Vehicle positioning methods, devices, vehicles and storage media

Abstract

This application provides a vehicle positioning method, device, vehicle, and storage medium, relating to the field of intelligent driving technology. The vehicle determines its initial positioning based on its initial location, wheel speed at each moment during driving, and driving direction. The vehicle downloads an offline point cloud image corresponding to the initial positioning from a cloud server. The vehicle acquires environmental point cloud images via radar. On the offline point cloud images, the vehicle finds multiple mapped point cloud images with the same shape and size as the environmental point cloud images. The vehicle determines the average and covariance of the differences between points on the environmental point cloud images and corresponding points on each mapped point cloud image. The vehicle determines its position with high accuracy based on the average and covariance of the differences corresponding to each mapped point cloud image, the average of the largest and smallest differences among the multiple averages, and the largest and smallest covariances among the multiple covariances.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and in particular to a vehicle positioning method, device, vehicle, and storage medium. Background Technology

[0002] With the development of technology, people's demands for the driving experience are also increasing, giving rise to intelligent driving. Intelligent driving can free the driver's hands and achieve automatic vehicle operation. For example, when the vehicle detects an obstacle or pedestrian ahead, it will automatically brake; when the vehicle detects a fork in the road ahead, it will automatically turn according to the navigation route. Understandably, to achieve intelligent driving, the vehicle's position needs to be located.

[0003] Currently, one method for vehicle positioning is satellite positioning. However, when a vehicle is in an area with many obstructions (such as a dock, port quay crane, or deep mountains), the signal strength of the satellite signal is low, resulting in low accuracy in vehicle positioning. Summary of the Invention

[0004] This application provides a vehicle positioning method, device, vehicle, and storage medium to solve the problem of low accuracy in vehicle positioning in the prior art.

[0005] In a first aspect, this application provides a vehicle positioning method, comprising: determining the initial positioning of the vehicle based on its initial position, wheel speed at each moment during driving, and driving direction; downloading an offline point cloud image corresponding to the initial positioning from a cloud server; acquiring an environmental point cloud image via radar, wherein the area represented by the offline point cloud image is larger than the area represented by the environmental point cloud image; finding multiple mapped point cloud images with the same shape and size as the environmental point cloud image on the offline point cloud image, wherein the distance between the geometric center of each mapped point cloud image and the geometric center of the environmental point cloud image is less than a set distance threshold; determining the average value and covariance of the differences between points on the environmental point cloud image and corresponding points on each mapped point cloud image; determining the confidence level of each mapped point cloud image matching the environmental point cloud image based on the average value and covariance of the differences corresponding to each mapped point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among the covariances; and determining the vehicle's position based on the confidence level of each mapped point cloud image and the coordinates of its geometric center.

[0006] In one possible implementation, the vehicle determines the confidence level of matching each mapped point cloud image with the environmental point cloud image based on the average and covariance of the differences corresponding to each mapped point cloud image, the average of the largest and smallest differences among the averages of multiple differences, and the largest and smallest covariances among multiple covariances. This includes the vehicle using the following formula:

[0007] Determine the confidence level of each mapped point cloud image matching the environment point cloud image; where m ij Let n be the average of the differences corresponding to each mapped point cloud image. ij The covariance of each mapped point cloud image, max(m ij ) represents the average of the largest difference among multiple differences, min(m ij ) represents the average of the minimum differences among multiple differences, max(n) ij ) is the largest covariance among multiple covariances, min(n) ij P is the smallest covariance among multiple covariances. ij , where is the confidence level.

[0008] Understandably, the accuracy of the confidence level of matching each mapped point cloud image with the environmental point cloud image determined by the above method is high.

[0009] In one possible implementation, the vehicle determines its position based on the confidence level of each mapped point cloud image and the coordinates of its geometric center, including: the vehicle determining its position according to a formula. Determine the vehicle's location; where P ij For the confidence level of each mapped point cloud image, s ij [0] represents the x-coordinate of the geometric center of each mapped point cloud image, s ij [1] The ordinate of the geometric center of each mapped point cloud image, where w is a preset correction factor. The location of the vehicle has been determined.

[0010] This ensures a high degree of accuracy in determining the vehicle's location.

[0011] In one possible implementation, the environmental point cloud image includes an online environmental reflectance point cloud image and an online height point cloud image. The vehicle determines the confidence level of matching each mapped point cloud image with the environmental point cloud image based on the average and covariance of the differences corresponding to each mapped point cloud image, the average of the largest and smallest differences among the averages of multiple differences, and the largest and smallest covariances among multiple covariances. This includes: the vehicle determining the average and covariance of the differences between points on the online environmental reflectance point cloud image and corresponding points on each mapped reflectance point cloud image; the vehicle determining the average and covariance of the differences between points on the online height point cloud image and corresponding points on each mapped height point cloud image; and the vehicle determining the confidence level of matching each mapped point cloud image based on the average and covariance of the differences corresponding to each mapped point cloud image, the average of the largest and smallest differences among the averages of multiple differences, and the largest and smallest covariances among multiple covariances. The confidence level of matching each mapped point cloud image with the environment point cloud image is determined by the sum of the differences and the minimum covariance. This includes: the vehicle determining the confidence level of matching each mapped reflectivity point cloud image with the environment reflectivity point cloud image based on the average and covariance of the differences corresponding to each mapped reflectivity point cloud image, the average of the largest and smallest differences among the averages of multiple differences, and the largest and smallest covariances among multiple covariances; the vehicle determining the confidence level of matching each mapped height point cloud image with the environment height point cloud image based on the average and covariance of the differences corresponding to each mapped height point cloud image, the average of the largest and smallest differences among the averages of multiple differences, and the largest and smallest covariances among multiple covariances; the vehicle determining the first weight and the second weight of the confidence level of the mapped reflectivity point cloud image; and the vehicle using the following formula: Determine the confidence level of each mapped point cloud image in matching the environmental point cloud image, where, The confidence score for each mapped height point cloud image. To map the confidence level of the reflectance point cloud image, u is the first weight, 1-u is the second weight, and P... ij The confidence level of each mapped point cloud image, where v is a preset smoothing coefficient.

[0012] Understandably, the accuracy of the confidence level of matching each mapped point cloud image with the environmental point cloud image determined by the above method is high.

[0013] In one possible implementation, the vehicle determines a first weight for the confidence score of the mapped reflectance point cloud image and a second weight for the confidence score of the mapped height point cloud image, including: the vehicle according to a formula... Determine the error distribution coefficient α of the confidence level for each mapped reflectance point cloud image. r The error distribution coefficient α of the confidence level of each mapped height point cloud imageh s ij [0] represents the x-coordinate of the geometric center of each mapped reflectance point cloud image, s ij [1] The ordinate of the geometric center of each mapped reflectance point cloud image; the vehicle is calculated according to the formula. and Determine the error fusion coefficient γ for each mapped reflectance point cloud image. r Error fusion coefficient γ of each mapped height point cloud image h , where α r [0] represents the error distribution coefficient of the confidence level of the abscissa of each point on the mapped reflectance point cloud image, α r [1] The error distribution coefficient of the confidence level of the ordinate of the points on each mapped reflectance point cloud image, α h [0] The error distribution coefficient of the confidence level of the x-coordinate of the points on each mapped height point cloud image, α r [1] The error distribution coefficient of the confidence level of the ordinate of the points on each mapped height point cloud image; the vehicle is calculated according to the formula. Determine the first weight u for the confidence of the mapped reflectance point cloud image, where γ h [0] represents the error fusion coefficient for the abscissa of each point on the mapped height point cloud image; γ h [1] represents the error fusion coefficient of the ordinate of the points on each mapped height point cloud image; γ r [0] represents the error fusion coefficient of the abscissa of each point on the mapped reflectance point cloud image, γ. r [1] is the error fusion coefficient of the ordinate of each point on the mapped reflectance point cloud image. The vehicle determines the second weight z of the confidence of the mapped height point cloud image according to the formula z = 1 - u.

[0014] Understandably, the above method allows for high accuracy in determining the first weight of the confidence level of the mapped reflectance point cloud image and the second weight of the confidence level of the mapped height point cloud image.

[0015] In one possible implementation, after the vehicle acquires environmental point cloud images via radar, the method provided in this application further includes: the vehicle filtering the environmental point cloud images.

[0016] This results in a more robust environmental point cloud image.

[0017] In one possible implementation, after the vehicle downloads the offline point cloud image corresponding to the initial positioning from the cloud server, the method provided in this application further includes: the vehicle identifying whether the boundary between the initial positioning and the corresponding offline point cloud image is within a set distance range; if it is within the set distance range, then downloading the next offline point cloud image adjacent to the corresponding offline point cloud image in the driving direction.

[0018] This allows for the download of an offline point cloud image in advance, saving time spent on obtaining vehicle locations later.

[0019] Secondly, this application also provides a vehicle positioning device, comprising: an initial positioning determination unit, used to determine the initial positioning of the vehicle based on its initial position at startup, wheel speed at each moment during driving, and driving direction; a point cloud image download unit, used to download an offline point cloud image corresponding to the initial positioning from a cloud server; a point cloud image acquisition unit, used to acquire an environmental point cloud image via radar, wherein the area represented by the offline point cloud image is larger than the area represented by the environmental point cloud image; and a point cloud image search unit, used to find multiple mapped point cloud images with the same shape and size as the environmental point cloud image on the offline point cloud image, wherein each mapped point cloud image has several The distance between the geometric center of the point cloud image and the geometric center of the environment point cloud image is less than a set distance threshold; the parameter determination unit is used to determine the average value and covariance of the differences between points on the environment point cloud image and corresponding points in each mapped point cloud image; the parameter determination unit is also used to determine the confidence level of each mapped point cloud image matching the environment point cloud image based on the average value and covariance of the differences corresponding to each mapped point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances; the position determination unit is used to determine the position of the vehicle based on the confidence level of each mapped point cloud image and the coordinates of the geometric center.

[0020] Thirdly, this application provides a server including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the server performs the method provided in the first aspect.

[0021] Fourthly, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the computer to perform the method provided in the first aspect.

[0022] Fifthly, this application also provides a computer program product, including a computer program that, when run, causes a computer to perform the method provided in the first aspect.

[0023] This application provides a vehicle positioning method, device, vehicle, and storage medium. The vehicle finds multiple mapped point cloud images with the same shape and size as the acquired environmental point cloud image on the offline point cloud image corresponding to the initial positioning. The vehicle determines the confidence level of each mapped point cloud image's match with the environmental point cloud image based on the average and covariance of the differences between points on the environmental point cloud image and corresponding points on each mapped point cloud image. Furthermore, the vehicle determines its position based on the confidence level of each mapped point cloud image and the coordinates of its geometric center. This method results in a highly accurate and robust determination of the vehicle's position. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart of the vehicle positioning method provided in the embodiments of this application;

[0026] Figure 2 This is a circuit connection block diagram of a vehicle provided in an embodiment of this application;

[0027] Figure 3 This is a functional block diagram of the vehicle positioning device provided in the embodiments of this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments made by those skilled in the art under the guidance of these embodiments are within the scope of protection of this application.

[0029] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Currently, one method for vehicle positioning is satellite positioning. However, when a vehicle is in an area with many obstructions (such as a dock, port quay crane, or deep mountains), the signal strength of the satellite signal is low, resulting in low accuracy in vehicle positioning.

[0031] Based on the aforementioned technical problems, the inventive concept of this application is as follows: On the offline point cloud image corresponding to the initial positioning, the vehicle finds multiple mapped point cloud images with the same shape and size as the collected environmental point cloud image; the vehicle determines the confidence level of each mapped point cloud image matching the environmental point cloud image based on the average value and covariance of the differences between the points on the environmental point cloud image and the corresponding points on each mapped point cloud image; furthermore, the vehicle determines its position based on the confidence level of each mapped point cloud image and the coordinates of its geometric center. In this way, the determined position of the vehicle has high accuracy and good robustness.

[0032] The technical solutions of this application and how they solve the aforementioned technical problems will be described in detail below with specific embodiments. These 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 this application will now be described with reference to the accompanying drawings.

[0033] Please see Figure 1 This application provides a vehicle positioning method, applied to a vehicle, wherein, as Figure 2 As shown, the vehicle includes a processor, radar, wheel speedometer, and inertial measurement unit, wherein the vehicle positioning method includes:

[0034] S101: The vehicle determines its initial positioning based on its initial position at startup, wheel speed at each moment during travel, and direction of travel.

[0035] The wheel speed sensor 202 can collect the wheel speed at each moment, and the inertial measurement unit 203 can collect the driving direction at each moment. For example, the vehicle can integrate the wheel speed and driving direction at each moment based on its initial position to obtain the vehicle's initial positioning in the test area. This ensures high reliability in determining the vehicle's initial positioning in the test area.

[0036] S102: The vehicle downloads the offline point cloud image corresponding to the initial positioning from the cloud server.

[0037] The offline point cloud image is an environmental point cloud image pre-collected by the vehicle using radar 204 in a designated area, where the designated area includes the initial positioning described above. It can be understood that the offline point cloud image corresponding to the initial positioning includes the offline point cloud image collected in the designated area of ​​the initial positioning.

[0038] S103: The vehicle acquires environmental point cloud images via radar 204, wherein the area represented by the offline point cloud image is larger than the area represented by the environmental point cloud image.

[0039] S104: The vehicle finds multiple mapped point cloud images on the offline point cloud image that have the same shape and size as the environmental point cloud image.

[0040] Among them, the distance between the geometric center of each mapped point cloud image and the geometric center of the environment point cloud image is less than a set distance threshold.

[0041] S105: The average and covariance of the differences between points on the vehicle-determined environment point cloud image and the corresponding points in each mapped point cloud image.

[0042] In this context, the points on the environmental point cloud image and the corresponding points on the mapped point cloud image refer to the points on the mapped point cloud image that have the same coordinates as the points on the environmental point cloud image.

[0043] S106: The vehicle determines the confidence level of matching each mapped point cloud image with the environmental point cloud image based on the average value and covariance of the differences corresponding to each mapped point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances.

[0044] For example, S106 can be implemented as follows: the vehicle uses an arithmetic expression. Determine the confidence level of each mapped point cloud image in matching the environmental point cloud image.

[0045] Where, m ij Let n be the average of the differences corresponding to each mapped point cloud image. ij The covariance of each mapped point cloud image, max(m ij) represents the average of the largest difference among multiple differences, min(m ij ) represents the average of the minimum differences among multiple differences, max(n) ij ) is the largest covariance among multiple covariances, min(n) ij P is the smallest covariance among multiple covariances. ij , where is the confidence level.

[0046] Understandably, the accuracy of the confidence level of matching each mapped point cloud image with the environmental point cloud image determined by the above method is high.

[0047] S107: The vehicle determines its position based on the confidence level of each mapped point cloud image and the coordinates of the geometric center.

[0048] Understandably, the vehicle's processor 201 is used to execute S101-S107 as described above.

[0049] For example, S107 can be specifically implemented as follows: the vehicle calculates according to the formula. Determine the vehicle's location. Wherein, P... ij For the confidence level of each mapped point cloud image, s ij [0] represents the x-coordinate of the geometric center of each mapped point cloud image, s ij [1] The ordinate of the geometric center of each mapped point cloud image, where w is a preset correction factor. The location of the vehicle is determined. This ensures a high degree of accuracy in determining the vehicle's location.

[0050] In summary, this application provides a vehicle localization method. The vehicle finds multiple mapped point cloud images with the same shape and size as the acquired environmental point cloud image on the offline point cloud image corresponding to the initial localization. The vehicle determines the confidence level of each mapped point cloud image's match with the environmental point cloud image based on the average and covariance of the differences between points on the environmental point cloud image and corresponding points on each mapped point cloud image. Furthermore, the vehicle determines its position based on the confidence level of each mapped point cloud image and the coordinates of its geometric center. This method results in high accuracy and robustness in determining the vehicle's position.

[0051] It should be noted that, in the above... Figure 1 Based on the corresponding embodiments, in one possible implementation, the environmental point cloud image includes an online environmental reflectance point cloud image and an online height point cloud image. A specific implementation of S105 may include:

[0052] The vehicle determines the average and covariance of the differences between points on the online environmental reflectance point cloud image and their corresponding points on each mapped reflectance point cloud image. The vehicle also determines the average and covariance of the differences between points on the online elevation point cloud image and their corresponding points on each mapped elevation point cloud image.

[0053] The specific implementation of S106 can include:

[0054] Step 1-1: The vehicle determines the confidence level of matching each mapped reflectance point cloud image with the environmental reflectance point cloud image based on the average value and covariance of the differences corresponding to each mapped reflectance point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances.

[0055] Steps 1-2: The vehicle determines the confidence level of matching each mapped height point cloud image with the environmental height point cloud image based on the average value and covariance of the differences corresponding to each mapped height point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances.

[0056] Steps 1-3: The vehicle determines the first weight of the confidence score of the mapped reflectance point cloud image and the second weight of the confidence score of the mapped height point cloud image.

[0057] Specifically, the vehicle can be based on the formula. and Determine the error distribution coefficient α of the confidence score for each mapped reflectance point cloud image and the error distribution coefficient α of the confidence score for each mapped height point cloud image. h s ij [0] represents the x-coordinate of the geometric center of each mapped reflectance point cloud image, s ij [1] The ordinate of the geometric center of each mapped reflectance point cloud image.

[0058] Vehicles according to the formula and Determine the error fusion coefficient γ for each mapped reflectance point cloud image. r Error fusion coefficient γ of each mapped height point cloud image h Among them, α r [0] represents the error distribution coefficient of the confidence level of the abscissa of each point on the mapped reflectance point cloud image, α r [1] The error distribution coefficient of the confidence level of the ordinate of the points on each mapped reflectance point cloud image, α h [0] The error distribution coefficient of the confidence level of the x-coordinate of the points on each mapped height point cloud image, α r[1] The error distribution coefficient of the confidence level of the ordinate of the points on the point cloud image at each mapped height.

[0059] Vehicles according to the formula Determine the first weight u for the confidence of the mapped reflectance point cloud image, where γ h [0] represents the error fusion coefficient for the abscissa of each point on the mapped height point cloud image; γ h [1] represents the error fusion coefficient of the ordinate of the points on each mapped height point cloud image; γ r [0] represents the error fusion coefficient of the abscissa of each point on the mapped reflectance point cloud image, γ. r [1] is the error fusion coefficient of the ordinate of the points on each mapped reflectance point cloud image.

[0060] Then, the vehicle determines the second weight z of the confidence of the mapped height point cloud image according to the formula z = 1 - u.

[0061] Understandably, the above method allows for high accuracy in determining the first weight of the confidence level of the mapped reflectance point cloud image and the second weight of the confidence level of the mapped height point cloud image.

[0062] Steps 1-4: The vehicle is calculated according to the formula: Determine the confidence level of each mapped point cloud image in matching the environmental point cloud image.

[0063] in, The confidence score for each mapped height point cloud image. To map the confidence level of the reflectance point cloud image, u is the first weight, 1-u is the second weight, and P... ij The confidence level of each mapped point cloud image, where v is a preset smoothing coefficient.

[0064] Understandably, the accuracy of the confidence level of matching each mapped point cloud image with the environmental point cloud image determined by the above method is high.

[0065] Please see Figure 3 This application also provides a vehicle positioning device 300. It should be noted that the basic principle and technical effects of the vehicle positioning device 300 provided in this application are the same as those in the above embodiments. For the sake of brevity, any parts not mentioned in this application can be referred to the corresponding content in the above embodiments. The vehicle positioning device 300 includes an initial positioning determination unit 301, a point cloud image download unit 302, a point cloud image acquisition unit 303, a point cloud image search unit 304, a parameter determination unit 305, and a position determination unit 306.

[0066] The initial positioning determination unit 301 is used to determine the initial positioning of the vehicle based on its position at startup, wheel speed at each moment during driving, and driving direction.

[0067] The point cloud image download unit 302 is used to download the offline point cloud image corresponding to the initial positioning from the cloud server.

[0068] The point cloud image acquisition unit 303 is used to acquire environmental point cloud images through radar 204, wherein the area represented by the offline point cloud image is larger than the area represented by the environmental point cloud image.

[0069] In one possible implementation, the apparatus 300 provided in this application embodiment may further include: a filtering unit for filtering environmental point cloud images.

[0070] In one possible implementation, the apparatus 300 provided in this application embodiment may further include: a data download unit, configured to identify whether the boundary between the initial positioning and the corresponding offline point cloud image is within a set distance range; if it is within the set distance range, then download the next offline point cloud image adjacent to the corresponding offline point cloud image in the driving direction.

[0071] The point cloud image search unit 303 is used to find multiple mapped point cloud images with the same shape and size as the environment point cloud image on the offline point cloud image, wherein the distance between the geometric center of each mapped point cloud image and the geometric center of the environment point cloud image is less than a set distance threshold.

[0072] The parameter determination unit 305 is used to determine the average value and covariance of the difference between the points on the environmental point cloud image and the corresponding points in each mapped point cloud image.

[0073] The parameter determination unit 305 is also used to determine the confidence level of each mapped point cloud image matching the environment point cloud image based on the average value and covariance of the difference corresponding to each mapped point cloud image, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the largest covariance and the smallest covariance among multiple covariances.

[0074] The position determination unit 306 is used to determine the position of the vehicle based on the confidence level of each mapped point cloud image and the coordinates of the geometric center.

[0075] In one possible implementation, the parameter determination unit 305 is specifically used to employ a formula:

[0076] Determine the confidence level of each mapped point cloud image in matching the environmental point cloud image.

[0077] Where, m ijLet n be the average of the differences corresponding to each mapped point cloud image. ij The covariance of each mapped point cloud image, max(m ij ) represents the average of the largest difference among multiple differences, min(n) ij ) represents the average of the minimum differences among multiple differences, max(n) ij ) is the largest covariance among multiple covariances, min(n) ij P is the smallest covariance among multiple covariances. ij , where is the confidence level.

[0078] In one possible implementation, the position determination unit 306 is configured to determine the position based on the formula. Determine the vehicle's location.

[0079] Among them, P ij For the confidence level of each mapped point cloud image, s ij [0] represents the x-coordinate of the geometric center of each mapped point cloud image, s ij [1] The ordinate of the geometric center of each mapped point cloud image, where w is a preset correction factor. The location of the vehicle has been determined.

[0080] In one possible implementation, the environmental point cloud image includes an online environmental reflectance point cloud image and an online height point cloud image. The parameter determination unit 305 is specifically used to determine the average and covariance of the differences between points on the online environmental reflectance point cloud image and their corresponding points on each mapped reflectance point cloud image; and to determine the average and covariance of the differences between points on the online height point cloud image and their corresponding points on each mapped height point cloud image.

[0081] The parameter determination unit 305 is further specifically used to determine the confidence level of matching each mapped reflectivity point cloud image with the environmental reflectivity point cloud image based on the average value and covariance of the differences corresponding to each mapped reflectivity point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances; to determine the confidence level of matching each mapped height point cloud image with the environmental height point cloud image based on the average value and covariance of the differences corresponding to each mapped height point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances; to determine the first weight of the confidence level of the mapped reflectivity point cloud image and the second weight of the confidence level of the mapped height point cloud image; according to the formula: Determine the confidence level of each mapped point cloud image in matching the environmental point cloud image, where, The confidence score for each mapped height point cloud image. To map the confidence level of the reflectance point cloud image, u is the first weight, 1-u is the second weight, and P... ij The confidence level of each mapped point cloud image, where v is a preset smoothing coefficient.

[0082] In one possible implementation, the parameter determination unit 305 is specifically used to determine the formula... and Determine the error distribution coefficient α of the confidence score for each mapped reflectance point cloud image and the error distribution coefficient α of the confidence score for each mapped height point cloud image. h s ij [0] represents the x-coordinate of the geometric center of each mapped reflectance point cloud image, s ij [1] The ordinate of the geometric center of each mapped reflectance point cloud image; according to the formula and Determine the error fusion coefficient γ for each mapped reflectance point cloud image. r Error fusion coefficient γ of each mapped height point cloud image h .

[0083] Where, α r [0] represents the error distribution coefficient of the confidence level of the abscissa of each point on the mapped reflectance point cloud image, α r [1] The error distribution coefficient of the confidence level of the ordinate of the points on each mapped reflectance point cloud image, α h [0] The error distribution coefficient of the confidence level of the x-coordinate of the points on each mapped height point cloud image, α r [1] The error distribution coefficient of the confidence level of the ordinate of the points on each mapped height point cloud image; the vehicle is calculated according to the formula. Determine the first weight u for the confidence of the mapped reflectance point cloud image, where γ h [0] represents the error fusion coefficient for the abscissa of each point on the mapped height point cloud image; γ h [1] represents the error fusion coefficient of the ordinate of the points on each mapped height point cloud image; γ r [0] represents the error fusion coefficient of the abscissa of each point on the mapped reflectance point cloud image, γ. r [1] is the error fusion coefficient of the ordinate of each point on the mapped reflectance point cloud image. The vehicle determines the second weight z of the confidence of the mapped height point cloud image according to the formula z = 1 - u.

[0084] This application also provides a vehicle, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it causes the vehicle to perform the methods provided in the embodiments of this application described above.

[0085] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the computer to perform the method provided in the above embodiments of this application.

[0086] This application also provides a computer program product, including a computer program that, when run, causes a computer to perform the methods provided in the above embodiments.

[0087] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A vehicle positioning method, characterized in that, The method includes: The vehicle's initial positioning is determined based on its initial position at startup, wheel speed at each moment during travel, and direction of travel. The vehicle downloads an offline point cloud image corresponding to its initial location from the cloud server. The vehicle acquires environmental point cloud images via radar, wherein the area represented by the offline point cloud image is larger than the area represented by the environmental point cloud image, and the environmental point cloud image includes online environmental reflectivity point cloud images and online height point cloud images; The vehicle finds multiple mapped point cloud images on the offline point cloud image that have the same shape and size as the environmental point cloud image, wherein the distance between the geometric center of each mapped point cloud image and the geometric center of the environmental point cloud image is less than a set distance threshold. The vehicle determines the average value and covariance of the differences between points on the environmental point cloud image and corresponding points in each mapped point cloud image; The vehicle determines the average and covariance of the differences between points on the online environmental reflectance point cloud image and corresponding points on each mapped reflectance point cloud image; The vehicle determines the average value and covariance of the differences between points on the online elevation point cloud image and corresponding points on each mapped elevation point cloud image; the vehicle determines the confidence level of each mapped point cloud image matching the environmental point cloud image based on the average value and covariance of the differences corresponding to each mapped point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among the multiple covariances. The vehicle determines the confidence level of each of the mapped reflectance point cloud images matching the online environment reflectance point cloud image based on the average value and covariance of the differences corresponding to each of the mapped reflectance point cloud images, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances. The vehicle determines the confidence level of each mapped height point cloud image matching the online height point cloud image based on the average value and covariance of the differences corresponding to each mapped height point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances. The vehicle determines a first weight for the confidence score of the mapped reflectance point cloud image and a second weight for the confidence score of the mapped height point cloud image. The vehicle is based on the following formula: Determine the confidence level of each of the mapped point cloud images to match the environmental point cloud image, wherein, The confidence level of each of the mapped height point cloud images. The confidence level of the mapped reflectance point cloud image. For the first weight, 1- This is the second weight. The confidence level of each of the mapped point cloud images, This is the preset smoothing coefficient; The vehicle determines its position based on the confidence level of each of the mapped point cloud images and the coordinates of its geometric center.

2. The method according to claim 1, characterized in that, The vehicle determines the confidence level of matching each mapped point cloud image with the environmental point cloud image based on the average and covariance of the differences corresponding to each mapped point cloud image, the average of the largest and smallest differences among the averages of multiple differences, and the largest and smallest covariances among multiple covariances, including: The vehicle uses the following formula: Determine the confidence level of each of the mapped point cloud images in matching the environmental point cloud image; in, The average of the differences corresponding to each of the mapped point cloud images. The covariance corresponding to each of the mapped point cloud images, It is the average of the largest difference among multiple differences. It is the average of the smallest difference among multiple differences. The largest covariance among multiple covariances. The smallest covariance among multiple covariances. The confidence level is denoted as .

3. The method according to claim 1, characterized in that, The vehicle determines its position based on the confidence level of each mapped point cloud image and the coordinates of its geometric center, including: The vehicle is based on the formula Determine the location of the vehicle; in, The confidence level of each of the mapped point cloud images. Let x be the x-coordinate of the geometric center of each of the mapped point cloud images. The ordinate of the geometric center of each of the mapped point cloud images, The preset correction factor. The location of the vehicle is determined.

4. The method according to claim 1, characterized in that, The vehicle determines a first weight for the confidence score of the mapped reflectance point cloud image and a second weight for the confidence score of the mapped height point cloud image, including: The vehicle is based on the formula and Determine the error distribution coefficients for the confidence scores of each of the mapped reflectance point cloud images. Error distribution coefficients of the confidence scores of each of the mapped height point cloud images , Let x be the x-coordinate of the geometric center of each of the mapped reflectance point cloud images. The ordinate of the geometric center of each of the mapped reflectance point cloud images; The vehicle is based on the formula and , Determine the error fusion coefficients for each of the mapped reflectance point cloud images. Error fusion coefficients of each of the mapped height point cloud images ,in, The error distribution coefficient represents the confidence level of the abscissa of each point in the mapped reflectance point cloud image. The error distribution coefficient of the confidence level of the ordinate of each point on the mapped reflectance point cloud image. The error distribution coefficient of the confidence level of the x-coordinate of each point on the mapped height point cloud image. The error distribution coefficient of the confidence level of the ordinate of each point on the mapped height point cloud image; The vehicle is based on the formula First weight for determining the confidence of the mapped reflectance point cloud image ,in, The error fusion coefficient is the x-coordinate of the points on each of the mapped height point cloud images; The error fusion coefficient is the ordinate of the points on each of the mapped height point cloud images; The error fusion coefficient is the x-coordinate of each point on the mapped reflectance point cloud image. The error fusion coefficient is the ordinate of the point on each of the mapped reflectance point cloud images; The vehicle is based on the formula z=1- The second weight z is used to determine the confidence level of the mapped height point cloud image.

5. The method according to any one of claims 1-4, characterized in that, After the vehicle acquires environmental point cloud images via radar, the method further includes: The vehicle filters the environmental point cloud image.

6. The method according to any one of claims 1-4, characterized in that, After the vehicle downloads the offline point cloud image corresponding to the initial positioning from the cloud server, the method further includes: The vehicle identifies whether the boundary between the initial positioning and the corresponding offline point cloud image is within a set distance range; If the distance is within the set range, then download the next offline point cloud image that is adjacent to the corresponding offline point cloud image in the driving direction.

7. A vehicle positioning device, characterized in that, The device includes: An initial positioning determination unit is used to determine the initial positioning of the vehicle based on its initial position at startup, wheel speed at each moment during driving, and driving direction. The point cloud image download unit is used to download the offline point cloud image corresponding to the initial positioning from the cloud server; A point cloud image acquisition unit is used to acquire environmental point cloud images via radar, wherein the area represented by the offline point cloud image is larger than the area represented by the environmental point cloud image, and the environmental point cloud image includes an online environmental reflectivity point cloud image and an online height point cloud image; A point cloud image search unit is used to find multiple mapped point cloud images with the same shape and size as the environmental point cloud image on the offline point cloud image, wherein the distance between the geometric center of each mapped point cloud image and the geometric center of the environmental point cloud image is less than a set distance threshold. The parameter determination unit is used to determine the average value and covariance of the differences between points on the environmental point cloud image and points corresponding to each mapped point cloud image; The parameter determination unit is specifically used to determine the average value and covariance of the differences between the points on the online environmental reflectance point cloud image and the corresponding points on each mapped reflectance point cloud image; The vehicle determines the average value and covariance of the differences between points on the online height point cloud image and corresponding points on each mapped height point cloud image; the parameter determination unit is further configured to determine the confidence level of each mapped point cloud image matching the environmental point cloud image based on the average value and covariance of the differences corresponding to each mapped point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among the multiple covariances. The parameter determination unit is specifically used to determine the confidence level of each of the mapped reflectance point cloud images matching the online environment reflectance point cloud image based on the average value and covariance of the difference corresponding to each of the mapped reflectance point cloud images, the average value of the largest difference and the average value of the smallest difference among the average values ​​of multiple differences, and the largest covariance and the smallest covariance among multiple covariances. The vehicle determines the confidence level of each mapped height point cloud image matching the online height point cloud image based on the average value and covariance of the differences corresponding to each mapped height point cloud image, the average value of the largest and smallest differences among the average values ​​of multiple differences, and the largest and smallest covariances among multiple covariances. The vehicle determines a first weight for the confidence score of the mapped reflectance point cloud image and a second weight for the confidence score of the mapped height point cloud image. The vehicle is based on the following formula: Determine the confidence level of each of the mapped point cloud images to match the environmental point cloud image, wherein, The confidence level of each of the mapped height point cloud images. The confidence level of the mapped reflectance point cloud image. For the first weight, 1- This is the second weight. The confidence level of each of the mapped point cloud images, The preset smoothing coefficient; The position determination unit is used to determine the position of the vehicle based on the confidence level of each of the mapped point cloud images and the coordinates of the geometric center.

8. A vehicle comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it causes the vehicle to perform the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it causes the computer to perform the method as described in any one of claims 1 to 6.

Images