling locations along the movement route of the vehicle and the terminal.

5. The method according to claim 4, characterized in that, Obtaining the location information of the terminal includes: Based on the vehicle's location information, the terminal coordinates are determined by pedestrian dead reckoning and / or hotspot positioning algorithms, and the floor where the terminal is located is determined by a barometric pressure sensor.

6. The method according to claim 5, characterized in that, The pedestrian dead reckoning calculation is based on the current position, orientation angle, and stride length to calculate the next position; Among them, determining the terminal coordinates based on the vehicle's location information using pedestrian dead reckoning and / or hotspot positioning algorithms includes: The orientation angle is obtained using an orientation sensor; The accelerometer determines whether the user's position has changed. Based on the change in the user's location, the terminal coordinates are determined according to the vehicle coordinates, the direction angle, and the step size.

7. The method according to claim 5, characterized in that, The floor where the terminal is located is determined by a barometric pressure sensor, including: The height difference between the vehicle and the terminal is calculated using a barometric pressure sensor, and the floor where the terminal is located is determined based on the height difference and the floor where the vehicle is located.

8. The method according to claim 3, characterized in that, The preset rules include: the hotspot signal strength at multiple sampling locations is continuously greater than a preset strength threshold, and / or the hotspot name is a preset model name, and / or the preset number of digits of the hotspot's unique identifier is a preset value.

9. The method according to claim 8, characterized in that, The plurality of original hotspots at each sampling location are filtered according to preset rules to obtain the plurality of first hotspots, including: Among the multiple original hotspots, hotspots whose signal strength at multiple sampling locations is continuously greater than a preset strength threshold, and / or hotspots whose names are preset model names, and / or hotspots whose unique identifiers have a preset number of digits, are removed, thus obtaining the multiple first hotspots.

10. The method according to claim 1, characterized in that, Clustering the region dataset of the first hotspot yields a clustered map containing multiple hotspot regions, including: Determine the cluster points of the regional dataset of the first hotspot, and calculate the K value, where the K value represents the number of hotspot centers in the hotspot region; The center point of the cluster is selected, and the cluster is clustered using the K-nearest neighbor classification algorithm to obtain a cluster map including multiple hotspot areas.

11. The method according to claim 10, characterized in that, Select the center point of each cluster, and cluster the clusters using the K-nearest neighbor classification algorithm to obtain the cluster map of the first hotspot, including: Select the center point of the cluster, and calculate the signal domain distance and point distance between the center point of the cluster and the cluster points; Calculate the Euclidean norm distance between the center point of the cluster and the cluster point based on the signal domain distance and the point distance; Select the cluster point with the smallest Euclidean norm distance and add it to the cluster, then update the cluster. The average value of the cluster is calculated as the center point of the cluster. The Euclidean norm distance between the center point of the cluster and the cluster point is repeatedly calculated until the sum of the Euclidean norm distances from the cluster point to the center point of the corresponding cluster is less than a preset threshold, thus obtaining the cluster map of the first hotspot.

12. A vehicle location method, characterized in that, The method includes: Acquire multiple secondary hotspots around the terminal; Based on the clustering maps of the multiple second hotspots and the first hotspot, the feature vectors of the second hotspots are determined. The clustering map of the first hotspot includes the hotspot center of the hotspot region and the multiple second hotspots. Based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle location, information guiding the terminal to move toward the vehicle is generated. The step of determining the feature vector of the second hotspot based on the clustering map of the plurality of second hotspots and the first hotspot includes: By matching the regional datasets of the multiple second hotspots and the first hotspot, the feature information of the second target region of the terminal is determined, and a feature vector of the second hotspot is constructed. The process of matching the regional datasets of the multiple second hotspots and the first hotspots to determine the second target region feature information of the terminal and constructing a second hotspot feature vector includes: Extract multiple second hotspot features from the multiple second hotspots; Based on the values ​​of the multiple second hotspot features, the second target region corresponding to the multiple second hotspots is determined according to the first hotspot region, and the second target region feature information of the terminal is determined. The second hotspot feature vector is converted based on the second target region feature information of the terminal; The method further includes: Based on the fact that all of the multiple second hotspot features are illegal values, the terminal interface is controlled to display abnormally to prompt the user to deviate from the planned trajectory; The step of generating information guiding the terminal to move toward the vehicle based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle location includes: Based on the second target area and the hotspot center of the hotspot area, obtain the hotspot center of the second target area; Calculate the distance between the second hotspot feature vector and the hotspot center of the second target region, and determine the target cluster that is closest to the second hotspot feature vector; Calculate the Euclidean distance between the second hotspot feature vector and the cluster points in the target cluster, and obtain the coordinates of the K target cluster points with the closest Euclidean distance; The terminal position is calculated based on the coordinates of the K target cluster points using a weighted sum and a preset formula. Based on the terminal location, the hotspot center of the hotspot area, and the vehicle location, information is generated to guide the terminal to move towards the vehicle.

13. The method according to claim 12, characterized in that, The preset formulas include: the formula for the unscented Kalman filter algorithm, the formula for the observation equation of the Kalman filter, the formula for the state equation of the Kalman filter, the formula for proportional sampling, the formula for variance, and the formula for covariance.

14. The method according to claim 12, characterized in that, Based on the terminal location, the hotspot center of the hotspot area, and the vehicle location, information guiding the terminal to move towards the vehicle is generated, including: Based on the terminal location, the hotspot center of the hotspot area, and the vehicle location, Kalman gain, state estimation, and covariance are calculated using pedestrian dead reckoning, hotspot localization algorithm, and filtering algorithm to generate information guiding the terminal to move toward the vehicle. The Kalman gain is the accurate ratio of pedestrian dead reckoning and hotspot localization algorithm, the state estimation includes the next position, and the covariance is the error between the calculated coordinates and the actual coordinates.

15. The method according to claim 14, characterized in that, There are multiple sampling locations between the terminal location and the vehicle location; Specifically, based on the terminal location, the hotspot center of the hotspot area, and the vehicle location, Kalman gain, state estimation, and covariance are calculated using pedestrian dead reckoning, hotspot localization algorithms, and filtering algorithms to generate information guiding the terminal to move towards the vehicle, including: Using the terminal location as the current sampling location, when the user moves to the next sampling location by listening to the pedestrian dead reckoning, the coordinates of the current sampling location, the direction angle corresponding to the current sampling location, and the step size are determined. Multiple second hotspots around the current sampling location are scanned to determine the hotspot center of the second hotspot area. The initial mean and initial covariance of the sampling points are sampled by proportional sampling to obtain a preset number of first target sampling points and the weights corresponding to the first target sampling points. The state equation of the Kalman filter is used to perform nonlinear transmission on the current sampling position to calculate the predicted mean and predicted covariance of the current sampling position. The predicted mean and predicted covariance of the sampling points are sampled by proportional sampling to obtain a preset number of second target sampling points and the corresponding weights of the second target sampling points. The observation equation of the Kalman filter is then used to pass the observation to the next sampling position to calculate the observation mean, observation variance, and observation covariance of the next sampling position. Based on the predicted mean of the current sampling location, the predicted covariance of the current sampling location, the observed mean of the next sampling location, the observed variance of the next sampling location, the observed covariance of the next sampling location, and the hotspot center of the second hotspot region, the Kalman gain, state estimate, and covariance are calculated to determine the target coordinates of the next sampling location and generate information to guide the terminal to move toward the vehicle.

16. A vehicle location map generation device, characterized in that, The device is applied to a terminal and includes: The first acquisition module is used to acquire multiple first hotspots searched by the terminal at multiple sampling locations in the moving route, wherein the terminal is moving relative to the vehicle; The clustering module is used to cluster the multiple first hotspots based on the coordinate information of the multiple sampling locations and the intensity information of the multiple first hotspots to obtain a clustered map including multiple hotspot regions, wherein different hotspot regions include different hotspots, and the clustered map includes the location information of the vehicle. The clustering module is further configured to sort multiple sampling locations in chronological order to determine multiple first hotspots at each sampling location; for each sampling location, sort the signal strengths of the multiple first hotspots in descending order; determine the dimension of the target hotspot feature vector based on the signal strengths of the multiple first hotspots; construct regional feature information based on the target hotspot feature vector to obtain a regional dataset of the first hotspots; and cluster the regional dataset of the first hotspots to obtain a clustered map including multiple hotspot regions. The clustering module is further configured to traverse the region feature information to obtain the first target region feature information corresponding to the first hotspot at the sampling location; extract the first hotspot feature of the first hotspot at the sampling location through the first target region feature information; based on the value of the first hotspot feature, process the first hotspot at the sampling location into a first hotspot feature vector according to the first target region feature information, and add the first hotspot feature vector to the region dataset corresponding to the first target region. The clustering module is further configured to, based on the fact that all the first hotspot features are invalid values, sort the signal strength of the first hotspots at the sampling locations corresponding to the first hotspot features in descending order, determine the dimension of the target hotspot feature vector based on the number of first hotspots at the sampling locations, and update the regional feature information constructed from the target hotspot feature vector to the regional dataset of the first hotspots.

17. A vehicle finding device, characterized in that, The device includes: The second acquisition module is used to acquire multiple secondary hotspots around the terminal; The determination module is used to determine the feature vector of the second hotspot based on the clustering map of the plurality of second hotspots and the first hotspot, wherein the clustering map of the first hotspot includes the hotspot center of the hotspot region and the plurality of second hotspots; The generation module is used to generate information guiding the terminal to move toward the vehicle based on the feature vectors of the multiple second hotspots, the hotspot center of the hotspot area, and the vehicle position. The determining module is also used to match the regional datasets of the multiple second hotspots and the first hotspot, determine the second target region feature information of the terminal, and construct the second hotspot feature vector; The determining module is further configured to extract multiple second hotspot features from the multiple second hotspots; based on the values ​​of the multiple second hotspot features, determine the second target region corresponding to the multiple second hotspots according to the first hotspot region, determine the second target region feature information of the terminal; and convert the second hotspot feature vector according to the second target region feature information of the terminal. The determining module is also used to control the terminal interface to display an abnormality based on the fact that all of the multiple second hotspot features are illegal values, so as to prompt the user to deviate from the planned trajectory; The generation module is further configured to: obtain the hotspot center of the second target area based on the second target area and the hotspot center of the hotspot area; calculate the distance between the second hotspot feature vector and the hotspot center of the second target area to determine the target cluster closest to the second hotspot feature vector; calculate the Euclidean distance between the second hotspot feature vector and the cluster points in the target cluster to obtain the coordinates of the K target cluster points closest to the Euclidean distance; calculate the terminal position based on the coordinates of the K target cluster points using a weighted sum and preset formula; and generate information guiding the terminal to move towards the vehicle based on the terminal position, the hotspot center of the hotspot area, and the vehicle position.

18. An electronic device, characterized in that, include: Memory; processor; as well as Computer programs; The computer program is stored in the memory and configured to be executed by the processor to implement the method as described in any one of claims 1-15.

19. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-15.

Images