Automobile wireless charging parking position judgment method and system

Abstract

The invention relates to the technical field of automobile wireless charging, in particular to an automobile wireless charging parking position judgment method. The method comprises the following steps: S1, establishing initial connection between the vehicle-mounted equipment and a wireless access point in a certain range, judging whether a machine learning model is prepared or not by a main control access point, and if the machine learning model is not prepared, executing the step S2; if the preparation is finished, executing the step S3; s2, identifying a parking position and a wireless access point corresponding to the parking position according to the power pulse, and collecting position fingerprint information by the main control access point through the wireless access point; and S3, measuring signal intensity values of the wireless access points, and identifying the parking positions and the corresponding wireless access points by the machine learning model through comparing the signal intensity values of the wireless access points with the position fingerprint information. In the training stage, training data can be automatically collected, a fingerprint information base is established, an existing mode of manually collecting fingerprint information is replaced, and the system operation cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile wireless charging, and in particular to a method for determining a parking position of an automobile wireless charging. Background Art

[0002] The automobile wireless charging system usually includes a ground end and an on-board end. The ground end includes a power transmitter and a WiFi access point network device (Access Point, AP), and the on-board end includes a power receiver and a WiFi terminal (STA). The STA positioning method used in the prior art is a WiFi fingerprint positioning method, which identifies the parking space of the car through the signal feature fingerprint of the AP on each parking space and connects to the AP corresponding to the parking space. However, this method requires collecting the signal feature fingerprint of each known parking space during the training phase in order to establish an accurate fingerprint database for comparison and identification. The prior art usually uses manual collection to obtain signal feature fingerprints, which greatly increases the operating cost of the wireless charging system. Summary of the invention

[0003] The purpose of the present invention is to provide a method for determining the parking position of a wireless charging vehicle to solve the above technical problems;

[0004] The present invention also aims to provide a system for determining the parking position of a wirelessly charged vehicle to solve the above technical problems.

[0005] The technical problem solved by the present invention can be achieved by adopting the following technical solutions:

[0006] A method for determining a parking position of a car using wireless charging, comprising:

[0007] Step S1, after the car is parked in the parking space, the vehicle-mounted device establishes an initial connection with a wireless access point within a certain range, and a master access point determines whether the machine learning model is ready. If not, step S2 is executed; if ready, step S3 is executed;

[0008] Step S2, the vehicle-mounted terminal receives the power pulse from the ground terminal, identifies the parking position and the wireless access point corresponding to the parking position according to the power pulse, the vehicle-mounted device connects to the corresponding wireless access point, and the master access point collects the position fingerprint information through the wireless access point;

[0009] Step S3, the vehicle-mounted end measures the signal strength value of the wireless access point within a certain range, the machine learning model identifies the parking position and the corresponding wireless access point by comparing the signal strength value of the wireless access point with the location fingerprint information, and the vehicle-mounted device connects to the corresponding wireless access point.

[0010] Preferably, in step S1, the vehicle-mounted device selects one of the wireless access points according to the signal strength value of the wireless access point to establish an initial connection.

[0011] Preferably, in step S1, the method for determining whether the machine learning model is ready includes:

[0012] The master access point determines whether the machine learning model has completed collecting the location fingerprint information or whether the location fingerprint information needs to be updated. If the location fingerprint information is collected and does not need to be updated, it is determined that the machine learning model is ready, and the master access point sends a preparation completion notification to the vehicle-mounted device, and executes step S3; otherwise, it is determined that the machine learning model is not ready, and the master access point sends a non-readiness completion notification to the vehicle-mounted device, and executes step S2.

[0013] Preferably, in step S1, the wireless access point includes the master access point.

[0014] Preferably, step S2 comprises,

[0015] Step S21, multiple ground-side power transmission devices generate power pulses respectively;

[0016] Step S22, the power receiving device at the vehicle end receives the power pulse, and identifies the parking position where the corresponding power transmitting device is located and the corresponding wireless access point according to the waveform of the received power pulse;

[0017] Step S23, determining whether the wireless access point currently connected to the vehicle-mounted device is the corresponding wireless access point, if so, the vehicle-mounted device maintains connection with the current wireless access point; if not, the vehicle-mounted device switches to connect with the corresponding wireless access point.

[0018] Preferably, in step S2, each ground end is a parking space, and each parking space is provided with a matching power sending device and a matching wireless access point; the vehicle end is provided with a matching power receiving device and a matching vehicle-mounted device.

[0019] Preferably, in step S2, the propagation distance of the power pulse is smaller than the distance between the power receiving device and the power sending device outside the parking position.

[0020] Preferably, step S2 further includes, step S24, the vehicle-mounted device measures the signal strength value RSSI of all the wireless access points, and the master access point collects the location fingerprint information

[0021] in represents the average value of the signal strength value RSSI from the nth wireless access point collected by the vehicle-mounted device at the lth parking space; l, n are positive integers belonging to [1, N], and N represents the number of the wireless access points.

[0022] Preferably, step S3 comprises,

[0023] Step S31: The vehicle-mounted terminal measures the signal strength value SRRI of the wireless access point within a certain range. m ;

[0024] Step S32, the master access point sends the location fingerprint information to the vehicle-mounted device in an inquiry manner or a mandatory manner;

[0025] Step S33, the machine learning model identifies the corresponding wireless access point by comparing the signal strength value of the wireless access point with the location fingerprint information;

[0026] Step S34, determining whether the wireless access point currently connected to the vehicle-mounted device is the corresponding wireless access point, if so, the vehicle-mounted device maintains the connection with the current wireless access point; if not, the vehicle-mounted device switches to connect with the corresponding wireless access point;

[0027] RSSI m represents the signal strength value of the mth wireless access point around the vehicle-mounted device, where m is a positive integer belonging to [1, N], and N represents the number of the wireless access points.

[0028] Automobile wireless charging parking position determination system, including:

[0029] Wireless access points are respectively arranged at the ground end, and the wireless access points include a master access point connected to the other wireless access points, and the master access point is used to determine whether the machine learning model is ready and collect location fingerprint information;

[0030] power transmission devices, respectively corresponding to the wireless access points, generating power pulses for identifying parking positions;

[0031] The vehicle-mounted device is provided at the vehicle end and is used to identify the parking position according to the power pulse, select the wireless access point corresponding to the parking position for connection and measure the signal strength value of the wireless access point;

[0032] a power receiving device, corresponding to the vehicle-mounted device, for receiving the power pulse;

[0033] The machine learning model is connected to the vehicle-mounted device and the master access point, and is used to compare the signal strength value of the wireless access point with the location fingerprint information to identify the parking location and the corresponding wireless access point.

[0034] Beneficial effects of the present invention: Due to the adoption of the above technical solution, the present invention can automatically collect training data and establish a fingerprint information database during the training stage, replacing the existing method of manually collecting fingerprint information and reducing system operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of the steps of a method for determining a parking position of a wireless charging vehicle according to an embodiment of the present invention;

[0036] Figure 2 This is a structural diagram of a system for determining a parking position of a wireless charging vehicle in an embodiment of the present invention;

[0037] Figure 3 This is a flow chart of vehicle-mounted terminal signal processing in an embodiment of the present invention;

[0038] Figure 4 This is a flow chart of ground terminal signal processing in an embodiment of the present invention;

[0039] Figure 5 This is a schematic diagram of step S2 in an embodiment of the present invention;

[0040] Figure 6 Schematic diagram of step S3 in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0043] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0044] Method for determining the parking position of a car with wireless charging, such as Figure 1 As shown, including,

[0045] Step S1, after the car 5 is parked in the parking space, the vehicle-mounted device 3 establishes an initial connection with the wireless access point 1 within a certain range, and a master access point 11 determines whether the machine learning model is ready. If not, step S2 is executed; if ready, step S3 is executed;

[0046] Step S2, the vehicle-mounted terminal receives the power pulse from the ground terminal, identifies the parking position and the wireless access point corresponding to the parking position according to the power pulse, the vehicle-mounted device 3 connects to the corresponding wireless access point, and the master access point 11 collects the position fingerprint information through the wireless access point 1;

[0047] In step S3, the vehicle-mounted end measures the signal strength value of the wireless access point 1 within a certain range. The machine learning model identifies the parking location and the corresponding wireless access point by comparing the signal strength value of the wireless access point 1 with the location fingerprint information, and the vehicle-mounted device 3 connects to the corresponding wireless access point.

[0048] Specifically, the method for determining the parking position of a wireless charging vehicle of the present invention is applied in a wireless charging system for a vehicle, such as Figure 2 As shown, it includes a ground end and a vehicle-mounted end. The ground end includes a power transmitting device and a wireless access point 1 (ie, AccessPoint, AP), and the vehicle-mounted end includes a power receiving device 4 and a vehicle-mounted device 3 (ie, STA). Each charging parking space 6 is installed with a set of ground-end devices, namely, a power transmitting device and a wireless access point 1. The system used in the present invention adopts an independent WiFi local area network and cannot exchange information with the Internet / Internet of Vehicles, as well as the in-vehicle communication system. Therefore, it is conducive to the independent development and operation of wireless charging products. Through the method and system for determining the parking position of a wireless charging vehicle provided by the present invention, training data can be automatically collected and a fingerprint information database can be established to replace the existing method of collecting fingerprint information manually, thereby reducing the system operating costs.

[0049] The function realized by the present invention is that when the car 5 is parked in a charging parking space 6, the vehicle-mounted device 3 of the vehicle-mounted end will try to connect to the wireless access point corresponding to the ground end of the parking space. After the connection is established, the ground end of the parking space starts power transmission, and the WiFi communication satisfies the information exchange between the vehicle-mounted end and the ground end during the charging process. If the wireless access point 1 is connected incorrectly, for example, it is connected to the wireless access point 1 of another parking space 6, the car 5 will not be able to charge.

[0050] Based on the functions of the above application scenarios, the present invention proposes a method for determining the parking position of a car wireless charging. The determination of the parking space 6 and the selection of the wireless access point 1 are essentially the same, that is, to let the wireless access point 1 find its own position (positioning), that is, which parking space 6 the car 5 is parked in. Because each parking space 6 corresponds to a wireless access point 1, finding the correct parking space 6 is equivalent to selecting the correct wireless access point 1.

[0051] The present invention is divided into two stages: training and inference. In the training stage of step S2, it is necessary to collect location fingerprint information at each known location. The location fingerprint information of the present invention is the signal strength from the wireless access point 1 received by the vehicle-mounted device 3. The present invention uses the MAC address of the access point as the identification code of the parking space 6. In the training stage, it is necessary to collect the location fingerprint information of each wireless access point 1 received by the vehicle-mounted device 3 in each parking space, so as to establish a fingerprint database. In the inference stage of step S3, the vehicle-mounted device 3 determines which parking space it belongs to by the signal strength of multiple wireless access points 1 received and accesses the corresponding wireless access point.

[0052] When the car 5 is parked in a parking space 6 , the car 5 needs to connect to the correct wireless access point 1 and start charging.

[0053] The information processing method of the vehicle-mounted device 3 is as follows: Figure 3 As shown, the following steps are included:

[0054] Step 1, after the car 5 parks in the parking space, the vehicle-mounted device 3 receives the parking instruction, searches for the nearby wireless access point 1 and establishes an initial connection;

[0055] Step 2, receiving information from the master access point 11 on whether the machine learning model is ready, if it is ready, entering the training phase and executing step 3; if it is not ready, executing step 4;

[0056] Step 3, detecting the power pulse and determining the parking position, switching the connected wireless access point 1 according to the parking position or sending information that no switching is required, the vehicle-mounted device 3 measures the signal strength values ​​of all wireless access points 1, and sends the obtained measurement data to the master access point 11;

[0057] Step 4, receiving the inquiry from wireless access point 1, replying to wireless access point 1 whether to receive the location fingerprint information, if the reply is yes, receiving the location fingerprint information, and executing step 5; if the reply is no, directly executing step 5; or receiving the location fingerprint information forcibly sent by wireless access point 1, and executing step 5;

[0058] Step 5: Infer and switch the connected wireless access point 1, or send information that no switching is required.

[0059] The information processing method of the wireless access point 1 is as follows: Figure 4 As shown, the following steps are included:

[0060] Step 1, wireless access point 1 performs access detection and notifies the master access point 11;

[0061] Step 2, the master access point 11 determines whether the machine learning model is ready and notifies the vehicle-mounted device 3. If not, it enters the training phase and executes step 3; if it is ready, it enters the inference phase and executes step 4;

[0062] Step 3, sending power pulses at multiple ground terminals, waiting for the vehicle-mounted device 3 to switch the wireless access point 1 connection or the information that no switching is required, and the master access point 11 receives the measurement data from the vehicle-mounted device 3 to train the machine learning model;

[0063] Step 4, the master access point 11 inquires whether to send the location fingerprint information of the machine learning model. If the reply received from the vehicle-mounted device 3 is yes, the location fingerprint information is sent; if the reply received from the vehicle-mounted device 3 is no, there is no need to send the location fingerprint information; or the master access point 11 forcibly sends the location fingerprint information to the vehicle-mounted device 3;

[0064] Step 5: The wireless access point 1 waits for the information of the vehicle-mounted device 3 that the connection is switched or that no switching of the connection is required.

[0065] In a preferred embodiment, in step S1 , the vehicle-mounted device 3 selects a wireless access point 1 according to the signal strength value of the wireless access point 1 to establish an initial connection.

[0066] Specifically, the connection between the vehicle-mounted device 3 and the wireless access point 1 is divided into an initial connection and a connection switching. The wireless access point 1 for the initial connection of the present invention is selected by signal strength. Therefore, the wireless access point 1 for establishing the initial connection with the vehicle-mounted device 3 is not necessarily the wireless access point 1 that correctly corresponds to the vehicle-mounted device 3. The purpose of establishing the initial connection is to ensure the communication between the master access point 11 and the vehicle-mounted device 3. When the communication process for identifying the corresponding wireless access point is completed, the vehicle-mounted device 3 can select the corresponding wireless access point to switch the connection by means of power pulses or machine learning; if the identified wireless access point corresponding to the vehicle-mounted device 3 is the wireless access point 1 for establishing the initial connection with the vehicle-mounted device 3, the vehicle-mounted device 3 sends a signal that no switching is required and maintains the connection with the current wireless access point 1.

[0067] In a preferred embodiment, in step S1, the method for determining whether the machine learning model is ready includes:

[0068] The master access point 11 determines whether the machine learning model has completed collecting location fingerprint information or whether the location fingerprint information needs to be updated. If the location fingerprint information is collected and does not need to be updated, it is determined that the machine learning model is ready, and the master access point 11 sends a preparation completion notification to the vehicle-mounted device, and executes step S3; otherwise, it is determined that the machine learning model is not ready, and the master access point 11 sends an unreadiness completion notification to the vehicle-mounted device, and executes step S2.

[0069] Specifically, the master access point 11 determines whether the machine learning model is ready, for example, whether the location fingerprint information collection is completed, whether the location fingerprint collection needs to be updated, etc., so as to determine whether it can enter the inference stage. If the machine learning model has not completed the collection of location fingerprint information, it enters the training stage, identifies the parking position through power pulses, and collects location fingerprint information.

[0070] In a preferred embodiment, in step S1 , the wireless access point 1 includes a master access point 11 .

[0071] In a preferred embodiment, Figure 5 As shown, step S2 includes,

[0072] Step S21, the power transmission devices 2 at the multiple ground terminals generate power pulses respectively;

[0073] Step S22, the power receiving device 4 at the vehicle end receives the power pulse, and identifies the parking position where the corresponding power transmitting device 2 is located and the corresponding wireless access point according to the waveform of the received power pulse;

[0074] Step S23, determining whether the wireless access point 1 currently connected to the vehicle-mounted device 3 is the corresponding wireless access point, if so, the vehicle-mounted device 3 maintains the connection with the current wireless access point 1; if not, the vehicle-mounted device 3 switches to connect with the corresponding wireless access point.

[0075] Specifically, in the training phase of step S2 of the present invention, the master access point 11 collects location fingerprint information through the wireless access point 1 of each parking space 6. represents the average value of RSSI from the nth AP collected by STA at the lth parking space 6, where It represents the average value of the signal strength value RSSI from the nth wireless access point 1 collected by the vehicle-mounted device 3 at the lth parking space 6; l, n are positive integers belonging to [1, N], and N represents the number of wireless access points 1 (that is, the number of parking spaces 6).

[0076] Since the collection of location fingerprint information requires the machine learning model to know the real location of the vehicle-mounted device 3 in advance, in the present invention, the power pulse of the power transmitting device 2 is used to allow the vehicle-mounted terminal to identify the parking space. The power transmitting device 2 of each parking space generates a different pulse waveform. When the vehicle-mounted terminal receives the power pulse and performs waveform detection, it can identify the wireless access point 1 supporting the power transmitting device 2, thereby identifying the parking location, and then the vehicle-mounted device 3 connects to the corresponding wireless access point.

[0077] In a preferred embodiment, in step S2, each ground end is a parking space 6, and each parking space 6 is provided with a matching power sending device 2 and a wireless access point 1; the vehicle end is provided with a matching power receiving device 4 and a vehicle-mounted device 3.

[0078] In a preferred embodiment, in step S2, the propagation distance of the power pulse is smaller than the distance between the power receiving device 4 and the power transmitting device 2 outside the parking position.

[0079] Specifically, the power pulse of the present invention has a preset transmission distance, so that the vehicle-mounted terminal can only receive the power pulse from the current parking space 6 .

[0080] In a preferred embodiment, step S2 also includes: step S24, the vehicle-mounted device 3 measures the signal strength value RSSI at all wireless access points 1, and the master access point 11 collects the location fingerprint information

[0081] in represents the average value of the signal strength value RSSI from the nth wireless access point 1 collected by the vehicle-mounted device 3 at the lth parking space 6; l, n are positive integers belonging to [1, N], and N represents the number of wireless access points 1.

[0082] In a preferred embodiment, Figure 6 As shown, step S3 includes,

[0083] Step S31: The vehicle-mounted terminal measures the signal strength value RSSI of the wireless access point 1 within a certain range. m ;

[0084] Step S32, the master access point 11 sends the location fingerprint information to the vehicle-mounted device 3 in an inquiry manner or a mandatory manner;

[0085] Step S33, the machine learning model identifies the corresponding wireless access point by comparing the signal strength value of wireless access point 1 with the location fingerprint information;

[0086] Step S34, determining whether the wireless access point 1 to which the vehicle-mounted device 3 is currently connected is the corresponding wireless access point, if so, the vehicle-mounted device 3 maintains the connection with the current wireless access point 1; if not, the vehicle-mounted device 3 switches to connect with the corresponding wireless access point;

[0087] RSSI m represents the signal strength value of the mth wireless access point 1 around the vehicle-mounted device 3 , where m is a positive integer belonging to [1, N], and N represents the number of wireless access points 1 .

[0088] Specifically, since the power pulse method can accurately locate the parking space 6 of the car 5, its disadvantage is that it increases the power consumption of the charging system, thereby increasing the system operating cost and introducing more harmful radiation. Therefore, the basic idea of ​​the present invention is to only use the power pulse method for the necessary training stage, and there is no need to send power pulses in the inference stage;

[0089] In the inference phase of step S3, the vehicle-mounted device 3 cannot determine the position through the power pulse, and it measures the signal strength RSSI from the surrounding mth AP on an unknown parking space. m (m∈[1,N]), N represents the number of wireless access points 1 (that is, the number of parking spaces 6), and the RSSI is compared using a machine learning model such as KNN or k-means algorithm m and location fingerprint information The onboard device 3 can calculate its position.

[0090] The master access point 11 determines whether the machine learning system is ready, for example, whether the location fingerprint collection is completed, whether the location fingerprint collection needs to be updated, etc. The master access point 11 can force the vehicle-mounted device 3 to receive the location fingerprint information, or let the vehicle-mounted device 3 choose whether to receive the location fingerprint information by asking. The master access point 11 asks the vehicle-mounted device 3 that has already obtained the location fingerprint information without having to obtain the information again, reducing unnecessary communication, thereby reducing the delay in charging startup.

[0091] Car wireless charging parking position judgment system, such as Figure 2 As shown, including,

[0092] The wireless access points 1 are respectively arranged at the ground end, and the wireless access points 1 include a master access point 11 connected to the other wireless access points 1, and the master access point 11 is used to determine whether the machine learning model is ready and collect location fingerprint information;

[0093] The power transmission devices 2 correspond to the wireless access points 1 respectively and generate power pulses for identifying the parking position;

[0094] The vehicle-mounted device 3 is provided at the vehicle end and is used to identify the parking position according to the power pulse, select the wireless access point corresponding to the parking position for connection and measure the signal strength value of the wireless access point 1;

[0095] The power receiving device 4, corresponding to the vehicle-mounted device 3, is used to receive the power pulse;

[0096] The machine learning model is connected to the vehicle-mounted device 3 and the master access point 11, and is used to compare the signal strength value of the wireless access point 1 with the location fingerprint information to identify the parking location and the corresponding wireless access point.

[0097] Specifically, the system of the present invention is as follows Figure 1 As shown, each parking space 6 includes a power transmission device 2 and a wireless access point 1, and a power receiving device 4 and a vehicle-mounted device 3 are installed on the car 5 that needs to be charged. The wireless access point 1 of each parking space 6 is connected to the ground-end power transmission device 2 belonging to it, and the wireless access points 1 of all parking spaces 6 are connected to a master access point 11. Through the method and system for determining the parking position of a car wireless charging provided by the present invention, training data can be automatically collected and a fingerprint information library can be established. This method reduces the operating cost of the wireless charging system and reduces harmful electromagnetic radiation.

[0098] The above description is only a preferred embodiment of the present invention, and does not limit the implementation mode and protection scope of the present invention. For those skilled in the art, it should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for determining a parking position of a car with wireless charging, characterized in that: include, Step S1, after the car (5) is parked in the parking space, the vehicle-mounted device (3) establishes an initial connection with the wireless access point (1) within a certain range, and a master access point (11) determines whether the machine learning model is ready. If not, step S2 is executed; if ready, step S3 is executed; Step S2, the vehicle-mounted terminal receives a power pulse from the ground terminal, identifies the parking position and the wireless access point corresponding to the parking position according to the power pulse, the vehicle-mounted device (3) connects to the corresponding wireless access point, and the master access point (11) collects the position fingerprint information through the wireless access point (1); In step S3, the vehicle-mounted terminal measures the signal strength value of the wireless access point (1) within a certain range, the machine learning model identifies the parking position and the corresponding wireless access point by comparing the signal strength value of the wireless access point (1) with the position fingerprint information, and the vehicle-mounted device (3) connects to the corresponding wireless access point.

2. The method for determining the parking position of a wireless charging vehicle according to claim 1, characterized in that: In step S1, the vehicle-mounted device (3) selects one of the wireless access points (1) to establish an initial connection according to the signal strength value of the wireless access point (1).

3. The method for determining the parking position of a wireless charging vehicle according to claim 1, characterized in that: In step S1, the method for determining whether the machine learning model is ready includes: The master access point (11) determines whether the machine learning model has completed collecting the location fingerprint information or whether the location fingerprint information needs to be updated. If the location fingerprint information has been collected and does not need to be updated, it is determined that the machine learning model is ready, and the master access point (11) sends a preparation completion notification to the vehicle-mounted device, and executes step S3; otherwise, it is determined that the machine learning model is not ready, and the master access point (11) sends a non-readiness completion notification to the vehicle-mounted device, and executes step S2.

4. The method for determining the parking position of a wireless charging vehicle according to claim 1, characterized in that: In step S1, the wireless access point (1) includes the master access point (11).

5. The method for determining the parking position of a wireless charging vehicle according to claim 1, characterized in that: Step S2 comprises, Step S21, multiple ground-side power transmission devices (2) generate power pulses respectively; Step S22, the power receiving device (4) at the vehicle end receives the power pulse, and identifies the parking position where the corresponding power transmitting device (2) is located and the corresponding wireless access point according to the waveform of the received power pulse; Step S23, determining whether the wireless access point (1) to which the vehicle-mounted device (3) is currently connected is the corresponding wireless access point; if so, the vehicle-mounted device (3) maintains the connection with the current wireless access point (1); if not, the vehicle-mounted device (3) switches to connect with the corresponding wireless access point.

6. The method for determining the parking position of a wireless charging vehicle according to claim 5, characterized in that: In step S2, each ground end is a parking space (6), and each parking space (6) is provided with a matching power transmission device (2) and a matching wireless access point (1); and the vehicle end is provided with a matching power receiving device (4) and a matching vehicle-mounted device (3).

7. The method for determining the parking position of a wireless charging vehicle according to claim 5, characterized in that: In step S2, the propagation distance of the power pulse is less than the distance between the power receiving device (4) and the power transmitting device (2) outside the parking position.

8. The method for determining the parking position of a wireless charging vehicle according to claim 1, characterized in that: Step S2 also includes, in step S24, the vehicle-mounted device (3) measures the signal strength value RSSI at all the wireless access points (1), and the master access point (11) collects the location fingerprint information in represents the average value of the signal strength value RSSI from the nth wireless access point (1) collected by the vehicle-mounted device (3) at the lth parking space (6); l, n are positive integers belonging to [1, N], and N represents the number of the wireless access points (1).

9. The method for determining the parking position of a wireless charging vehicle according to claim 1, characterized in that: Step S3 comprises, Step S31, the vehicle-mounted terminal measures the signal strength value RSSI of the wireless access point (1) within a certain range m ; Step S32, the master access point (11) sends the location fingerprint information to the vehicle-mounted device (3) in an inquiry manner or a forced manner; Step S33, the machine learning model identifies the corresponding wireless access point by comparing the signal strength value of the wireless access point (1) with the location fingerprint information; Step S34, determining whether the wireless access point (1) to which the vehicle-mounted device (3) is currently connected is the corresponding wireless access point, and if so, the vehicle-mounted device (3) maintains the connection with the current wireless access point (1); if not, the vehicle-mounted device (3) switches to connect with the corresponding wireless access point; RSSI m represents the signal strength value of the mth wireless access point (1) around the vehicle-mounted device (3), where m is a positive integer belonging to [1, N], and N represents the number of the wireless access points (1).

10. A car wireless charging parking position determination system, characterized in that: include, Wireless access points (1) are respectively arranged at the ground end, the wireless access points (1) comprising a master access point (11) connected to the other wireless access points (1), the master access point (11) being used to determine whether the machine learning model is ready and to collect location fingerprint information; A power transmission device (2), corresponding to each of the wireless access points (1), generating a power pulse for identifying a parking position; The vehicle-mounted device (3) is arranged at the vehicle end and is used to identify the parking position according to the power pulse, select the wireless access point corresponding to the parking position for connection and measure the signal strength value of the wireless access point (1); A power receiving device (4), corresponding to the vehicle-mounted device (3), and configured to receive the power pulse; The machine learning model is connected to the vehicle-mounted device (3) and the master access point (11) and is used to compare the signal strength value of the wireless access point (1) with the location fingerprint information to identify the parking location and the corresponding wireless access point.

Images