Vehicle key locating method and apparatus, and vehicle and storage medium

Abstract

Provided in the present application are a vehicle key locating method and apparatus, and a vehicle and a storage medium. The method comprises: according to distances between a vehicle key and a plurality of locating anchor points of a vehicle, and a preset distance threshold value, determining a first location of the key, which first location indicates that the key is located in the vicinity of the vehicle; according to the distances and a pre-trained first machine-learning binary classifier, determining a second location of the key, which second location indicates that the key is located outside of the vehicle; according to the distances and a pre-trained second machine-learning binary classifier, determining a third location of the key, which third location indicates that the key is located on a lateral side of the vehicle; and according to the distances, a pre-trained machine-learning regression model and a preset triangle locating algorithm, determining final location coordinates of the key relative to the vehicle. A vehicle region is divided into areas of different levels, and a final location is determined in different ways, thereby realizing precise locating of a vehicle key.

Description

Vehicle key positioning method, device, vehicle and storage medium

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on June 16, 2023, with application number 2023107192340 and application name “Vehicle Key Positioning Method, Device, Vehicle and Storage Medium”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of automotive electronics technology, and in particular to a vehicle key positioning method, device, vehicle, and storage medium. Background Art

[0003] With technological advancements, the original mechanical car key has gradually evolved into a smart key for contactless entry and ignition. With the application of security-enhanced Ultra-Wide Band (UWB) wireless access technology to car keys, car keys have entered a new era that leverages the advantages of various wireless technologies. While UWB wireless access technology offers protection against relay attacks and high ranging and positioning accuracy, it is susceptible to reflection, diffraction, and multipath interference with the vehicle itself or surrounding conductive objects during operation, causing interference and affecting the positioning accuracy of car keys.

[0004] In the existing technology, in order to avoid the above problems, machine learning methods are usually used to locate car keys. However, the machine learning network is large in scale, has a large amount of calculation and requires high software and hardware data. The changeable actual environment outside the car is very different from the ideal environment when the training sample data is collected, which leads to large errors in actual positioning outside the car. The large range of the outside space and the narrow inside space make the gap between the number of training sample sets inside and outside the car too large, which also leads to problems such as low recognition rate inside the car.

[0005] In summary, how to accurately and stably achieve the positioning of car keys is a problem that needs to be solved urgently in this field.

[0006] Summary of the Invention

[0007] The present application provides a vehicle key positioning method, device, vehicle and storage medium to solve the problem of how to accurately and stably position the car key.

[0008] In a first aspect, the present application provides a vehicle key positioning method, applied to a vehicle, comprising:

[0009] determining a first position of the vehicle key based on distances between the vehicle key and a plurality of positioning anchor points of the vehicle and a preset distance threshold, wherein the first position is used to indicate that the vehicle key is in a far field or a near field of the vehicle;

[0010] If the first position indicates that the vehicle key is in the near field of the vehicle, a pre-trained first machine learning binary classifier determines a second position of the vehicle key based on the distance between the vehicle key and a plurality of positioning anchor points of the vehicle, wherein the second position is used to indicate that the vehicle key is outside or inside the vehicle, and the first machine learning binary classifier is an algorithm pre-trained based on a data set collected within the near field of the vehicle for determining the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points;

[0011] If the second position indicates that the vehicle key is outside the vehicle, a pre-trained second machine learning binary classifier determines a third position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, wherein the third position is used to indicate that the vehicle key is located to the side or body of the vehicle. The second machine learning binary classifier is an algorithm pre-trained based on a data set collected within the range outside the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points;

[0012] If the third position indicates that the vehicle key is on the side of the vehicle, the final position coordinates of the vehicle key relative to the vehicle are determined based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm. The machine learning regression model is an algorithm that is pre-trained based on a data set collected within the side range of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0013] In combination with the first aspect, in some embodiments, if the third position indicates that the vehicle key is to the side of the vehicle, determining the final position coordinates of the vehicle key relative to the vehicle based on the distances between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm includes:

[0014] The pre-trained machine learning regression model determines a first position coordinate of the vehicle key based on the distance between the vehicle key and a plurality of positioning anchor points of the vehicle;

[0015] determining a second position coordinate of the vehicle key using the triangulation positioning algorithm based on the distance between the vehicle key and a plurality of positioning anchor points of the vehicle;

[0016] The final position coordinates of the vehicle key relative to the vehicle are determined based on the first position coordinates and the second position coordinates.

[0017] In combination with the first aspect, in some embodiments, the method further includes:

[0018] If the first position indicates that the vehicle key is in the far field of the vehicle, the triangulation positioning algorithm determines the final position coordinates of the vehicle key relative to the vehicle based on the distances between the vehicle key and multiple positioning anchor points of the vehicle.

[0019] In combination with the first aspect, in some embodiments, the method further includes:

[0020] If the second position indicates that the vehicle key is inside the vehicle, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained first machine learning multi-classifier determines that the vehicle key is located in a first sub-area of ​​the vehicle, where the first sub-area includes the main driver's area, the co-driver's area, the rear seat area, or the trunk. The first machine learning multi-classifier is an algorithm that is pre-trained based on a data set collected within the interior of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0021] In combination with the first aspect, in some embodiments, the method further includes:

[0022] If the third position indicates that the vehicle key is on the vehicle body, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained second machine learning multi-classifier determines that the vehicle key is located in a second sub-area of ​​the vehicle, where the second sub-area includes the hood, the roof, or the trunk. The second machine learning multi-classifier is an algorithm that is pre-trained based on a data set collected within the vehicle body to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points.

[0023] In conjunction with the first aspect, in some embodiments, before determining the first position of the vehicle key based on the distances between the vehicle key and multiple positioning anchor points of the vehicle and a preset distance threshold, the method further includes:

[0024] Acquiring ranging signals between the vehicle key and multiple positioning anchor points of the vehicle;

[0025] The distance between the vehicle key and a plurality of positioning anchor points of the vehicle is determined according to the ranging signal.

[0026] In a second aspect, the present application provides a vehicle key positioning device, comprising:

[0027] a first determining module, configured to determine a first position of the vehicle key based on distances between the vehicle key and a plurality of positioning anchor points of the vehicle and a preset distance threshold, wherein the first position is used to indicate that the vehicle key is in a far field or a near field of the vehicle;

[0028] a second determination module, configured to, if the first position indicates that the vehicle key is in the near field of the vehicle, determine a second position of the vehicle key using a pre-trained first machine learning binary classifier based on distances between the vehicle key and a plurality of positioning anchor points of the vehicle, wherein the second position is used to indicate that the vehicle key is outside or inside the vehicle, the first machine learning binary classifier being an algorithm pre-trained based on a data set collected within the near field of the vehicle for determining the position of the vehicle key based on the distances between the vehicle key and the positioning anchor points;

[0029] a third determination module, configured to, if the second position indicates that the vehicle key is outside the vehicle, determine a third position of the vehicle key using a pre-trained second machine learning binary classifier based on the distances between the vehicle key and a plurality of positioning anchor points of the vehicle, wherein the third position indicates that the vehicle key is located to the side or body of the vehicle, the second machine learning binary classifier being an algorithm pre-trained based on a data set collected within a range outside the vehicle and used to determine the position of the vehicle key based on the distances between the vehicle key and the positioning anchor points;

[0030] and a fourth determination module, for determining, if the third position indicates that the vehicle key is on the side of the vehicle, the final position coordinates of the vehicle key relative to the vehicle based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained machine learning regression model, and a pre-set triangulation positioning algorithm.

[0031] In conjunction with the second aspect, in some embodiments, the fourth determining module includes:

[0032] a first determining unit, configured to determine a first position coordinate of the vehicle key based on the distance between the vehicle key and a plurality of positioning anchor points of the vehicle and the pre-trained machine learning regression model, wherein the machine learning regression model is an algorithm pre-trained based on a data set collected from a lateral range of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points;

[0033] a second determining unit, configured to determine a second position coordinate of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle and the triangulation positioning algorithm;

[0034] The third determining unit is configured to determine the final position coordinates of the vehicle key relative to the vehicle based on the first position coordinates and the second position coordinates.

[0035] In conjunction with the second aspect, in some embodiments, the apparatus further includes:

[0036] a fifth determination module, configured to determine, if the first position indicates that the vehicle key is in the far field of the vehicle, the final position coordinates of the vehicle key relative to the vehicle based on the distances between the vehicle key and multiple positioning anchor points of the vehicle and the triangulation positioning algorithm.

[0037] In conjunction with the second aspect, in some embodiments, the apparatus further includes:

[0038] The sixth determination module is used to determine, if the second position indicates that the vehicle key is inside the vehicle, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, using a pre-trained first machine learning multi-classifier to determine that the vehicle key is located in a first sub-area of ​​the vehicle, the first sub-area including the main driver's area or the co-driver's area or the rear seat area or the trunk, and the first machine learning multi-classifier is an algorithm that is pre-trained based on a data set collected within the interior of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0039] In conjunction with the second aspect, in some embodiments, the apparatus further includes:

[0040] The seventh determination module is used to determine, if the third position indicates that the vehicle key is on the vehicle body, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, using a pre-trained second machine learning multi-classifier to determine that the vehicle key is located in a second sub-area of ​​the vehicle, the second sub-area including the hood, the roof, or the trunk. The second machine learning multi-classifier is an algorithm pre-trained based on a data set collected within the vehicle body range to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0041] In conjunction with the second aspect, in some embodiments, before the first determining module, the apparatus further includes:

[0042] an acquisition module, configured to acquire ranging signals between the vehicle key and a plurality of positioning anchor points of the vehicle;

[0043] An eighth determination module is configured to determine the distance between the vehicle key and a plurality of positioning anchor points of the vehicle according to the ranging signal.

[0044] In a third aspect, the present application further provides a vehicle, comprising: a vehicle body, a storage unit disposed in the vehicle body, an electronic control unit, and a plurality of positioning anchor points;

[0045] The storage unit stores computer-executable instructions;

[0046] The electronic control unit executes the computer-executable instructions stored in the storage unit to implement the method described in any one of the above aspects.

[0047] In a fourth aspect, the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in any of the above aspects.

[0048] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which implements the method described in any one of the first aspects when executed by a processor.

[0049] The vehicle key positioning method, device, vehicle, and storage medium provided herein communicate with a vehicle key and multiple positioning anchors mounted on the vehicle body to obtain the distances between the vehicle key and the multiple positioning anchors. Based on the distances between the vehicle key and the multiple positioning anchors and a preset distance threshold, the vehicle key's first position is determined. If the first position indicates the vehicle key is in the near field of the vehicle, a pre-trained first machine learning binary classifier is used to determine the vehicle key's second position. If the second position indicates the vehicle key is outside the vehicle, a pre-trained second machine learning binary classifier is used to determine the vehicle key's third position. If the third position indicates the vehicle key is to the side of the vehicle, the final position coordinates of the vehicle key relative to the vehicle are determined based on the distances between the vehicle key and the multiple positioning anchors using a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm. By dividing the vehicle domain into multiple hierarchical regions, the success rate and accuracy of vehicle key position recognition are improved. The machine learning regression model and triangulation positioning algorithm are used to integrate and revise the vehicle key's position coordinates, achieving accurate and stable vehicle key recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] FIG1 is a diagram illustrating an application scenario of a vehicle key positioning method provided by an embodiment of the present application;

[0052] FIG2 is a flow chart of a first embodiment of a vehicle key positioning method according to an embodiment of the present application;

[0053] FIG3 is a schematic diagram of a specific process of a vehicle key positioning method provided in an embodiment of the present application;

[0054] FIG4 is a schematic diagram of vehicle area division according to an embodiment of the present application;

[0055] FIG5 is a flow chart of a second embodiment of a vehicle key positioning method according to an embodiment of the present application;

[0056] FIG6 is a flow chart of a third embodiment of a vehicle key positioning method according to an embodiment of the present application;

[0057] FIG7 is a flow chart of a fourth embodiment of a vehicle key positioning method according to an embodiment of the present application;

[0058] FIG8 is a schematic diagram of the functional area division of a vehicle provided in an embodiment of the present application;

[0059] FIG9 is a schematic structural diagram of a first embodiment of a vehicle key positioning device provided in an embodiment of the present application;

[0060] FIG10 is a schematic structural diagram of a second embodiment of a vehicle key positioning device provided in an embodiment of the present application;

[0061] FIG11 is a schematic structural diagram of a third embodiment of a vehicle key positioning device provided in an embodiment of the present application;

[0062] FIG12 is a schematic structural diagram of a vehicle provided in an embodiment of the present application.

[0063] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0064] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0065] With the continuous development of the automotive industry, the form and use of keys have also undergone tremendous changes. Today, the most common car key is the smart key. So-called smart keys come in a variety of forms, and keys are no longer just keys. For example, a card key can unlock the vehicle simply by sensing it at the door. You can also bind your key to a smart terminal, so next time you go out, you don't need to carry a key with you; you can carry the smart terminal to complete all operations. Communication between smart keys and vehicles is usually achieved through UWB wireless access technology, which has the technical characteristics of preventing relay attacks and high ranging and positioning accuracy. For the problem of smart key positioning, machine learning methods are usually used to collect a large number of training samples in different areas outside the vehicle and positions inside the vehicle, and then positioning is performed through the trained model. However, the changeable actual environment outside the vehicle is very different from the ideal environment when the training sample data was collected, resulting in large errors in actual positioning outside the vehicle. In addition, the large range of space outside the vehicle and the narrow space inside the vehicle make the gap between the number of training sample sets inside and outside the vehicle too large, which leads to problems such as low recognition rate inside the vehicle. If the smart key is positioned by the least squares method, if the smart key is near or inside the vehicle body, due to factors such as obstruction by the vehicle body and multiple reflections of radio waves inside the vehicle, the positioning output is unstable and the positioning accuracy is low.

[0066] In response to the above problems, the present application provides a vehicle key positioning method, which realizes accurate and stable positioning of vehicle keys. Specifically, for the problem of smart key positioning, it is usually necessary to train the machine learning model by collecting a large number of vehicle domain training samples, and then perform positioning by the trained model. However, due to the variability of the actual environment outside the vehicle and the large difference in the amount of data collected inside and outside the vehicle, problems such as low recognition rate inside the vehicle are caused. If the smart key is positioned by the least squares method, due to factors such as the obstruction of the vehicle body and multiple reflections of the radio waves inside the vehicle, the positioning output around the vehicle and inside the vehicle is unstable and the positioning accuracy is low. Taking these problems into consideration, the inventor studied whether it is possible to use the multipath distribution and delay spread characteristics of UWB signal propagation in different environments inside / outside the vehicle, and comprehensively adopt ranging / machine learning classification (recognition) / machine learning regression (positioning) / triangulation positioning algorithm for partition positioning. Based on this, the technical solution of the present application is proposed.

[0067] FIG1 is an application scenario diagram of the vehicle key positioning method provided by an embodiment of the present application. As shown in FIG1 , the application scenario includes at least a vehicle and a vehicle key. A plurality of positioning anchor points are provided on the vehicle, and the positioning anchor points are UWB1, UWB2, UWB3, and UWB4 as shown in FIG2 . The positioning anchor points UWB1, UWB2, UWB3, and UWB4 are all located outside the vehicle body, and UWB1 and UWB4 are located at the two corners of the front of the vehicle, and UWB2 and UWB3 are located at the two corners of the rear of the vehicle. The positioning anchor points are located outside the vehicle body and at the four corners of the front and rear of the vehicle, which can ensure that the ranging signal between the key and the positioning anchor point is not blocked by the vehicle body, or is not reflected or diffracted by the vehicle body. The positioning anchor point is deployed with a positioning device, which can determine the distance between the positioning anchor point and the vehicle key, or the positioning anchor point is an anchor point with a positioning function. The positioning device can be a UWB device including a UWB chip or a UWB module, or a positioning device including a BLE module or a BLE chip. The vehicle key also includes a BLE module or BLE chip. The vehicle key can be a physical key or a virtual key on a user terminal. The user terminal can be a smart terminal such as a mobile phone or tablet.

[0068] This application does not limit the specific form and type of the above-mentioned physical devices.

[0069] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0070] FIG2 is a flow chart of a first embodiment of a vehicle key locating method according to an embodiment of the present application. As shown in FIG2 , the vehicle key locating method according to the present application is applied to a vehicle and specifically includes:

[0071] S101: Determine a first position of the vehicle key based on distances between the vehicle key and multiple positioning anchor points of the vehicle and a preset distance threshold.

[0072] In this step, UWB wireless technology realizes communication based on electromagnetic wave propagation. At different positions of the vehicle, due to the different presence of conductors, the effectiveness of signal transmission between the vehicle key and the vehicle is different. In order to accurately identify the specific position of the vehicle key and thus accurately activate the function of the area, the different positions of the vehicle are divided into levels, and the first position of the vehicle key is determined based on the distance between the vehicle key and multiple positioning anchor points of the vehicle and the preset distance threshold.

[0073] Specifically, as shown in Figure 3, the vehicle key communicates with multiple positioning anchor points installed on the vehicle, and then obtains the ranging signals between the vehicle key and the multiple positioning anchor points of the vehicle through the multiple positioning anchor points. Then, based on the ranging signals, the distances between the vehicle key and the multiple positioning anchor points of the vehicle are determined. The shortest distance is selected from the distances between the vehicle key and the multiple positioning anchor points, and the shortest distance is compared with a preset distance threshold to determine the first position of the vehicle key. If the shortest distance is greater than the preset distance threshold, the first position indicates that the vehicle is in the far field of the vehicle. If the shortest distance is less than the preset distance threshold, the first position indicates that the vehicle is in the near field of the vehicle.

[0074] In a specific embodiment, as shown in FIG4 , taking a preset distance threshold of 3 meters and 4 positioning anchor points as an example, the 4 positioning anchor points are respectively set on the corners of the front and rear of the vehicle body. The area with the 4 positioning anchor points as the center and a radius of 3 meters is the vehicle near field, such as the shaded area in FIG4 , and any area outside the shaded area is the vehicle far field.

[0075] S102: If the first position indicates that the vehicle key is in the near field of the vehicle, a pre-trained first machine learning binary classifier is used to determine a second position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle.

[0076] In this step, by comparing the distance between the vehicle key and multiple positioning anchor points of the vehicle with a preset distance threshold, if it is determined that the first position indicates that the vehicle key is in the near field of the vehicle, the near field of the vehicle is still a relatively complex environment, which may be outside the vehicle close to the body or inside the vehicle. As in the above steps, in order to accurately identify and locate, the near field of the vehicle is further divided into different levels, and the distance between the vehicle key and multiple positioning anchor points of the vehicle is used as input and input into the pre-trained first machine learning binary classifier to determine the second position of the vehicle key.

[0077] Specifically, as shown in FIG3 , the pre-trained first machine learning binary classifier is a model obtained by training the machine learning binary classifier based on a data set pre-collected within the near-field range of the vehicle. The distance between the acquired vehicle key and multiple positioning anchor points of the vehicle is input into the pre-trained first machine learning binary classifier. After processing, the output is a first category value, wherein the first category value represents a corresponding value outside the vehicle or a corresponding value inside the vehicle. If the output is a corresponding value outside the vehicle, the second position indicates that the vehicle key is outside the vehicle. If the output is a corresponding value inside the vehicle, the second position indicates that the vehicle key is inside the vehicle. The category value can be represented by a number, for example, the outside of the vehicle corresponds to 0 and the inside of the vehicle corresponds to 1. If the output is 0, it means that the second position indicates that the vehicle key is outside the vehicle. If the output is 1, it means that the second position indicates that the vehicle key is inside the vehicle.

[0078] Optionally, by communicating between the vehicle key and multiple positioning anchor points, the received signal strength indication (RSSI) corresponding to each of the multiple positioning anchor points is obtained. The RSSI is used to characterize the strength of the communication capability between the corresponding positioning anchor point and the vehicle key. The RSSI and the distance between the vehicle key and the multiple positioning anchor points of the vehicle are input into the pre-trained first machine learning binary classifier to improve the accuracy of the output.

[0079] S103: If the second position indicates that the vehicle key is outside the vehicle, a pre-trained second machine learning binary classifier is used to determine a third position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle.

[0080] In this step, the vehicle near field is hierarchically divided through the above steps. If the second position obtained indicates that the vehicle key is outside the vehicle, it means that the vehicle key is located in the surrounding environment of the vehicle. However, in the process of communication between the vehicle key and multiple positioning anchor points, when the ranging signal sent by the vehicle key is a radio frequency signal, the radio frequency signal will be reflected, multipath interfered, and diffracted when encountering metal parts of the vehicle body, surrounding conductors such as the human body, etc. These phenomena will cause unstable positioning. Therefore, the external area of ​​the vehicle is further divided into levels, and the distance between the vehicle key and multiple positioning anchor points of the vehicle is used as input and input into the pre-trained second machine learning binary classifier to determine the third position of the vehicle key.

[0081] Specifically, the pre-trained second machine learning binary classifier is a model obtained by training the machine learning binary classifier based on a data set collected in advance within the range outside the vehicle. The specific process of model analysis is the same as the analysis process of the pre-trained first machine learning binary classifier in the above steps, and will not be repeated here.

[0082] Optionally, as shown in FIG3 , if the second position indicates that the vehicle key is outside the vehicle, a pre-trained machine learning binary classifier can be used to determine whether the vehicle key is within a preset distance range S1 around the vehicle body. If the vehicle key is outside the preset distance range S1 around the vehicle body, it indicates that the vehicle body has little impact on the signal transmission between the vehicle key and multiple positioning anchor points, and a triangulation positioning algorithm is then used to locate and interpret the vehicle key. If the vehicle key is within the preset distance range S1 around the vehicle body, the third position of the vehicle key is determined using the above method. For example, within a distance of 1 meter around the vehicle body, the positioning of the vehicle key is easily affected by the installation positions of multiple positioning anchor points. If positioning anchor points are also installed inside the vehicle, the positioning of the vehicle key within a distance of 1 meter around the vehicle body requires the distance between the vehicle key and the anchor points inside the vehicle. Therefore, before determining the third position of the vehicle key, a pre-trained machine learning binary classifier can be used to determine whether the vehicle key is within or outside a distance of 1 meter from the vehicle body.

[0083] S104: If the third position indicates that the vehicle key is on the side of the vehicle, the final position coordinates of the vehicle key relative to the vehicle are determined based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm.

[0084] In this step, the vehicle's exterior is divided into layers. If the third position indicates the key is to the side of the vehicle, due to obstruction by the vehicle body, a pre-trained machine learning regression model and a pre-set triangulation algorithm are used to analyze the distances between the key and multiple positioning anchor points of the vehicle to determine the final position coordinates of the key relative to the vehicle. The side of the vehicle primarily includes the left, rear, and right sides of the vehicle. The position coordinates are three-dimensional.

[0085] Specifically, as shown in A3 in Figure 3, if the vehicle key is located on any of the above-mentioned sides of the vehicle, the pre-trained machine learning regression model is trained based on the data set collected within the side range of the vehicle. The machine learning regression model is similar to the machine learning classifier analysis process. The distance between the vehicle key and multiple positioning anchor points of the vehicle is input into the pre-trained machine learning regression model. After processing, the first position coordinate of the vehicle key is output. According to the distance between the vehicle key and multiple positioning anchor points of the vehicle, an anchor point set is determined from multiple positioning anchor points. The distance between the positioning anchor point in the anchor point set and the key is corrected to obtain a corrected distance. The multiple corrected distances are used for positioning interpretation to determine the second position coordinate of the vehicle key. The first position coordinate and the second position coordinate are averaged to finally obtain the final position coordinate of the vehicle key relative to the vehicle, wherein the anchor point set corresponds one-to-one to the side of the vehicle.

[0086] Optionally, as shown in A3 in Figure 3, if the third position indicates that the vehicle key is on the side of the vehicle, the vehicle key can be determined to be on the left side, rear side, or right side of the vehicle based on a pre-trained machine learning multi-classifier. After determining that the vehicle key is on any one of the left side, rear side, or right side of the vehicle, the corresponding anchor point set is determined by the above method and then its positioning is solved.

[0087] The vehicle key positioning method provided in this embodiment determines the first position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle and a preset distance threshold. If the first position indicates that the vehicle key is in the near field of the vehicle, a pre-trained first machine learning binary classifier is used to determine the second position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle. If the second position indicates that the vehicle key is outside the vehicle, a pre-trained second machine learning binary classifier is used to determine the third position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle. If the third position indicates that the vehicle key is on the side of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm are used to determine the final position coordinates of the vehicle key relative to the vehicle based on the distance between the vehicle key and multiple positioning anchor points of the vehicle. By dividing the surrounding environment of the vehicle into regions, and using the pre-trained machine learning regression model and the pre-set triangulation positioning algorithm to process the side near the vehicle, the positioning accuracy of the vehicle key is improved and the recognition efficiency is enhanced.

[0088] FIG5 is a flow chart of a second embodiment of a vehicle key locating method according to an embodiment of the present application. As shown in FIG5 , based on the above embodiment, step S103 includes:

[0089] S1031: Determine the first position coordinates of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle using a pre-trained machine learning regression model.

[0090] In this step, in order to determine the first position coordinates of the vehicle key, the distances between the vehicle key and multiple positioning anchor points of the vehicle are input into a pre-trained machine learning regression model, and after processing, the first position coordinates of the vehicle key are obtained.

[0091] Specifically, the pre-trained machine learning regression model can be obtained after training based on any one of a linear regressor, a logistic regressor, and a stepwise regressor. Taking the pre-trained machine learning regression model based on the linear regressor training as an example, the distance between the vehicle key and multiple positioning anchor points of the vehicle is input into the pre-trained machine learning regression model, and the distance between the input vehicle key and the multiple positioning anchor points of the vehicle is set to X, and the output is Y. Then, according to the formula Y=b+ωX, the first position coordinates of the vehicle key are obtained. In the formula, b is the intercept, ω is the weight, and b and ω are both obtained during the training process.

[0092] S1032: Determine the second position coordinates of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle using a triangulation positioning algorithm.

[0093] In this step, for example, the inaccurate and unstable positioning caused by multipath interference, diffraction, and reflection can be reduced, and the second position coordinates of the vehicle key are determined through a triangulation positioning algorithm.

[0094] Specifically, based on the distances between the vehicle key and multiple positioning anchor points of the vehicle, at least two positioning anchor points whose ranging signal transmission mode is a non-obstructed transmission mode are selected from the multiple positioning anchor points as reference anchor points. If the number of reference anchor points exceeds two and the determined reference anchor points are collinear, the two reference anchor points with the largest distance between the reference anchor points are determined as target reference anchor points. If the number of reference anchor points exceeds two and the determined reference anchor points are not collinear, two adjacent reference anchor points whose connecting line is parallel to the ground are determined as target reference anchor points. If the number of reference anchor points exceeds two and the determined reference anchor points are not collinear, two adjacent reference anchor points whose sum of the distances between the two reference anchor points and the vehicle key is the smallest are determined as target reference anchor points. A local coordinate system is determined with the connecting line of the two target reference anchor points as the X-axis and a target reference anchor point on the X-axis as the coordinate origin. Based on the distances between each of the two target reference anchor points and the vehicle key, and the distance between the two target reference anchor points, a first coordinate of the vehicle key in the local coordinate system is determined. The first coordinate is then transformed to obtain a second position coordinate.

[0095] S1033: Determine the final position coordinates of the vehicle key relative to the vehicle based on the first position coordinates and the second position coordinates.

[0096] In this step, after obtaining the first position coordinates and the second position coordinates of the vehicle key through different methods, in order to make the positioning of the vehicle key more accurate and stable, the first position coordinates and the second position coordinates are averaged to obtain the final position coordinates of the vehicle key relative to the vehicle.

[0097] The vehicle key positioning method provided in this embodiment determines the first position coordinates of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle and a pre-trained machine learning regression model, determines the second position coordinates of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle and a triangulation positioning algorithm, and determines the final position coordinates of the vehicle key relative to the vehicle based on the first position coordinates and the second position coordinates. For the position of the body key on the side of the vehicle, the pre-trained machine learning regression model and the triangulation positioning algorithm results are fused to improve the stability of the vehicle key and increase the accuracy.

[0098] FIG6 is a flow chart of a third embodiment of a vehicle key locating method according to an embodiment of the present application. As shown in FIG6 , based on the first embodiment, the vehicle key locating method according to the present application further includes:

[0099] S105: If the first position indicates that the vehicle key is in the far field of the vehicle, a triangulation positioning algorithm is used to determine the final position coordinates of the vehicle key relative to the vehicle based on the distances between the vehicle key and multiple positioning anchor points of the vehicle.

[0100] In this step, as shown in A1 in Figure 3, by determining the preliminary level of the vehicle key, if the first position indicates that the vehicle key is in the far field of the vehicle, the final position coordinates of the vehicle key relative to the vehicle are determined based on the distance between the vehicle key and multiple positioning anchor points of the vehicle and the triangulation positioning algorithm. The specific processing process is the same as the processing process of step S1032 in the above embodiment and will not be repeated here.

[0101] S106: If the second position indicates that the vehicle key is inside the vehicle, a pre-trained first machine learning multi-classifier determines that the vehicle key is located in a first sub-region of the vehicle based on the distance between the vehicle key and multiple positioning anchor points of the vehicle.

[0102] S107: If the third position indicates that the vehicle key is located on the vehicle body, a pre-trained second machine learning multi-classifier determines that the vehicle key is located in a second sub-area of ​​the vehicle based on the distance between the vehicle key and multiple positioning anchor points of the vehicle.

[0103] In step S106 and step S107, by dividing the vehicle area hierarchically, the second position indicates that the vehicle key is inside the vehicle, and the third position indicates that the vehicle key is on the vehicle body. When the vehicle key is inside the vehicle or on the vehicle body, it is interfered by insufficient radio frequency signal transmission. In order to accurately determine the specific location of the vehicle key, a data set within the vehicle interior and a data set within the vehicle body are pre-collected. For the second position indicating that the vehicle key is inside the vehicle, as shown in A2 in Figure 3, based on the data set within the vehicle interior, the machine learning multi-classifier is trained separately to obtain a pre-trained first machine learning multi-classifier. Then, the distance between the vehicle key and multiple positioning anchor points of the vehicle is input into the pre-trained first machine learning multi-classifier, and a second category value is output. The second category value represents the value corresponding to the vehicle key being located in the first sub-area of ​​the vehicle. The first sub-area includes the main driver's area, the co-driver's area, the rear seat area, or the trunk. For the third position indicating that the vehicle key is on the vehicle body, as shown in A3 in Figure 3, based on the data set within the vehicle body range, the machine learning multi-classifiers are trained separately to obtain a pre-trained second machine learning multi-classifier, and then the distance between the vehicle key and multiple positioning anchor points of the vehicle is input into the pre-trained second machine learning multi-classifier, and a third category value is output. The third category value represents the numerical value corresponding to the vehicle key being located in the second sub-area of ​​the vehicle, and the second sub-area includes the hood, the roof, or the trunk.

[0104] Optionally, in step S106, the interior space of the vehicle is small and the positioning of the vehicle key is very inaccurate. In order to make the positioning of the vehicle key more accurate, as shown in A2 in Figure 3, a data set within the front cabin and trunk of the vehicle is pre-collected to train the machine learning classifier. The trained model is used to determine whether the vehicle key is located in the front cabin or trunk of the vehicle. The specific process is the same as the processing process of the aforementioned machine learning classifier and will not be repeated here.

[0105] In step S107 , the vehicle body also includes an area with a certain reserved distance above the vehicle body, such as an area with a certain reserved distance above the vehicle hood, roof, or trunk.

[0106] In a specific embodiment, taking the example of the corresponding value of the main driver's area being 3, the corresponding value of the passenger driver's area being 4, the corresponding value of the rear seat being 5, and the corresponding value of the trunk being 6, if the output is 3, it is determined that the vehicle key is located in the main driver's area of ​​the vehicle; if the output is 4, it is determined that the vehicle key is located in the passenger seat of the vehicle; if the output is 5, it is determined that the vehicle key is located in the rear seat of the vehicle; if the output is 6, it is determined that the vehicle key is located in the trunk of the vehicle.

[0107] The vehicle key positioning method provided in this embodiment, if the first position indicates that the vehicle key is in the far field of the vehicle, the triangulation positioning algorithm is used to determine the final position coordinates of the vehicle key relative to the vehicle based on the distance between the vehicle key and multiple positioning anchor points of the vehicle; if the second position indicates that the vehicle key is inside the vehicle, the pre-trained first machine learning multi-classifier is used to determine that the vehicle key is located in the first sub-area of ​​the vehicle based on the distance between the vehicle key and multiple positioning anchor points of the vehicle; if the third position indicates that the vehicle key is on the vehicle body, the pre-trained second machine learning multi-classifier is used to determine that the vehicle key is located in the second sub-area of ​​the vehicle based on the distance between the vehicle key and multiple positioning anchor points of the vehicle. By dividing the vehicle into regional hierarchies, the recognition and accuracy of vehicle key positioning are improved.

[0108] FIG7 is a flow chart of a fourth embodiment of a vehicle key locating method according to an embodiment of the present application. As shown in FIG7 , based on the first embodiment above, before step S101, the method further includes:

[0109] S108: Acquire ranging signals between the vehicle key and multiple positioning anchor points of the vehicle.

[0110] In this step, after the vehicle key is connected to the vehicle, the positioning devices on multiple positioning anchor points of the vehicle start the ranging operation, the vehicle key sends ranging signals to the multiple positioning anchor points, and the vehicle obtains the ranging signals between the vehicle key and the multiple positioning anchor points of the vehicle.

[0111] S109: Determine the distance between the vehicle key and multiple positioning anchor points of the vehicle based on the ranging signal.

[0112] In this step, after the vehicle obtains the ranging signal, the distance between the vehicle key and multiple positioning anchor points of the vehicle is obtained by multiplying the propagation time by the speed of light.

[0113] The vehicle key positioning method provided in this embodiment obtains the ranging signal between the vehicle key and multiple positioning anchor points of the vehicle, and determines the distance between the vehicle key and the multiple positioning anchor points of the vehicle based on the ranging signal, thereby realizing wireless communication between the vehicle key and the vehicle.

[0114] Figure 8 shows that after the vehicle key enters S6, it activates the welcome lights and music, and based on the key's left / right position, it also activates the ground light on the user's side. After the vehicle key enters S2 meters (the contour line ranges from 0 to 2 meters), there are many functional zones, such as the left front door, left rear door, right front door, right rear door, and tailgate left / right. When in the left / right door zone, the corresponding door is opened. When in the tailgate zone, the user's dwelling sequence and dwell time in the tailgate sub-zone can form a dwell trajectory pattern for the key holder, which can be used to activate the tailgate / lighting, etc. Leveraging UWB's high-precision positioning capabilities, different vehicle models can define different functions, the sub-functional zones that implement these functions, and the action sequences that implement these functions.

[0115] In the vehicle key locating method provided in the embodiment of the present application, for the machine learning classifier involved in the above embodiment, it is worth noting that the machine learning-neural network judges with different functions need to collect training sample sets separately and train the network separately to obtain dedicated network weights; the area for collecting training sample data must be consistent with the area that needs to be actually classified. For example, the classifier for judging inside / outside the car needs to collect data at various positions inside the car (front cabin + trunk) as inside-car samples, and collect data at various positions outside the car (including the front hood, trunk lid, outside of the window, roof within the contour line, and different positions within different ranges of 0-1 meter, 1-2 meters, 2-3 meters, and 3-6 meters outside the contour line as shown in Figure 4, and a certain height range) as outside-car samples, and obtain the weights of the network after training; the network hyperparameters of the classifiers with different functions (such as the number of hidden layers, the number of nodes per layer, etc.) need to be set through experiments based on indicators such as recognition accuracy, and can be the same or different.

[0116] FIG9 is a schematic structural diagram of a first embodiment of a vehicle key locating device according to an embodiment of the present application. As shown in FIG9 , the vehicle key locating device 200 includes:

[0117] The first determination module 201 is used to determine a first position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle and a preset distance threshold. The first position is used to indicate that the vehicle key is in the far field or near field of the vehicle.

[0118] The second determination module 202 is used to determine the second position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle if the first position indicates that the vehicle key is in the near field of the vehicle. The second position is used to indicate that the vehicle key is outside or inside the vehicle. The first machine learning binary classifier is an algorithm that is pre-trained based on a data set collected within the near field of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0119] The third determination module 203 is used to determine the third position of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle if the second position indicates that the vehicle key is outside the vehicle, using a pre-trained second machine learning binary classifier. The third position is used to indicate that the vehicle key is on the side or body of the vehicle. The second machine learning binary classifier is an algorithm that is pre-trained based on a data set collected within the external range of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0120] The fourth determination module 204 is used to determine the final position coordinates of the vehicle key relative to the vehicle if the third position indicates that the vehicle key is on the side of the vehicle, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm. The machine learning regression model is an algorithm that is pre-trained based on a data set collected within the side range of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0121] FIG10 is a schematic diagram of the structure of a second embodiment of a vehicle key locating device according to an embodiment of the present application. As shown in FIG10 , the fourth determining module 204 includes:

[0122] The first determination unit 2041 is used to determine the first position coordinates of the vehicle key based on the distance between the vehicle key and multiple positioning anchor points of the vehicle and a pre-trained machine learning regression model.

[0123] The second determining unit 2042 is configured to determine the second position coordinates of the vehicle key according to the distances between the vehicle key and multiple positioning anchor points of the vehicle and a triangulation positioning algorithm.

[0124] The third determining unit 2043 is configured to determine the final position coordinates of the vehicle key relative to the vehicle based on the first position coordinates and the second position coordinates.

[0125] FIG11 is a schematic structural diagram of a third embodiment of a vehicle key locating device according to an embodiment of the present application. As shown in FIG11 , the vehicle key locating device 200 further includes:

[0126] The fifth determination module 205 is configured to determine the final position coordinates of the vehicle key relative to the vehicle using a triangulation positioning algorithm based on the distances between the vehicle key and multiple positioning anchor points of the vehicle if the first position indicates that the vehicle key is in the far field of the vehicle.

[0127] The sixth determination module 206 is used to determine, if the second position indicates that the vehicle key is inside the vehicle, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, using a pre-trained first machine learning multi-classifier to determine that the vehicle key is located in a first sub-area of ​​the vehicle, the first sub-area including the main driver's area, the front passenger area, the rear seat area, or the trunk. The first machine learning multi-classifier is an algorithm that is pre-trained based on a data set collected within the interior of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

[0128] The seventh determination module 207 is used to determine, if the third position indicates that the vehicle key is on the vehicle body, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, using a pre-trained second machine learning multi-classifier to determine that the vehicle key is located in a second sub-area of ​​the vehicle, the second sub-area including the hood, roof, or trunk. The second machine learning multi-classifier is an algorithm pre-trained based on a data set collected within the vehicle body to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points.

[0129] The acquisition module 208 is configured to acquire ranging signals between the vehicle key and multiple positioning anchor points of the vehicle.

[0130] The eighth determination module 209 is configured to determine the distance between the vehicle key and a plurality of positioning anchor points of the vehicle according to the ranging signal.

[0131] The vehicle key locating device provided in this embodiment is used to execute the vehicle key locating method in any of the aforementioned method embodiments. Its implementation principle and technical effects are similar and will not be described in detail here.

[0132] FIG12 is a schematic structural diagram of a vehicle provided in an embodiment of the present application. As shown in FIG12 , the vehicle includes: a vehicle body 300 , a storage unit 301 , an electronic control unit 302 , and a plurality of positioning anchor points 303 ;

[0133] The storage unit 301 stores computer-executable instructions.

[0134] The electronic control unit 302 executes the computer-executable instructions stored in the storage unit 301 to implement the method in any one of the embodiments.

[0135] The plurality of positioning anchor points 303 are used to enable communication between the vehicle and the vehicle key.

[0136] An embodiment of the present application further provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the method in any one of the embodiments.

[0137] The computer-readable storage medium mentioned above may be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic disk, or optical disk. The computer-readable storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0138] Optionally, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0139] An embodiment of the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when at least one processor executes the computer program, it can implement the technical solution provided by any of the above method embodiments.

[0140] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0141] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A vehicle key positioning method, characterized in that: Applied to vehicles, including: Determine a first position of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle and a preset distance threshold, wherein the first position is used to indicate that the vehicle key is in a far field or a near field of the vehicle; If the first position indicates that the vehicle key is in the near field of the vehicle, a pre-trained first machine learning binary classifier determines a second position of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle, wherein the second position is used to indicate that the vehicle key is outside the vehicle or inside the vehicle, and the first machine learning binary classifier is an algorithm pre-trained according to a data set collected within the near field of the vehicle for determining the position of the vehicle key according to the distance between the vehicle key and the positioning anchor points; If the second position indicates that the vehicle key is outside the vehicle, a pre-trained second machine learning binary classifier determines a third position of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle, wherein the third position is used to indicate that the vehicle key is on the side or body of the vehicle, and the second machine learning binary classifier is an algorithm pre-trained according to a data set collected within the external range of the vehicle for determining the position of the vehicle key according to the distance between the vehicle key and the positioning anchor points; If the third position indicates that the vehicle key is on the side of the vehicle, the final position coordinates of the vehicle key relative to the vehicle are determined based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm. The machine learning regression model is an algorithm that is pre-trained based on a data set collected within the side range of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

2. The method according to claim 1, characterized in that If the third position indicates that the vehicle key is at the side of the vehicle, determining the final position coordinates of the vehicle key relative to the vehicle according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm, including: The pre-trained machine learning regression model determines a first position coordinate of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle; The triangulation positioning algorithm determines the second position coordinates of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle; The final position coordinates of the vehicle key relative to the vehicle are determined according to the first position coordinates and the second position coordinates.

3. The method according to claim 1 or 2, characterized in that: The method further comprises: If the first position indicates that the vehicle key is in the far field of the vehicle, the triangulation positioning algorithm determines the final position coordinates of the vehicle key relative to the vehicle according to the distances between the vehicle key and multiple positioning anchor points of the vehicle.

4. The method according to claim 1 or 2, characterized in that: The method further comprises: If the second position indicates that the vehicle key is inside the vehicle, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained first machine learning multi-classifier determines that the vehicle key is located in a first sub-area of ​​the vehicle, where the first sub-area includes the main driver's area, the co-driver's area, the rear seat area, or the trunk. The first machine learning multi-classifier is an algorithm that is pre-trained based on a data set collected within the interior of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points.

5. The method according to claim 1 or 2, characterized in that: The method further comprises: If the third position indicates that the vehicle key is on the vehicle body, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained second machine learning multi-classifier determines that the vehicle key is located in a second sub-area of ​​the vehicle, and the second sub-area includes the hood, the roof, or the trunk. The second machine learning multi-classifier is an algorithm that is pre-trained based on a data set collected within the vehicle body and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points.

6. The method according to claim 1, characterized in that Before determining the first position of the vehicle key according to the distances between the vehicle key and the plurality of positioning anchor points of the vehicle and a preset distance threshold, the method further includes: Acquiring a ranging signal between the vehicle key and a plurality of positioning anchor points of the vehicle; The distance between the vehicle key and a plurality of positioning anchor points of the vehicle is determined according to the ranging signal.

7. A vehicle key positioning device, characterized in that: include: A first determination module, configured to determine a first position of the vehicle key according to a distance between the vehicle key and a plurality of positioning anchor points of the vehicle and a preset distance threshold, wherein the first position is used to indicate that the vehicle key is in a far field or a near field of the vehicle; a second determination module, configured to determine, if the first position indicates that the vehicle key is in the near field of the vehicle, a second position of the vehicle key according to a pre-trained first machine learning binary classifier based on the distance between the vehicle key and a plurality of positioning anchor points of the vehicle, wherein the second position is used to indicate that the vehicle key is outside the vehicle or inside the vehicle, and the first machine learning binary classifier is an algorithm pre-trained based on a data set collected within the near field of the vehicle for determining the position of the vehicle key according to the distance between the vehicle key and the positioning anchor points; a third determination module, configured to determine, if the second position indicates that the vehicle key is outside the vehicle, a third position of the vehicle key based on the distance between the vehicle key and a plurality of positioning anchor points of the vehicle using a pre-trained second machine learning binary classifier, wherein the third position is used to indicate that the vehicle key is located on the side or body of the vehicle, and the second machine learning binary classifier is an algorithm pre-trained based on a data set collected within the external range of the vehicle and used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor points; The fourth determination module is used to determine the final position coordinates of the vehicle key relative to the vehicle if the third position indicates that the vehicle key is on the side of the vehicle, based on the distance between the vehicle key and multiple positioning anchor points of the vehicle, a pre-trained machine learning regression model and a pre-set triangulation positioning algorithm. The machine learning regression model is an algorithm that is pre-trained based on a data set collected within the side range of the vehicle and is used to determine the position of the vehicle key based on the distance between the vehicle key and the positioning anchor point.

8. The device according to claim 7, characterized in that The fourth determining module includes: A first determination unit, configured to determine a first position coordinate of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle and the pre-trained machine learning regression model; A second determining unit, configured to determine a second position coordinate of the vehicle key according to the distance between the vehicle key and a plurality of positioning anchor points of the vehicle and the triangulation positioning algorithm; The third determining unit is used to determine the final position coordinates of the vehicle key relative to the vehicle according to the first position coordinates and the second position coordinates.

9. A vehicle, characterized in that: include: A vehicle body, a storage unit disposed in the vehicle body, an electronic control unit, and a plurality of positioning anchor points; The storage unit stores computer-executable instructions; The electronic control unit executes the computer execution instructions stored in the storage unit to achieve the following The method according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the vehicle key locating method according to any one of claims 1 to 6.

Images