A navigation processing method, device, navigation system and medium

Abstract

The present invention discloses a navigation processing method, device, navigation system and medium, and relates to the field of navigation technology. The grid position of a preset time period and the posture information of the vehicle in each preset time period are recorded in the map information. In the vehicle guidance process, each target space feature point is mapped on the gridded spherical surface to obtain the target position of each target space feature point mapped on the gridded spherical surface. After matching the target position with the grid position of the preset time period recorded in the map information, the posture information of the vehicle in the current time period during the guidance process is determined according to the posture of the preset time period corresponding to the maximum sum of the successfully matched grid areas, so as to realize the positioning of the vehicle in the guidance process, and determine the speed of the vehicle according to the posture of the adjacent time period; there is no measurement error caused by a third-party measurement tool, the measurement accuracy is improved, the measurement cost is reduced, and it can be shared for positioning for a variety of vehicle models, and has universality.

Description

Technical Field

[0001] The present invention relates to the field of navigation technology, and in particular to a navigation processing method, device, navigation system and medium. Background Art

[0002] With the rapid development of autonomous driving technology, the demand for environmental perception and precise positioning is growing. As a key component of the navigation system, the performance of mapping and positioning technology directly affects the reliability and safety of the navigation system.

[0003] Related navigation systems use monocular or multi-cameras to track targets, and combine third-party measurement tools (such as inertial measurement units (IMUs)) to measure target movement information, such as movement direction and movement distance, and then navigate based on the target movement information. Since third-party testing tools are needed during the navigation process, measurement errors will be introduced and navigation costs will increase.

[0004] It can be seen that providing a new navigation method to improve navigation accuracy and reduce navigation costs is a technical problem that people in this field urgently need to solve. Summary of the invention

[0005] The purpose of the present invention is to provide a navigation processing method, device, navigation system and medium to solve the technical problems of low navigation accuracy and high navigation cost.

[0006] In order to solve the above technical problems, the present invention provides a navigation processing method, comprising:

[0007] During the vehicle guidance process, at least one target spatial feature point of the current time period is obtained; wherein the spatial feature point is a point within the common field of view of multiple visual sensors in any time period;

[0008] Obtaining the target position of each target spatial feature point mapped on the rasterized spherical surface; wherein the number of the rasterized spherical surfaces is at least two;

[0009] Matching the target position with the grid position of the preset time period recorded in the map information; wherein the map information also records the position information of the vehicle in each preset time period;

[0010] The sum of the areas of the grids that are successfully matched in each preset time period is obtained respectively, and the target preset time period is determined according to the preset time period corresponding to the largest sum of all the areas;

[0011] The posture of the vehicle in the current period is determined according to the posture of the target preset period, and the speed of the vehicle is determined according to the posture of the target preset period and the posture of the previous preset period of the target preset period.

[0012] Preferably, obtaining the map information includes:

[0013] During the mapping process, images of each visual sensor at each time period are obtained;

[0014] Generate spatial feature points between different visual sensors in each time period according to the images of each visual sensor in each time period, the internal parameters and external parameters of each visual sensor;

[0015] Determine the posture information and position change information of the adjacent time periods according to the spatial feature points of the adjacent time periods; wherein the position change includes at least a translation change and a rotation change;

[0016] Determine the posture information of the local sub-coordinate system in the vehicle body coordinate system of the next period in the adjacent time period according to the posture information, position change information of the adjacent time period and the posture of the local sub-coordinate system in the vehicle body coordinate system of the previous period in the adjacent time period; wherein the local sub-coordinate system of each period is the vehicle body coordinate system at the starting time point of the corresponding period;

[0017] Mapping the spatial feature points of a preset time period onto the target gridded spherical surface, and obtaining the grid position information of the mapped spatial feature points on the target gridded spherical surface; wherein the preset time period is a time period selected from all time periods, and the target gridded spherical surface is the gridded spherical surface corresponding to each preset time period;

[0018] Obtaining the position and posture information of the local sub-coordinate system under the vehicle body coordinate system in each preset time period;

[0019] The grid position information of the spatial feature points in each preset time period and the position and posture information of the local sub-coordinate system in the vehicle body coordinate system are used as the map information.

[0020] Preferably, generating spatial feature points between different visual sensors in each time period according to the images of each visual sensor in each time period, the internal parameters and the external parameters of each visual sensor comprises:

[0021] Acquire target images collected by multiple visual sensors at different time periods, and extract characteristic pixel points of each target image;

[0022] Determine the position of the characteristic pixel points of each target image in each time period in the corresponding visual sensor coordinate system according to the internal parameters of each visual sensor, and obtain the position and posture of the visual sensor coordinate system in the local sub-coordinate system of each time period; wherein the external parameters of the visual sensor are the position and posture of the visual sensor coordinate system in the local sub-coordinate system of each time period;

[0023] Determine the position of the feature pixel points of each target image in each time period in the corresponding local sub-coordinate system according to the position of the feature pixel points of each target image in each time period in the corresponding visual sensor coordinate system and the posture of the visual sensor coordinate system in the local sub-coordinate system of each time period;

[0024] Generate characteristic rays pointing from the optical center of the visual sensor to the position of characteristic pixel points in the same local sub-coordinate system;

[0025] Obtain the length and common perpendicular point of the straight line where two characteristic rays of different visual sensors are located in the same period;

[0026] When it is detected that the common perpendicular point is located on the characteristic ray and the length of the common perpendicular line segment is less than or equal to the distance threshold, the center point of the line segment formed by the two common perpendicular points is determined as the spatial characteristic point.

[0027] Preferably, obtaining the position and posture of the visual sensor coordinate system in the local sub-coordinate system at each time period includes:

[0028] The position and posture of the visual sensor coordinate system in the vehicle coordinate system are obtained according to the external parameters of each visual sensor;

[0029] Obtain the position change of the vehicle body coordinate system at the target moment relative to the local sub-coordinate system of the corresponding time period, and determine the posture change matrix of the vehicle body coordinate system of the corresponding time period at the target moment according to the position change; wherein the target moment is the moment corresponding to the target image collected in each time period;

[0030] The pose of each visual sensor coordinate system in the local sub-coordinate system at each time period is determined according to the pose of the visual sensor coordinate system in the vehicle coordinate system and the pose change matrix of the vehicle coordinate system at the target moment.

[0031] Preferably, determining the preset time period includes:

[0032] During the mapping process, obtain the vehicle's mileage;

[0033] Dividing the mileage into a plurality of sections according to a preset distance interval; wherein each section corresponds to at least one time period;

[0034] The time period with the largest number of spatial feature points among all time periods corresponding to each segment is taken as the preset time period.

[0035] Preferably, obtaining the target position of each target spatial feature point mapped on the rasterized spherical surface includes:

[0036] Acquire a first target position of each target space feature point mapped on a first rasterized spherical surface and acquire a second target position of each target space feature point mapped on a second rasterized spherical surface; wherein the first rasterized spherical surface is located in a current vehicle body coordinate system, and the second rasterized spherical surface is located in a vehicle body coordinate system after the current vehicle body coordinate system is transformed in position, and the position transformation includes at least one of rotation or translation;

[0037] The step of respectively obtaining the sum of the areas of the grids that successfully match each preset time period, and determining the target preset time period according to the preset time period corresponding to the largest sum of the areas among all the sums of the areas comprises:

[0038] Acquire the sum of the areas of the grids that are successfully matched in each preset time period in the current vehicle body coordinate system, and use the preset time period corresponding to the largest sum of all the areas as the first target preset time period determined in the current vehicle body coordinate system;

[0039] Obtaining the sum of the areas of the grids that are successfully matched in each preset time period in the vehicle body coordinate system after the position transformation, and taking the preset time period corresponding to the largest sum of all the area sums as the second target preset time period determined in the vehicle body coordinate system after the position transformation;

[0040] Obtaining the maximum sum of the sum of the areas of the successfully matched grids corresponding to the first target preset time period and the sum of the areas of the successfully matched grids corresponding to each of the second target preset time periods;

[0041] The first target preset time period or the second target preset time period corresponding to the sum of the maximum areas is used as the target preset time period.

[0042] Preferably, determining the posture of the vehicle in the current period according to the posture of the target preset period includes:

[0043] Obtaining an offset of the vehicle body coordinate system corresponding to the target preset time period relative to the current vehicle body coordinate system;

[0044] The posture of the vehicle in the current period is determined according to the posture of the target preset period and in combination with the offset.

[0045] In order to solve the above technical problems, the present invention further provides a navigation processing device, comprising:

[0046] The first acquisition module is used to acquire at least one target spatial feature point in the current time period during the vehicle guidance process; wherein the spatial feature point is a point within the common field of view of multiple visual sensors in any time period;

[0047] A second acquisition module is used to acquire the target position of each target spatial feature point mapped on the rasterized spherical surface; wherein the number of the rasterized spherical surfaces is at least two;

[0048] A matching module, used to match the target position with the grid position of the preset time period recorded in the map information; wherein the map information also records the position information of the vehicle in each preset time period;

[0049] A first determination module is used to obtain the sum of the areas of the grids that are successfully matched in each preset time period, and determine the target preset time period according to the preset time period corresponding to the largest sum of all the areas;

[0050] The second determination module is used to determine the posture of the vehicle in the current period according to the posture of the target preset period, and to determine the speed of the vehicle according to the posture of the target preset period and the posture of the previous preset period of the target preset period.

[0051] In order to solve the above technical problems, the present invention further provides a navigation system, comprising:

[0052] Memory for storing computer programs;

[0053] A processor is used to implement the steps of the above-mentioned navigation processing method when executing the computer program.

[0054] In order to solve the above technical problem, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above navigation processing method are implemented.

[0055] The navigation processing method provided by the present invention first records the grid positions of preset time periods and the position and posture information of the vehicle in each preset time period in the map information. During the vehicle guidance process, at least one target space feature point of the current time period (a point within the public field of view of multiple visual sensors in the current time period) is obtained, and each target space feature point is mapped on the gridded spherical surface to obtain the target position of each target space feature point mapped on the gridded spherical surface. After matching the target position with the grid position of the preset time period recorded in the map information, the position information of the vehicle in the current time period during the guidance process is determined according to the position of the preset time period corresponding to the maximum sum of the successfully matched grid areas, thereby realizing the positioning of the vehicle in the guidance process, and determining the speed of the vehicle according to the position of the target preset time period and the position of the previous preset time period of the target preset time period; secondly, during positioning, the preset time period is determined by matching the grid on the rasterized spherical surface, thereby determining the position of the vehicle according to the position of the vehicle in the preset time period recorded in the map information, without the need to track the target and combine third-party measurement tools to measure the target movement information, thereby reducing measurement errors and reducing measurement costs; thirdly, the grid position of the preset time period recorded in the map information occupies a small space, can be shared by multiple vehicle models for positioning, and has universality; and compared with the use of a single rasterized spherical surface, at least two rasterized spherical surfaces are used for navigation in this method, thereby improving the accuracy of navigation.

[0056] In addition, the present invention also provides a navigation processing device, a navigation system and a computer-readable storage medium, which have the same or corresponding technical features as the above-mentioned navigation processing method and have the same effects as above. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A flowchart of a navigation processing method is provided for an embodiment of the present invention;

[0059] Figure 2 A schematic diagram of multi-camera mapping provided by an embodiment of the present invention;

[0060] Figure 3 A schematic diagram of generating spatial feature points (also called matching points) provided by an embodiment of the present invention;

[0061] Figure 4 A schematic diagram of a double spherical space feature point storage structure provided by an embodiment of the present invention;

[0062] Figure 5 A schematic diagram of rasterization of double spherical mapping points provided by an embodiment of the present invention;

[0063] Figure 6 A schematic diagram of a spherical coordinate system and a vehicle body coordinate system provided by an embodiment of the present invention;

[0064] Figure 7 A schematic diagram of a mapping of a spatial feature point to a storage surface of a double sphere provided by an embodiment of the present invention;

[0065] Figure 8 A schematic diagram of positioning provided by an embodiment of the present invention;

[0066] Fig. 9 A structural diagram of a navigation system provided in another embodiment of the present invention. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] The core of the present invention is to provide a navigation processing method, device, navigation system and medium to solve the technical problems of low navigation accuracy and high navigation cost.

[0069] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 A flowchart of a navigation processing method is provided for an embodiment of the present invention, such as Figure 1 As shown, the method includes:

[0070] S10: During the vehicle guidance process, obtaining at least one target spatial feature point in the current time period; wherein the spatial feature point is a point within the common field of view of multiple visual sensors in any time period;

[0071] S11: Obtaining the target position of each target spatial feature point mapped on the rasterized spherical surface; wherein the number of rasterized spherical surfaces is at least two;

[0072] S12: Matching the target position with the grid position of the preset time period recorded in the map information; wherein the map information also records the position information of the vehicle in each preset time period;

[0073] S13: respectively obtaining the sum of the areas of the grids that are successfully matched in each preset time period, and determining the target preset time period according to the preset time period corresponding to the largest sum of all the areas;

[0074] S14: determining the posture of the vehicle in the current period according to the posture of the target preset period, and determining the speed of the vehicle according to the posture of the target preset period and the posture of the previous preset period before the target preset period.

[0075] Vehicle navigation includes two processes: mapping and positioning. In order to locate the vehicle, first of all, a map needs to be built. Obtaining map information includes:

[0076] During the mapping process, images of each visual sensor at each time period are obtained;

[0077] Generate spatial feature points between different visual sensors in each time period according to the images of each visual sensor in each time period, the internal parameters and external parameters of each visual sensor;

[0078] Determine the posture information and position change information of the adjacent time periods according to the spatial feature points of the adjacent time periods; wherein the position change includes at least a translation change and a rotation change;

[0079] Determine the posture information of the local sub-coordinate system in the vehicle body coordinate system of the next period in the adjacent time period according to the posture information, position change information of the adjacent time period and the posture of the local sub-coordinate system in the vehicle body coordinate system of the previous period in the adjacent time period; wherein the local sub-coordinate system of each period is the vehicle body coordinate system at the starting time point of the corresponding period;

[0080] Mapping the spatial feature points of the preset time period onto the target gridded spherical surface, and obtaining the grid position information of the mapped spatial feature points on the target gridded spherical surface; wherein the preset time period is a time period selected from all time periods, and the target gridded spherical surface is the gridded spherical surface corresponding to each preset time period;

[0081] Obtaining the position and posture information of the local sub-coordinate system under the vehicle body coordinate system in each preset time period;

[0082] The grid position information of the spatial feature points in each preset time period and the posture information of the local sub-coordinate system under the vehicle body coordinate system are used as map information.

[0083] There is no limitation on the divided time periods, which are determined according to the specific circumstances. Adjacent time periods include the previous time period and the next time period in the total time period corresponding to the adjacent time period. For example, if time period A and time period B constitute adjacent time periods, time period A is the previous time period in the adjacent time period, and time period B is the next time period in the adjacent time period. The visual sensor used may be a camera. The preset time period may be each divided time period, or it may be a target time period selected from all time periods. In order to reduce the amount of data of the stored map information and improve the efficiency of map building, the preset time period is determined in the embodiment of the present invention, including: obtaining the mileage of the vehicle during the map building process; dividing the mileage into multiple sections according to the preset distance interval; wherein each section corresponds to at least one time period; and taking the time period with the largest number of spatial feature points in all time periods corresponding to each section as the preset time period. That is, the selected preset time period is the time period with the largest number of spatial feature points in all time periods selected as the preset time period.

[0084] In order to obtain spatial feature points, in implementation, the spatial feature points between different visual sensors in each time period are generated according to the images of each visual sensor in each time period, the internal parameters and external parameters of each visual sensor, including:

[0085] Acquire target images collected by multiple visual sensors at different time periods, and extract characteristic pixel points of each target image;

[0086] Determine the position of the characteristic pixel points of each target image in each time period in the corresponding visual sensor coordinate system according to the internal parameters of each visual sensor, and obtain the position and posture of the visual sensor coordinate system in the local sub-coordinate system of each time period; wherein the external parameters of the visual sensor are the position and posture of the visual sensor coordinate system in the local sub-coordinate system of each time period;

[0087] Determine the position of the feature pixel points of each target image in each time period in the corresponding local sub-coordinate system according to the position of the feature pixel points of each target image in each time period in the corresponding visual sensor coordinate system and the posture of the visual sensor coordinate system in the local sub-coordinate system of each time period;

[0088] Generate characteristic rays pointing from the optical center of the visual sensor to the position of characteristic pixel points in the same local sub-coordinate system;

[0089] Obtain the length and common perpendicular point of the straight line where two characteristic rays of different visual sensors are located in the same period;

[0090] When it is detected that the common perpendicular point is located on the characteristic ray and the length of the common perpendicular line segment is less than or equal to the distance threshold, the center point of the line segment formed by the two common perpendicular points is determined as the spatial characteristic point.

[0091] Since the layout of the cameras is distributed and the position and posture of the vehicle itself is often changing in real time, in order to improve the accuracy of positioning, the visual sensor coordinate system needs to be corrected. Specifically, obtaining the position and posture of the visual sensor coordinate system in the local sub-coordinate system at each time period includes:

[0092] The position and posture of the visual sensor coordinate system in the vehicle coordinate system are obtained according to the external parameters of each visual sensor;

[0093] Obtain the position change of the vehicle body coordinate system at the target time relative to the local sub-coordinate system of the corresponding time period, and determine the posture change matrix of the vehicle body coordinate system at the target time of the corresponding time period according to the position change; wherein the target time is the time corresponding to the target image collected in each time period;

[0094] The pose of each visual sensor coordinate system in the local sub-coordinate system at each time period is determined according to the pose of the visual sensor coordinate system in the vehicle coordinate system and the pose change matrix of the vehicle coordinate system at the target moment.

[0095] In order to make those skilled in the art better understand the present invention, Figure 2 The mapping process of the present invention is described with reference to the specific implementation modes. Figure 2 A schematic diagram of a multi-camera mapping method provided by an embodiment of the present invention is shown in FIG. Figure 2 As shown, maps of each update cycle are constructed respectively, and the maps of each update cycle are aggregated to generate a map. The method of constructing each update cycle map is the same. Here, the specific process of constructing the update cycle map is explained by taking the construction of the first update cycle map as an example.

[0096] Building the first update cycle map specifically includes the following process:

[0097] Extract feature points of images collected by each camera;

[0098] Generate a set of spatial feature points based on the internal and external parameters of the camera and the extracted image feature points;

[0099] Mapping the spatial feature points to the double grid sphere according to the spatial feature point set generated by the internal and external parameters of the camera;

[0100] Adjacent cycle matching positioning combined with vehicle odometer recursive posture;

[0101] Generate map information for the first update cycle (the map information for this update cycle includes the positions of spatial feature points mapped to the double grid sphere and the position and posture of the vehicle for this update cycle).

[0102] In the specific implementation, the map construction specifically includes the following processes:

[0103] Step 1: Record multiple camera images and vehicle odometer information. The specific operations are as follows:

[0104] The mapping process refers to the driver driving the mapping vehicle (with cameras, and the camera's viewpoints overlap in front of or behind the vehicle ) constructs and records the characteristic information of the working scene. Here, the vehicle body coordinate system at the starting moment (hereinafter referred to as the b system) is used as the map coordinate system (hereinafter referred to as the m system).

[0105] First, record the multi-camera images of the first period of the map construction process. The specific operation is as follows: When building the map, from the starting time point (The starting time of the map is triggered by the driver) to the time when all camera images are updated is the first period. The multi-camera images recorded in this period are recorded as ,in Each element in represents the corresponding The latest image updated by each camera during this period. Recording start time and The update timestamp of each image in . Also record The vehicle mileage output by the vehicle odometer at the moment , angular velocity and line speed , and remember .

[0106] Finally, record the multi-camera images and vehicle odometer information of the second period, the third period, and the final period until the end of the map building. The specific operation is as follows: As in the first step, clear all camera images from the time point Until the time when all camera images are updated again For the second period, generate multi-camera images of this period and their update timestamps , and its corresponding vehicle mileage and Similarly, multiple camera images and vehicle odometer information are generated at different times during the entire mapping project ( , , , ,in The whole mapping process is divided into time periods).

[0107] Step 2: Correct the position and posture of each camera coordinate system in each time period. The specific operations are as follows:

[0108] Each time period in the map building process is regarded as a small local subgraph. The coordinate system of the local subgraph is the b system at the starting time point of the time period. For the convenience of expression, the coordinate system of the local subgraph is referred to as the local sub-coordinate system, which is recorded as Department, among which Indicates The camera coordinate system is recorded as Department, among which It refers to A camera.

[0109] First, construct the poses of each camera coordinate system in the b system. The specific operations are as follows: For example, a camera ) external parameters (camera installation parameters), generate the b system Department location and posture Similarly, the positions of other cameras can be obtained and posture .

[0110] Then, the B-system position change and attitude matrix of each camera image update time in the local sub-coordinate system of each period are generated. The specific operation is as follows: According to the difference between the update timestamp of each camera image in each period and the start time of the period, the delay time of each camera image update can be obtained. , here is the The first Take a camera as an example. The midline velocity and angular velocity can be obtained from the The b system when the camera image is updated is relative to System translation change and rotation changes , as shown in equation 2-1. Next, generate the corresponding posture matrix , as shown in Equation 2-2 and Equation 2-3. Similarly, the first All the posture matrices in the time period By analogy, the posture matrix of the b system can be generated when each camera image is updated in each period.

[0111] Finally, the position and posture of each camera coordinate system in the local sub-coordinate system are generated for each time period. The specific operation is as follows: according to the posture matrix of the b system and the position and posture of the camera coordinate system in the b system when the camera image is updated (here the first Period, camera as an example), we can get Tie The posture of the and location , as shown in equation 2-4. Similarly, the pose of each camera at each time period can be generated.

[0112] 2-1

[0113] 2-2

[0114] 2-3

[0115] 2-4

[0116] Among them, Equation 2-2 and Equation 2-3 together constitute the posture matrix. The subscripts indicate the When the camera image is updated The attitude matrix of system b under system a.

[0117] Step 3: Generate characteristic rays of multi-camera images in each time period, and then triangulate the characteristic rays to generate spatial feature points. The specific operations are as follows:

[0118] Pixel coordinate system (hereinafter referred to as The system is on the imaging plane, and its unit is pixel, such as pixel ( i , j ) indicates that it is i OK j After extracting the feature pixels of the image (common methods include accelerated robust features (Features from Accelerated Segment Test, FAST), oriented fast and rotated BRIEF (Oriented Fast and Rotated Brief, ORB), perform the following operations.

[0119] First, generate the coordinate position of the feature pixel point of each camera in each time period in the local sub-coordinate system. The specific operation is as follows: Calculate Department (here The period camera image as an example) under the characteristic pixel points ( i , j ),exist The position of the tie , as shown in equation 3-1. At this point, we have Tie The poses under the system are and , then we can get the feature pixel point in The lower position is Similarly, the position of any characteristic pixel point in each camera in each time period and the corresponding local subsystem in the time period can be generated.

[0120] Then, a characteristic ray is generated from the camera optical center to the characteristic pixel position in the same local sub-coordinate system. The specific operation is as follows: The period The characteristic pixels of the image ( i , j ) as an example, construct the corresponding characteristic ray, that is, From the camera optical center point Point to feature pixel Similarly, all characteristic rays of the image can be obtained, as shown in equation 3-2. Similarly, characteristic rays of other time periods can be obtained.

[0121] Finally, triangulation is performed based on the characteristic rays to generate spatial feature points. The specific implementation is as follows: First, calculate the length of the common perpendicular segment of the straight line where the two characteristic rays of different cameras in the same period are located And the common perpendicular point. Then determine whether the common perpendicular points are all on the ray. If the common perpendicular points are all on the ray, and ( is the set distance threshold), then the midpoint of the two common perpendicular points , which are the spatial feature points. Similarly, all the spatial feature points between different cameras in any period of time can be obtained, as shown in equation 3-3:

[0122] 3-1

[0123] 3-2

[0124] 3-3

[0125] in, It's a camera The matrix composed of the internal parameters is the camera internal parameter matrix. express Department The characteristic straight line of a camera is , Indicates the horizontal coordinate of the starting point (camera The horizontal axis of Indicates the horizontal coordinate of the end point (if the camera is installed in front of the rear axle, it is + , and the rear is - ), N represents the total number of characteristic rays generated by the camera. Indicates The total number of spatial feature points in the time period, , Respectively indicate The spatial feature point set of the time period and the The coordinates of the spatial feature points. Figure 3 A schematic diagram of generating spatial feature points (also called matching points) provided by an embodiment of the present invention. Figure 3 As shown, camera 01 and camera 05 generate spatial feature point 01; camera 03 and camera 04 generate spatial feature point 02; camera 02 and camera 04 generate spatial feature point 03.

[0126] Step 4: Generate the poses of each local sub-coordinate system under the m system, and generate each update cycle in the mapping process. The specific operations are as follows:

[0127] First, the position and posture changes of adjacent time periods are generated. The specific operations are as follows: Use matching algorithms (such as iterative closest point (ICP), normal distributions transform (NDT)) to generate adjacent time periods (here the first Time period and The translation change of the time period , attitude matrix and rotation changes , as shown in Equations 4-1 and 4-2.

[0128] Then, the position and posture of the local sub-coordinate system in each period under the m system are generated. The specific operation is as follows: system (which coincides with the m system), and set the position and attitude respectively and Then, the position and attitude of the m system in the second period are calculated based on the translation change, attitude matrix and rotation change, respectively: and , as shown in Equation 4-3. Similarly, the position and attitude of the local sub-coordinate system at each time period under the m system at any time period can be obtained.

[0129] Finally, according to the mileage, multiple segments are divided, and the key time periods of the segments are screened out, and then each update cycle is delineated. The specific operations are as follows: According to the mileage in the mileage information, according to the distance interval The total mileage is divided into several sections (each section contains at least one time period). In each section, the time period with the largest number of spatial feature points among all time periods is taken as the key time period of the section, as shown in 4-4. Since a section and the key time period of the section are unique, they are collectively referred to as update cycles below.

[0130] 4-1

[0131] 4-2

[0132] 4-3

[0133] 4-4

[0134] Among them, A represents the set of spatial feature points of all update cycles in the entire mapping process, and B represents the position and posture of the vehicle in the m system corresponding to the starting time of each update cycle; , All of them indicate that The spatial feature points of the update cycle Indicates The update cycle uses the ( ) update period data; Represents the total number of update cycles in the entire mapping process ( ).

[0135] Step 5: Map the spatial feature points onto the grid double sphere of the update cycle to generate a normalized shareable map , the specific steps are as follows:

[0136] First, at the beginning of each update cycle, a double sphere is constructed. The specific operation is as follows: its coordinate system is composed of the b system at the beginning of the key period corresponding to each segment (which is also the local sub-coordinate system of each update cycle, such as Update period local sub-coordinate system ) is translated, the origins of the double spherical coordinate systems (O1 and O2) are respectively perpendicular to the chassis plane, passing through the rear axle centerline, and at a height of , the distance from the vehicle center axis is , and the origins of the two bi-spherical coordinate systems are symmetrical about the plane passing through the central axis and perpendicular to the ground, as shown in Figure 4. Figure 4 A schematic diagram of a dual spherical space feature point storage structure provided by an embodiment of the present invention. The sphere is a sphere with the center of the sphere as the origin of the coordinate system. (Debugging setting) is to rasterize the heading angle and pitch angle of the double sphere at intervals, and then obtain the grid double sphere, as shown in 5, Figure 5 A schematic diagram of rasterization of double sphere mapping points provided by an embodiment of the present invention. Constructing a grid sphere specifically includes the following process: 1) The origin of the rectangular coordinate system is the center of the sphere, and the length is set as the radius to construct the sphere; 2) The circle where the sphere intersects with the XOZ plane With the Z axis as the rotation axis, the The overlap line with the sphere during gradient rotation is called the vertical line. 3) Similarly, the circle intersecting with the XOY plane With the Y axis as the rotation axis, the is the interval rotation. The overlapping line with the sphere when the gradient rotates is called the horizontal line. The small blocks on the sphere divided by the horizontal and vertical lines are the grid surfaces. The set of all grid surfaces is the grid sphere. The above segmentation operation is called spherical rasterization. The left grid sphere and the right grid sphere are combined to form a double grid sphere or a grid double sphere. The coordinate system corresponding to the right grid sphere is recorded as Coordinate system, the coordinate system corresponding to the left grid sphere is recorded as Coordinate system, where Indicates Then, the spatial feature points of each update period are mapped onto the grid double sphere of the corresponding update period.

[0137] The specific operations are as follows: Figure 6 A schematic diagram of a spherical coordinate system and a vehicle body coordinate system provided in an embodiment of the present invention. Figure 7 A schematic diagram of a mapping of a spatial feature point to a storage surface of a double sphere provided by an embodiment of the present invention. Two line segments are generated by connecting the spatial feature point and the origin of the two spheres, and the grid blocks that intersect with the line segments are marked as highlighted, and the grid positions are recorded and stored. , ,in Represents mapping to The pitch angle Split segment, The heading angle Split segment, Represents the spherical area of ​​the grid block. Similarly Represents mapping to The pitch angle Split segment, The heading angle 1 segment, Represents the spherical area of ​​the grid block. Through the above, the mapping of the spatial feature point to the grid double sphere is completed, that is, the spatial feature point is projected onto the grid.

[0138] Finally, a normalized shareable map is generated , as follows: According to the above, after completing the mapping, build the map First, the storage matrix of the double spherical grid , ,in Each element in represents the set of spatial feature points of each update period mapped to the grid sphere on the right. Each element in represents the set of spatial feature points of each update period mapped to the grid sphere on the left, and its elements are the grid positions ( , ). Finally, normalize and generate a shareable map , It is the pose information in the m system corresponding to each update cycle, completing the mapping process.

[0139] The above process completes the process of mapping, and positioning can be performed based on the map information. During the vehicle guidance process, at least one target spatial feature point of the current time period is obtained, and the target position of each target spatial feature point mapped on the rasterized spherical surface is obtained, and then the target position is matched with the grid position of the preset time period recorded in the map information. In order to improve the success rate of matching, obtaining the target position of each target spatial feature point mapped on the rasterized spherical surface includes:

[0140] Obtaining a first target position of each target space feature point mapped on a first rasterized spherical surface and obtaining a second target position of each target space feature point mapped on a second rasterized spherical surface; wherein the first rasterized spherical surface is located in a current vehicle body coordinate system, and the second rasterized spherical surface is located in a vehicle body coordinate system after the current vehicle body coordinate system is transformed in position, and the position transformation includes at least one of rotation or translation;

[0141] The sum of the areas of the grids that are successfully matched in each preset time period is obtained respectively, and the target preset time period is determined according to the preset time period corresponding to the largest sum of all the areas, including:

[0142] The sum of the areas of the grids that are successfully matched in each preset time period is obtained in the current vehicle body coordinate system, and the preset time period corresponding to the largest sum of all the areas is used as the first target preset time period determined in the current vehicle body coordinate system;

[0143] The sum of the areas of the grids that are successfully matched in each preset time period is obtained in the vehicle body coordinate system after the position transformation, and the preset time period corresponding to the largest sum of all the area sums is used as the second target preset time period determined in the vehicle body coordinate system after the position transformation;

[0144] Obtaining the maximum sum of the sum of the areas of the successfully matched grids corresponding to the first target preset time period and the sum of the areas of the successfully matched grids corresponding to each second target preset time period;

[0145] The first target preset time period or the second target preset time period corresponding to the sum of the maximum areas is used as the target preset time period.

[0146] Since the position of the vehicle body coordinate system is changed, in order to determine the posture of the vehicle, in implementation, determining the posture of the vehicle in the current period according to the posture of the target preset period includes:

[0147] Obtaining the offset of the vehicle body coordinate system corresponding to the target preset time period relative to the current vehicle body coordinate system;

[0148] The vehicle's position and posture in the current period are determined based on the position and posture of the target preset period and combined with the offset.

[0149] Figure 8 A schematic diagram of positioning provided by an embodiment of the present invention, such as Figure 8 As shown in the figure, positioning specifically includes the following processes:

[0150] Extract feature points of images collected by each camera;

[0151] Generate a set of spatial feature points based on the internal and external parameters of the camera and the extracted image feature points;

[0152] According to the camera internal and external parameters, the vehicle coordinate system is transformed and translated to generate multiple sets of grid double spheres with different vehicle postures;

[0153] Map the spatial feature points onto each group of grid double spheres respectively;

[0154] Compared with each cycle in the shared map, the pose of the update cycle of the map record corresponding to the maximum overlapping area is considered to coincide with the translation and rotation pose;

[0155] According to the posture of the corresponding period in the shared map and the changes in the rotation and translation of the vehicle body, the posture of the vehicle coordinate system at the current moment is obtained, and the matching positioning output is performed.

[0156] In the specific implementation, after continuing the five steps in the above-mentioned mapping process, positioning specifically includes the following processes:

[0157] Step 6: According to the map , matching positioning, as described below:

[0158] First, generate the spatial feature points in the guidance process. The specific operation is as follows: As described in the first, second and third steps above, generate a set of spatial feature points corresponding to the vehicle multi-camera images corresponding to the current period in the guidance process. The spatial feature points are all in the local sub-coordinate system (hereinafter referred to as system).

[0159] Then, the spatial feature points are matched. The specific operation is as follows: During the first matching, the spatial feature points of the current period in the guidance process are The pose of the system (i.e. , as shown in Equation 6-1), and , The spatial feature points of each update cycle are matched on the double grid sphere (compared to see if there are overlapping areas), and the areas of all overlapping grid blocks are accumulated and recorded as (where subscript express Tie axis, The deflection gradients in the longitudinal and heading directions are all integers). Select the update period with the largest cumulative area of ​​all matches during the first match and the corresponding grid block cumulative area, recorded as , while recording The pose information corresponding to this update cycle is recorded as .

[0160] Then, according to the distance and angle gradient transformation, the optimal matching position is determined. The specific operation is as follows: Set the distance change gradient to The maximum deviation is ( is an integer); set the angle change gradient to The maximum deviation is ( is an integer). According to the above method, the maximum grid block cumulative area and the corresponding posture under different offsets are calculated and recorded as , as shown in Equation 6-2. The element with the largest cumulative area (where the subscripts represent the offset gradients of position and heading angle) The corresponding map The position in is taken as the optimal matching position. For the convenience of expression, it is assumed that Middle The update cycle is the optimal matching position, and its corresponding postures are , Here we will Middle The pose and After The poses after gradient migration are considered to be coincident.

[0161] Finally, select The element with the largest cumulative area and the generated m system The pose and visual odometer of the system are as follows: The translation vector and rotation vector corresponding to the distance offset and heading deviation of the above-mentioned maximum cumulative element are , yes After The pose matrix corresponding to the gradient offset pose , as shown in Equation 6-3. According to the best matching m-frame posture and position and heading gradient deflection, we can get the m-frame in the guidance process Coordinate position of the system , attitude angle , Linear speed and angular velocity As shown in Equation 6-4 and Equation 6-5.

[0162] , , 6-1

[0163] ; 6-2

[0164] ; 6-3

[0165] 6-4

[0166] 6-5

[0167] in, Represents the total number of feature points in the current space. , Indicates the shared map with the best match. The pose in the m system during each update cycle. , Indicates the position and posture generated in the previous update cycle at the current moment.

[0168] The navigation processing method provided by the embodiment of the present invention first records the grid positions of preset time periods and the position and posture information of the vehicle in each preset time period in the map information. During the vehicle guidance process, at least one target space feature point of the current time period (a point within the public field of view of multiple visual sensors in the current time period) is obtained, and each target space feature point is mapped on the gridded spherical surface to obtain the target position of each target space feature point mapped on the gridded spherical surface. After matching the target position with the grid position of the preset time period recorded in the map information, the position information of the vehicle in the current time period during the guidance process is determined according to the position of the preset time period corresponding to the maximum sum of the successfully matched grid areas, thereby realizing the positioning of the vehicle in the guidance process, and determining the speed of the vehicle according to the position of the target preset time period and the position of the previous preset time period of the target preset time period; secondly, during positioning, the preset time period is determined by matching the grid on the rasterized spherical surface, thereby determining the position of the vehicle according to the position of the vehicle in the preset time period recorded in the map information, without the need to track the target and combine third-party measurement tools to measure the target movement information, thereby reducing measurement errors and reducing measurement costs; thirdly, the grid position of the preset time period recorded in the map information occupies a small space, can be shared by multiple vehicle models for positioning, and has universality; and compared with the use of a single rasterized spherical surface, at least two rasterized spherical surfaces are used for navigation in this method, thereby improving the accuracy of navigation.

[0169] In the above embodiments, the navigation processing method is described in detail, and the present invention also provides corresponding embodiments of the navigation processing device and the navigation system. It should be noted that the present invention describes the embodiments of the device part from two perspectives, one is based on the functional module perspective, and the other is based on the hardware perspective.

[0170] A navigation processing device provided by an embodiment of the present invention, based on the perspective of functional modules, includes:

[0171] The first acquisition module is used to acquire at least one target spatial feature point in the current time period during the vehicle guidance process; wherein the spatial feature point is a point within the common field of view of multiple visual sensors in any time period;

[0172] The second acquisition module is used to acquire the target position of each target spatial feature point mapped on the rasterized spherical surface; wherein the number of the rasterized spherical surfaces is at least two;

[0173] A matching module, used to match the target position with the grid position of the preset time period recorded in the map information; wherein the map information also records the position information of the vehicle in each preset time period;

[0174] A first determination module is used to obtain the sum of the areas of the grids that are successfully matched in each preset time period, and determine the target preset time period according to the preset time period corresponding to the largest sum of all the areas;

[0175] The second determination module is used to determine the posture of the vehicle in the current period according to the posture of the target preset period, and to determine the speed of the vehicle according to the posture of the target preset period and the posture of the previous preset period of the target preset period.

[0176] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, which will not be repeated here.

[0177] Fig. 9 This is a structural diagram of a navigation system provided by another embodiment of the present invention. This embodiment is based on the hardware perspective, such as Fig. 9 As shown, the navigation system includes:

[0178] A memory 20, for storing computer programs;

[0179] The processor 21 is used to implement the steps of the navigation processing method mentioned in the above embodiment when executing the computer program.

[0180] Among them, the processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 21 can be implemented in at least one hardware form of a digital signal processor (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 21 may be integrated with a graphics processing unit (GPU), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 21 may also include an artificial intelligence (AI) processor, which is used to process computing operations related to machine learning.

[0181] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 20 is at least used to store the following computer program 201, wherein, after the computer program is loaded and executed by the processor 21, it can implement the relevant steps of the navigation processing method disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include, but is not limited to, the data involved in the above-mentioned navigation processing method, etc.

[0182] In some embodiments, the navigation system may further include a display screen 22 , an input / output interface 23 , a communication interface 24 , a power source 25 , and a communication bus 26 .

[0183] Those skilled in the art will understand that Fig. 9 The structure shown in the figure does not constitute a limitation of the navigation system, and may include more or fewer components than shown in the figure.

[0184] The navigation system provided by the embodiment of the present invention includes a memory and a processor. When the processor executes the program stored in the memory, it can implement the following method: a navigation processing method, the effect is the same as above.

[0185] Finally, the present invention also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps recorded in the above method embodiment are implemented.

[0186] It is understandable that if the method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc. Various media that can store program codes.

[0187] The computer-readable storage medium provided by the present invention includes the above-mentioned navigation processing method, and the effect is the same as above.

[0188] The above is a detailed introduction to a navigation processing method, device, navigation system and medium provided by the present invention. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the embodiments can be referenced to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the scope of protection of the present invention.

[0189] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A navigation processing method, characterized in that: include: During the vehicle guidance process, at least one target spatial feature point of the current time period is obtained; wherein the spatial feature point is a point within the common field of view of multiple visual sensors in any time period; Obtaining the target position of each target spatial feature point mapped on the rasterized spherical surface; wherein the number of the rasterized spherical surfaces is at least two; Matching the target position with the grid position of the preset time period recorded in the map information; wherein the map information also records the position information of the vehicle in each preset time period; The sum of the areas of the grids that are successfully matched in each preset time period is obtained respectively, and the target preset time period is determined according to the preset time period corresponding to the largest sum of all the areas; Determine the posture of the vehicle in the current period according to the posture of the target preset period, and determine the speed of the vehicle according to the posture of the target preset period and the posture of the previous preset period of the target preset period; The step of obtaining the target position of each target spatial feature point mapped on the rasterized spherical surface comprises: Obtaining the target position of each target spatial feature point mapped on the gridded spherical surface in the current vehicle body coordinate system and the target position on the gridded spherical surface in the vehicle body coordinate system after the current vehicle body coordinate system is transformed; The step of respectively obtaining the sum of the areas of the grids that successfully match each preset time period, and determining the target preset time period according to the preset time period corresponding to the largest sum of the areas among all the sums of the areas comprises: Acquire the sum of the areas of the grids that are successfully matched in each preset time period in the current vehicle body coordinate system, and use the preset time period corresponding to the largest sum of all the areas as the first target preset time period determined in the current vehicle body coordinate system; Obtaining the sum of the areas of the grids that are successfully matched in each preset time period in the vehicle body coordinate system after the position transformation, and taking the preset time period corresponding to the largest sum of all the area sums as the second target preset time period determined in the vehicle body coordinate system after the position transformation; Obtaining the maximum sum of the sum of the areas of the successfully matched grids corresponding to the first target preset time period and the sum of the areas of the successfully matched grids corresponding to each of the second target preset time periods; The first target preset time period or the second target preset time period corresponding to the sum of the maximum areas is used as the target preset time period.

2. The navigation processing method according to claim 1, characterized in that: Acquiring the map information includes: During the mapping process, images of each visual sensor at each time period are obtained; Generate spatial feature points between different visual sensors in each time period according to the images of each visual sensor in each time period, the internal parameters and external parameters of each visual sensor; Determine the posture information and position change information of the adjacent time periods according to the spatial feature points of the adjacent time periods; wherein the position change includes at least a translation change and a rotation change; Determine the posture information of the local sub-coordinate system in the vehicle body coordinate system of the next period in the adjacent time period according to the posture information, position change information of the adjacent time period and the posture of the local sub-coordinate system in the vehicle body coordinate system of the previous period in the adjacent time period; wherein the local sub-coordinate system of each period is the vehicle body coordinate system at the starting time point of the corresponding period; Mapping the spatial feature points of a preset time period onto the target gridded spherical surface, and obtaining the grid position information of the mapped spatial feature points on the target gridded spherical surface; wherein the preset time period is a time period selected from all time periods, and the target gridded spherical surface is the gridded spherical surface corresponding to each preset time period; Obtaining the position and posture information of the local sub-coordinate system under the vehicle body coordinate system in each preset time period; The grid position information of the spatial feature points in each preset time period and the position and posture information of the local sub-coordinate system in the vehicle body coordinate system are used as the map information.

3. The navigation processing method according to claim 2, characterized in that: The generating of spatial feature points between different visual sensors in each time period according to the images of each visual sensor in each time period, the internal parameters and the external parameters of each visual sensor comprises: Acquire target images collected by multiple visual sensors at different time periods, and extract characteristic pixel points of each target image; Determine the position of the characteristic pixel points of each target image in each time period in the corresponding visual sensor coordinate system according to the internal parameters of each visual sensor, and obtain the position and posture of the visual sensor coordinate system in the local sub-coordinate system of each time period; wherein the external parameters of the visual sensor are the position and posture of the visual sensor coordinate system in the local sub-coordinate system of each time period; Determine the position of the feature pixel points of each target image in each time period in the corresponding local sub-coordinate system according to the position of the feature pixel points of each target image in each time period in the corresponding visual sensor coordinate system and the posture of the visual sensor coordinate system in the local sub-coordinate system of each time period; Generate characteristic rays pointing from the optical center of the visual sensor to the position of characteristic pixel points in the same local sub-coordinate system; Obtain the length and common perpendicular point of the straight line where two characteristic rays of different visual sensors are located in the same period; When it is detected that the common perpendicular point is located on the characteristic ray and the length of the common perpendicular line segment is less than or equal to the distance threshold, the center point of the line segment formed by the two common perpendicular points is determined as the spatial characteristic point.

4. The navigation processing method according to claim 3, characterized in that: The obtaining of the position and posture of the visual sensor coordinate system in the local sub-coordinate system at each time period includes: The position and posture of the visual sensor coordinate system in the vehicle coordinate system are obtained according to the external parameters of each visual sensor; Obtain the position change of the vehicle body coordinate system at the target moment relative to the local sub-coordinate system of the corresponding time period, and determine the posture change matrix of the vehicle body coordinate system of the corresponding time period at the target moment according to the position change; wherein the target moment is the moment corresponding to the target image collected in each time period; The pose of each visual sensor coordinate system in the local sub-coordinate system at each time period is determined according to the pose of the visual sensor coordinate system in the vehicle coordinate system and the pose change matrix of the vehicle coordinate system at the target moment.

5. The navigation processing method according to claim 2, characterized in that: Determining the preset time period includes: During the mapping process, obtain the vehicle's mileage; Dividing the mileage into a plurality of sections according to a preset distance interval; wherein each section corresponds to at least one time period; The time period with the largest number of spatial feature points among all time periods corresponding to each segment is taken as the preset time period.

6. The navigation processing method according to any one of claims 1 to 5, characterized in that: The step of obtaining the target position of each target spatial feature point mapped on the rasterized spherical surface comprises: Obtain a first target position of each target space feature point mapped on a first rasterized spherical surface, and obtain a second target position of each target space feature point mapped on a second rasterized spherical surface; wherein the first rasterized spherical surface is located in a current vehicle body coordinate system, and the second rasterized spherical surface is located in a vehicle body coordinate system after the current vehicle body coordinate system is transformed in position, and the position transformation includes at least one of rotation or translation.

7. The navigation processing method according to claim 6, characterized in that: Determining the posture of the vehicle in the current period according to the posture of the target preset period includes: Obtaining an offset of the vehicle body coordinate system corresponding to the target preset time period relative to the current vehicle body coordinate system; The posture of the vehicle in the current period is determined according to the posture of the target preset period and in combination with the offset.

8. A navigation processing device, characterized in that: include: The first acquisition module is used to acquire at least one target spatial feature point in the current time period during the vehicle guidance process; wherein the spatial feature point is a point within the common field of view of multiple visual sensors in any time period; A second acquisition module is used to acquire the target position of each target spatial feature point mapped on the rasterized spherical surface; wherein the number of the rasterized spherical surfaces is at least two; A matching module, used to match the target position with the grid position of the preset time period recorded in the map information; wherein the map information also records the position information of the vehicle in each preset time period; A first determination module is used to obtain the sum of the areas of the grids that are successfully matched in each preset time period, and determine the target preset time period according to the preset time period corresponding to the largest sum of all the areas; A second determination module is used to determine the posture of the vehicle in the current period according to the posture of the target preset period, and to determine the speed of the vehicle according to the posture of the target preset period and the posture of the previous preset period of the target preset period; The second acquisition module is specifically used for: Obtaining the target position of each target spatial feature point mapped on the gridded spherical surface in the current vehicle body coordinate system and the target position on the gridded spherical surface in the vehicle body coordinate system after the current vehicle body coordinate system is transformed; The first determining module is specifically used for: Acquire the sum of the areas of the grids that are successfully matched in each preset time period in the current vehicle body coordinate system, and use the preset time period corresponding to the largest sum of all the areas as the first target preset time period determined in the current vehicle body coordinate system; Obtaining the sum of the areas of the grids that are successfully matched in each preset time period in the vehicle body coordinate system after the position transformation, and taking the preset time period corresponding to the largest sum of all the area sums as the second target preset time period determined in the vehicle body coordinate system after the position transformation; Obtaining the maximum sum of the sum of the areas of the successfully matched grids corresponding to the first target preset time period and the sum of the areas of the successfully matched grids corresponding to each of the second target preset time periods; The first target preset time period or the second target preset time period corresponding to the sum of the maximum areas is used as the target preset time period.

9. A navigation system, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the navigation processing method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the navigation processing method according to any one of claims 1 to 7 are implemented.

Images