A position data correction processing method, device, equipment and storage medium

Abstract

The application discloses a position data correction processing method, device and equipment and a storage medium. The method comprises the following steps: acquiring position sequence information of at least one to-be-tested track point in a target area; performing multi-dimensional anomaly detection on the at least one to-be-tested track point based on the position sequence information, to obtain an anomaly detection result of the at least one to-be-tested track point; determining a target abnormal track point based on the anomaly detection result; and performing position correction processing on the target abnormal track point based on adjacent track points of the target abnormal track point and a target abnormal type corresponding to the target abnormal track point, to obtain target position sequence information. According to the embodiments of the application, multi-dimensional anomaly detection can be performed on track points, and corresponding correction processing can be performed on abnormal track points according to different abnormal types. The corrected track position information is more accurate, and the accuracy of the track position information can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a position data correction processing method and device, equipment and a storage medium. BACKGROUND

[0002] With the rapid development of artificial intelligence technology, the demand for high-precision maps for vehicle automatic driving is becoming more and more urgent. High-precision maps provide high-reliability map data for automatic driving, which can ensure the safe application of automatic driving.

[0003] In the prior art, high-precision maps are mainly made by using data collected by a vehicle-mounted laser scanning measurement system. Due to the influence of many factors such as vehicle mechanical characteristics, road flatness, driver driving habits, and complex on-site environment, the vehicle-mounted laser scanning measurement system is prone to abnormal situations such as jitter. For example, in an open area, the GNSS (Global Navigation Satellite System) positioning accuracy is relatively high, but is greatly affected by signal quality. In complex outdoor scenes such as urban canyons, under viaducts, and in tunnels, the GNSS signal is blocked by trees or buildings, resulting in unstable positioning accuracy. In a short time range, the IMU (Inertial Measurement Unit) measurement accuracy is relatively high, but as the mileage increases, the cumulative error gradually increases, so that the obtained trajectory data often has large deviations, and there are often repeated trajectory points, jumping trajectory points and other error data, resulting in distorted, misaligned and other uneven jumping problems of the generated point cloud data, which greatly affects the precision, efficiency and quality of the generated map. SUMMARY

[0004] To solve the above problems of the prior art, the present application provides a position data correction processing method, device, equipment and storage medium, which can perform multi-dimensional anomaly detection on trajectory points and perform corresponding correction processing on abnormal trajectory points according to different anomaly types. The corrected trajectory position information has higher relative accuracy and can improve the accuracy of the trajectory position information. The technical solutions of the present application are as follows:

[0005] According to an aspect of the embodiment of the present application, a position data correction processing method is provided, comprising:

[0006] Obtaining position sequence information of at least one to-be-measured trajectory point in a target area;

[0007] perform multi-dimensional anomaly detection on the at least one to-be-tested trajectory point based on the position sequence information, to obtain an anomaly detection result of the at least one to-be-tested trajectory point, the anomaly detection result representing a probability that the at least one to-be-tested trajectory point belongs to a plurality of abnormal trajectory points;

[0008] determine a target abnormal trajectory point based on the anomaly detection result;

[0009] perform position correction processing on the target abnormal trajectory point in the position sequence information based on adjacent trajectory points of the target abnormal trajectory point and a target abnormal type corresponding to the target abnormal trajectory point, to obtain target position sequence information.

[0010] Optionally, the anomaly detection result includes at least two of a first anomaly detection result corresponding to a first abnormal type, a second anomaly detection result corresponding to a second abnormal type, and a third anomaly detection result corresponding to a third abnormal type.

[0011] The first anomaly detection result is a first included angle between a first line corresponding to the at least one to-be-tested trajectory point and a second line corresponding to the at least one to-be-tested trajectory point, the first line corresponding to any to-be-tested trajectory point can be a line between the any to-be-tested trajectory point and a first adjacent trajectory point corresponding to the any to-be-tested trajectory point, the first adjacent trajectory point being a previous trajectory point in the adjacent trajectory points corresponding to the any to-be-tested trajectory point, and the second line corresponding to any to-be-tested trajectory point can be a line between the any to-be-tested trajectory point and a second adjacent trajectory point corresponding to the at least one to-be-tested trajectory point, the second adjacent trajectory point being a subsequent trajectory point in the adjacent trajectory points corresponding to the at least one to-be-tested trajectory point.

[0012] The second anomaly detection result is a second included angle between the first line and a first direction.

[0013] The third anomaly detection result is angle difference information between a roll angle corresponding to the at least one to-be-tested trajectory point and a roll angle corresponding to the second adjacent trajectory point.

[0014] Optionally, the position sequence information includes coordinate information and a roll angle, and the performing, based on the position sequence information, of the multi-dimensional anomaly detection on the at least one to-be-tested trajectory point to obtain the anomaly detection result of the at least one to-be-tested trajectory point includes:

[0015] determining the first included angle based on coordinate information of the at least one to-be-tested trajectory point, coordinate information of the first adjacent trajectory point, and coordinate information of the second adjacent trajectory point.

[0016] determining the second included angle based on the coordinate information of the at least one to-be-tested trajectory point and the coordinate information of the first adjacent trajectory point.

[0017] determine the angle difference information based on the roll angle of the at least one to-be-tested trajectory point and the roll angle of the second adjacent trajectory point.

[0018] Optionally, the position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, to obtain target position sequence information includes:

[0019] In a case where the target abnormal type is a first abnormal type, the position information of the target abnormal trajectory point is corrected to a midpoint position corresponding to a third connecting line in a first direction based on the position information of a first adjacent trajectory point and the position information of a second adjacent trajectory point, to obtain the target position sequence information.

[0020] The third connecting line is a connecting line between the first adjacent trajectory point and the second adjacent trajectory point, the first adjacent trajectory point is a previous trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point, and the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

[0021] Optionally, the position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, to obtain target position sequence information includes:

[0022] In a case where the target abnormal type is a second abnormal type, the position information of the target abnormal trajectory point is corrected to a midpoint position corresponding to a fourth connecting line in a second direction based on the position information of a first adjacent trajectory point and the position information of a second adjacent trajectory point, to obtain the target position sequence information.

[0023] The fourth connecting line is a connecting line between the first adjacent trajectory point and the second adjacent trajectory point, the first adjacent trajectory point is a previous trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point, and the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

[0024] Optionally, the position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, to obtain target position sequence information includes:

[0025] In a case where the target abnormal type is a third abnormal type, a roll angle corresponding to the target abnormal trajectory point is corrected based on position information of a second adjacent trajectory point, to obtain the target position sequence information.

[0026] The second adjacent trajectory point is a latter trajectory point among adjacent trajectory points corresponding to the target abnormal trajectory point.

[0027] Optionally, in a case where the target abnormal trajectory point includes a plurality of abnormal trajectory points, the position sequence information is corrected based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, to obtain target position sequence information, including:

[0028] The plurality of abnormal trajectory points are traversed based on a time sequence order of the plurality of abnormal trajectory points, and in a case where any abnormal trajectory point is traversed, the following process is performed:

[0029] The current traversed trajectory point is corrected based on an adjacent trajectory point of the current traversed trajectory point and a target abnormal type corresponding to the current traversed trajectory point, to obtain current position sequence information.

[0030] In a case where the plurality of abnormal trajectory points are traversed, the current position sequence information is taken as the target position sequence information.

[0031] Optionally, after the position sequence information is corrected based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, to obtain target position sequence information, the method further includes:

[0032] Target point cloud data is generated based on the target position sequence information.

[0033] A map of the target region is generated based on the target point cloud data.

[0034] According to another aspect of the disclosed embodiments, a position data correction processing device is provided, including:

[0035] A position sequence information acquisition module is configured to acquire position sequence information of at least one to-be-measured trajectory point of a target region.

[0036] An abnormality detection result generation module is configured to perform multi-dimensional abnormality detection on the at least one to-be-measured trajectory point based on the position sequence information, to obtain an abnormality detection result of the at least one to-be-measured trajectory point, the abnormality detection result representing a probability that the at least one to-be-measured trajectory point belongs to a plurality of abnormal trajectory points.

[0037] a target abnormal trajectory point determination module configured to determine a target abnormal trajectory point based on the abnormality detection result;

[0038] a correction processing module configured to perform position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, to obtain target position sequence information.

[0039] According to another aspect of the embodiments disclosed in the present application, there is provided an apparatus comprising a processor, a memory for storing instructions executable by the processor, wherein the processor is configured to execute the instructions to implement the position data correction processing method according to any one of the above aspects.

[0040] According to another aspect of the embodiments disclosed in the present application, there is provided a computer-readable storage medium storing instructions which, when executed by a processor of an electronic device, cause the electronic device to perform the position data correction processing method according to any one of the embodiments disclosed in the present application.

[0041] According to another aspect of the embodiments disclosed in the present application, there is provided a computer program product comprising instructions which, when executed on a computer, cause the computer to perform the position data correction processing method according to any one of the embodiments disclosed in the present application.

[0042] The technical solutions provided by the embodiments disclosed in the present application bring at least the following beneficial effects:

[0043] The position data correction processing method provided by the present application can accurately and efficiently determine abnormal trajectory points and their abnormal types by obtaining position sequence information of trajectory points, performing multi-dimensional abnormality detection on the trajectory points based on the position sequence information, and performing corresponding correction processing on the abnormal trajectory points based on their corresponding abnormal types and adjacent trajectory points, so that the abnormal trajectory points can be accurately corrected, the corrected trajectory position information has higher relative accuracy, and the accuracy of the trajectory position information can be improved.

[0044] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not intended to limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0045] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure and are not intended to limit the present disclosure.

[0046] Figure 1 is a flowchart of a position data correction processing method according to an exemplary embodiment;

[0047] Figure 2 is a diagram illustrating a first included angle according to an example embodiment;

[0048] Figure 3 is a diagram illustrating a second included angle according to an example embodiment;

[0049] Figure 4 is a block diagram of a position data correction processing device according to an example embodiment;

[0050] Figure 5 is a block diagram of a terminal device for position data correction processing according to an example embodiment;

[0051] Figure 6 is a block diagram of a server device for position data correction processing according to an example embodiment. DETAILED DESCRIPTION

[0052] In order to make the ordinary person skilled in the art better understand the technical solutions disclosed in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by the person skilled in the art without creative labor fall within the scope of protection of the present application.

[0053] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0054] The present application provides a position data correction processing method, which can be applied to the correction of trajectory position data.

[0055] Figure 1 is a flowchart of a position data correction processing method according to an example embodiment, as shown in Figure 1 The position data correction processing method includes the following steps.

[0056] S101: Obtain position sequence information of at least one to-be-tested trajectory point in a target area.

[0057] In one specific embodiment, the target area can be an area for which a map needs to be generated; and the position sequence information can be position information of the at least one to-be-tested trajectory point arranged in a time sequence of the at least one to-be-tested trajectory point. In actual application, the position sequence information can be collected by a mobile measurement system, which can include an inertial navigation system, a GNSS (Global Navigation Satellite System), a wheel speed meter, etc. Specifically, the position sequence information can reflect pose information of the inertial navigation system. Specifically, the position sequence information can include coordinate information and a roll angle, etc. Specifically, the coordinate information can be coordinates P(x, y, z) of the at least one to-be-tested trajectory point P in a WGS-84 coordinate system (World Geodetic System-1984 Coordinate System), where x can be a north direction coordinate or a latitude of the to-be-tested trajectory point, y can be an east direction coordinate or a longitude of the to-be-tested trajectory point, and z can be an elevation direction coordinate of the to-be-tested trajectory point. Specifically, the elevation direction can be a direction along a normal of a WGS-84 ellipsoid. The roll angle of any to-be-tested trajectory point is a roll angle of a trajectory collection vehicle at the to-be-tested trajectory point.

[0058] S103: Based on the position sequence information, performing multi-dimensional anomaly detection on the at least one to-be-tested trajectory point to obtain an anomaly detection result of the at least one to-be-tested trajectory point.

[0059] In one specific embodiment, the anomaly detection result represents an anomaly type to which the at least one to-be-tested trajectory point belongs. Specifically, the anomaly type can be an anomaly type of a trajectory point, and the anomaly type can include a first anomaly type, a second anomaly type, and a third anomaly type.

[0060] In one optional embodiment, the anomaly detection result can include at least two of a first anomaly detection result corresponding to the first anomaly type, a second anomaly detection result corresponding to the second anomaly type, and a third anomaly detection result corresponding to the third anomaly type.

[0061] The first abnormality detection result is a first included angle between a first line corresponding to the at least one to-be-detected trajectory point and a second line corresponding to the at least one to-be-detected trajectory point, the first line corresponding to any to-be-detected trajectory point can be a line between the any to-be-detected trajectory point and a first adjacent trajectory point corresponding to the any to-be-detected trajectory point, the first adjacent trajectory point is a previous trajectory point in the adjacent trajectory points corresponding to the any to-be-detected trajectory point, and the second line corresponding to any to-be-detected trajectory point can be a line between the any to-be-detected trajectory point and a second adjacent trajectory point corresponding to the at least one to-be-detected trajectory point, the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the at least one to-be-detected trajectory point;

[0062] The second abnormality detection result is a second included angle between the first line and a first direction;

[0063] The third abnormality detection result is angle difference information between a roll angle corresponding to the at least one to-be-detected trajectory point and a roll angle corresponding to the second adjacent trajectory point.

[0064] Specifically, the first abnormality type is an abnormality type corresponding to a first abnormality detection dimension, the second abnormality type is an abnormality type corresponding to a second abnormality detection dimension, and the third abnormality type is an abnormality type corresponding to a third abnormality detection dimension. The first direction can be a horizontal direction.

[0065] In one specific embodiment, the abnormality types can include an elevation jump, a horizontal jump, and a posture jump. Specifically, the first abnormality type can be a horizontal jump, the second abnormality type can be an elevation jump, and the third abnormality type can be a posture jump.

[0066] In one specific embodiment, the elevation jump is reflected in that the elevation direction coordinate of a trajectory point suddenly increases or decreases, causing the point cloud to appear in a waterfall fault or up-down distortion in this range, and the trajectory point needs to be corrected in the elevation direction to make the change moderate and consistent with the real road plane or gentle slope; the horizontal jump is reflected in that the position of a trajectory point in the horizontal direction suddenly jumps out of the real trajectory line, causing the point cloud to appear in left-right distortion and deformation in this range, and the planar position of the trajectory point needs to be corrected in the horizontal direction to make the change moderate and consistent with the real road straight line or gentle curve; and the posture jump is reflected in that the roll angle of a trajectory point suddenly increases or decreases, causing the point cloud data to exist in a scissors-like included angle with the actual point cloud data, and the posture of the trajectory point needs to be corrected to make the change moderate and consistent with the real road horizontal plane without deviation.

[0067] According to the above embodiment, the application can detect trajectory points with one or more abnormal types of height jump, horizontal jump and attitude jump, and the corrected and smoothed trajectory position information has higher relative accuracy, so as to ensure that the calculated point cloud is not deformed or mutated, improve the accuracy and accuracy of the vehicle-mounted laser scanning measurement system, and further improve the efficiency and quality of high-precision map vectorization.

[0068] In an optional embodiment, the position sequence information includes coordinate information and roll angle, and the multi-dimensional anomaly detection on the at least one to-be-detected trajectory point based on the position sequence information can include:

[0069] determining the first included angle based on the coordinate information of the at least one to-be-detected trajectory point, the coordinate information of the first adjacent trajectory point and the coordinate information of the second adjacent trajectory point;

[0070] determining the second included angle based on the coordinate information of the at least one to-be-detected trajectory point and the coordinate information of the first adjacent trajectory point;

[0071] determining the angle difference information based on the roll angle of the at least one to-be-detected trajectory point and the roll angle of the second adjacent trajectory point.

[0072] In a specific embodiment, the calculation formula of the first included angle can be

[0073]

[0074] wherein,

[0075]

[0076]

[0077] wherein, P n represents the nth to-be-detected trajectory point, P n+1 represents the n+1th to-be-detected trajectory point, P n+2 represents the n+2th to-be-detected trajectory point, ∠P n P n+1 P n+2 represents the first included angle corresponding to the n+1th to-be-detected trajectory point, P n (X n ,Y n ,Z n ) represents the coordinate information of the nth to-be-detected trajectory point, P n+1 (X n+1 ,Y n+1 ,Z n+1) represents the coordinate information of the n+1th to-be-detected trajectory point, and P n+2 (X n+2 ,Y n+2 ,Z n+2 ) represents the coordinate information of the n+2th to-be-detected trajectory point, and P Figure 2 .

[0078] In one specific embodiment, the calculation formula of the second included angle can be

[0079]

[0080] , P n represents the n th to-be-detected trajectory point, P n+1 represents the n+1th to-be-detected trajectory point, P n+2 represents the n+2th to-be-detected trajectory point, and ∠P n P n+1 represents the second included angle corresponding to the n+1th to-be-detected trajectory point, P n (X n ,Y n ,Z n ) represents the coordinate information of the n th to-be-detected trajectory point, P n+1 (X n+1 ,Y n+1 ,Z n+1 ) represents the coordinate information of the n+1th to-be-detected trajectory point, and P Figure 3 .

[0081] In one specific embodiment, the calculation formula of the angle difference information can be

[0082] ΔRoll = |Roll n+1 -Roll n |

[0083] , ΔRoll represents the angle difference information, Roll n represents the roll angle of the n th to-be-detected trajectory point, and Roll n+1 represents the roll angle of the n+1th to-be-detected trajectory point.

[0084] In the above embodiments, the position sequence information of the trajectory points is obtained, and the trajectory points are detected in multiple dimensions based on the position sequence information. Different abnormal types have corresponding abnormal detection methods, which improves the accuracy and comprehensiveness of the abnormal detection process. In addition, multiple abnormal detections of trajectory points can be performed simultaneously without affecting each other. Moreover, the multiple abnormal detection methods are simple and convenient to calculate, which can improve efficiency and rationality.

[0085] S105: determining a target abnormal trajectory point based on the abnormal detection result.

[0086] In one specific embodiment, the target abnormal trajectory point described above can indicate that at least one of the to-be-tested trajectory points has at least one type of abnormality in the abnormality detection result. Specifically, in the case where the first abnormality detection result of a certain to-be-tested trajectory point does not satisfy the first preset condition, the to-be-tested trajectory point is taken as the target abnormal trajectory point, and the to-be-tested trajectory point is the target abnormal trajectory point having the first abnormal type;

[0087] In the case where the second abnormality detection result of a certain to-be-tested trajectory point does not satisfy the second preset condition, the to-be-tested trajectory point is taken as the target abnormal trajectory point, and the to-be-tested trajectory point is the target abnormal trajectory point having the second abnormal type;

[0088] In the case where the third abnormality detection result of a certain to-be-tested trajectory point does not satisfy the third preset condition, the to-be-tested trajectory point is taken as the target abnormal trajectory point, and the to-be-tested trajectory point is the target abnormal trajectory point having the third abnormal type.

[0089] The first preset condition, the second preset condition and the third preset condition described above can be set according to actual application conditions. Specifically, the first preset condition, the second preset condition and the third preset condition can be set according to consideration of road-related standards, actual road conditions and vehicle mechanical characteristics. For example, the first preset condition is that the angle of the first included angle is in the range of [130°, 180°], the second preset condition is that the angle of the second included angle is in the range of [0°, 30°], and the third preset condition is that the angle difference information is in the range of [0°, 20°].

[0090] S107: Based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, performing position correction processing on the target abnormal trajectory point in the position sequence information to obtain target position sequence information.

[0091] In one optional embodiment, the position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point to obtain the target position sequence information can include:

[0092] In the case where the target abnormal type is the first abnormal type, based on the position information of the first adjacent trajectory point and the position information of the second adjacent trajectory point, the position information of the target abnormal trajectory point is corrected to the midpoint position corresponding to the first direction of the third connecting line to obtain the target position sequence information;

[0093] The third line is a line between the first adjacent track point and the second adjacent track point, the first adjacent track point is a previous track point among the adjacent track points corresponding to the target abnormal track point, and the second adjacent track point is a next track point among the adjacent track points corresponding to the target abnormal track point.

[0094] In one embodiment, P' n+1 (X n+1 ,Y n+1 ,Z n+1 ) represents coordinate information of a midpoint corresponding to the third line in the first direction, the third line being a line between the nth track point and the n+2th track point, wherein X n+1 =(X n +X n+2 ) / 2, Y n+1 =(Y n +Y n+2 ) / 2.

[0095] In one embodiment, the position correction processing of the target abnormal track point in the position sequence information based on the adjacent track point of the target abnormal track point and the target abnormal type corresponding to the target abnormal track point, to obtain the target position sequence information, can further include:

[0096] In the case where the target abnormal type is the second abnormal type, the position information of the target abnormal track point is corrected to a midpoint position corresponding to a fourth line in a second direction based on the position information of the first adjacent track point and the position information of the second adjacent track point, to obtain the target position sequence information.

[0097] The fourth line is a line between the first adjacent track point and the second adjacent track point, the first adjacent track point is a previous track point among the adjacent track points corresponding to the target abnormal track point, and the second adjacent track point is a next track point among the adjacent track points corresponding to the target abnormal track point.

[0098] In one embodiment, P' n+1 (X n+1 ,X n+1 ,Z n+1 ) represents coordinate information of a midpoint corresponding to the fourth line in the second direction, the fourth line being a line between the nth track point and the n+2th track point, wherein Z n+1 =(Z n +Z n+2 ) / 2, and specifically, the second direction is the elevation direction.

[0099] In an optional embodiment, the position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point to obtain the target position sequence information can further include:

[0100] In a case where the target abnormal type is a third abnormal type, the roll angle corresponding to the target abnormal trajectory point is corrected based on the position information of the second adjacent trajectory point to obtain the target position sequence information.

[0101] The second adjacent trajectory point is a later trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

[0102] In a specific embodiment, in a case where the roll angle of the second adjacent trajectory point is greater than the roll angle of the target abnormal trajectory point, the roll angle of the target abnormal trajectory point is increased by a first preset angle, and in a case where the roll angle of the second adjacent trajectory point is less than the roll angle of the target abnormal trajectory point, the roll angle of the target abnormal trajectory point is decreased by a second preset angle. The first preset angle and the second preset angle can be set according to actual application conditions.

[0103] In the above embodiments, in a case of different abnormal types, the corresponding correction processing is performed on the abnormal trajectory point, which can improve the accuracy of the correction processing.

[0104] In an optional embodiment, in a case where the target abnormal trajectory point includes a plurality of abnormal trajectory points, the position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point to obtain the target position sequence information can include:

[0105] Based on the time sequence order corresponding to the plurality of abnormal trajectory points, the plurality of abnormal trajectory points are traversed, and in a case where any abnormal trajectory point is traversed, the following process is performed:

[0106] The current position sequence information is obtained by performing the position correction processing on the current traversal trajectory point based on the adjacent trajectory point of the current traversal trajectory point and the target abnormal type corresponding to the current traversal trajectory point.

[0107] In a case where the plurality of abnormal trajectory points are traversed, the current position sequence information is taken as the target position sequence information.

[0108] In an optional embodiment, after the position correction processing is performed on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point, the target position sequence information is obtained, the method can further include:

[0109] generating target point cloud data based on the target position sequence information;

[0110] generating a map of the target region based on the target point cloud data.

[0111] In an optional embodiment, generating target point cloud data based on the target position sequence information can include: determining target pose information of the target lidar based on the target position sequence information and preset mapping information; and generating the target point cloud data based on the target pose information and time information.

[0112] In a specific embodiment, the target position sequence information can be target pose information of an inertial navigation system, the target pose information of the target lidar can include position information and rotation angle information of the target lidar, the time information can be a time length from emitting laser to receiving laser of the target lidar, the target point cloud data can be point cloud data of objects in the target region collected by the target lidar, and the preset mapping information can represent a corresponding relationship between the pose information of the inertial navigation system and the pose information of the lidar.

[0113] In an optional embodiment, before generating the target point cloud data based on the target position sequence information, the method can further include:

[0114] fitting at least one trajectory point based on the target position sequence information to obtain a fitting curve; and performing smoothing processing on the fitting curve to obtain updated position sequence information;

[0115] Correspondingly, the generating of the target point cloud data based on the target position sequence information can include:

[0116] generating the target point cloud data based on the updated position sequence information.

[0117] In a specific embodiment, the specific refinement of the generating of the target point cloud data based on the updated position sequence information can refer to the specific refinement of the generating of the target point cloud data based on the target position sequence information, which will not be repeated here.

[0118] In a specific embodiment, the generating of the map of the target region based on the target point cloud data can include: performing vectorization processing on the target point cloud data to obtain a vectorized map; and taking the vectorized map as the map of the target region.

[0119] In the above embodiments, the position sequence information updated after the correction processing is combined to generate point cloud data to generate a map, which can effectively improve the accuracy and generation efficiency of the generated map.

[0120] As can be seen from the technical solutions provided by the above embodiments of the present specification, in the present specification, the position sequence information of the trajectory points is acquired, and multi-dimensional anomaly detection is performed on the trajectory points based on the position sequence information, which can accurately and efficiently determine the abnormal trajectory points and their abnormal types. Based on the corresponding abnormal types and their adjacent trajectory points, the abnormal trajectory points are subjected to corresponding correction processing, which can accurately correct the abnormal trajectory points and improve the accuracy of the trajectory data. In addition, the various anomaly detections of the trajectory points can be performed simultaneously without affecting each other, and the various anomaly detections and correction processing procedures are simple and convenient to calculate, which can improve the efficiency and rationality of the anomaly detection and correction processing procedures. Furthermore, the position sequence information updated after the correction processing is combined to generate point cloud data to generate a map, which can effectively improve the accuracy and generation efficiency of the generated map.

[0121] Figure 4 is a block diagram of a position data correction processing device according to an exemplary embodiment. Referring to Figure 4 The device comprises:

[0122] A position sequence information acquisition module 410 is configured to acquire position sequence information of at least one to-be-tested trajectory point in a target area.

[0123] An anomaly detection result generation module 420 is configured to perform multi-dimensional anomaly detection on the at least one to-be-tested trajectory point based on the position sequence information to obtain an anomaly detection result of the at least one to-be-tested trajectory point, the anomaly detection result representing a probability that the at least one to-be-tested trajectory point belongs to a plurality of abnormal trajectory points.

[0124] A target abnormal trajectory point determination module 430 is configured to determine a target abnormal trajectory point based on the anomaly detection result.

[0125] A correction processing module 440 is configured to perform position correction processing on the target abnormal trajectory point in the position sequence information based on the adjacent trajectory points of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point to obtain target position sequence information.

[0126] Optionally, the anomaly detection result comprises at least two of a first anomaly detection result corresponding to a first abnormal type, a second anomaly detection result corresponding to a second abnormal type, and a third anomaly detection result corresponding to a third abnormal type.

[0127] The first abnormality detection result is a first included angle between a first line corresponding to the at least one to-be-tested trajectory point and a second line corresponding to the at least one to-be-tested trajectory point, the first line corresponding to any to-be-tested trajectory point can be a line between the any to-be-tested trajectory point and a first adjacent trajectory point corresponding to the any to-be-tested trajectory point, the first adjacent trajectory point is a previous trajectory point in the adjacent trajectory points corresponding to the any to-be-tested trajectory point, the second line corresponding to any to-be-tested trajectory point can be a line between the any to-be-tested trajectory point and a second adjacent trajectory point corresponding to the at least one to-be-tested trajectory point, the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the at least one to-be-tested trajectory point; the second abnormality detection result is a second included angle between the first line and a first direction; and the third abnormality detection result is angle difference information between a roll angle corresponding to the at least one to-be-tested trajectory point and a roll angle corresponding to the second adjacent trajectory point.

[0128] Optionally, the position sequence information includes coordinate information and a roll angle, and correspondingly, the abnormality detection result generation module 420 can include:

[0129] a first included angle determination unit, configured to determine the first included angle based on coordinate information of the at least one to-be-tested trajectory point, coordinate information of the first adjacent trajectory point, and coordinate information of the second adjacent trajectory point;

[0130] a second included angle determination unit, configured to determine the second included angle based on the coordinate information of the at least one to-be-tested trajectory point and the coordinate information of the first adjacent trajectory point;

[0131] an angle difference information determination unit, configured to determine the angle difference information based on a roll angle of the at least one to-be-tested trajectory point and a roll angle of the second adjacent trajectory point.

[0132] Optionally, the correction processing module 440 can include:

[0133] a first correction processing unit, configured to, in a case where the target abnormality type is a first abnormality type, correct, based on position information of the first adjacent trajectory point and position information of the second adjacent trajectory point, position information of the target abnormal trajectory point to a midpoint position corresponding to a third line in a first direction, to obtain the target position sequence information;

[0134] The third line is a line between the first adjacent trajectory point and the second adjacent trajectory point, the first adjacent trajectory point is a previous trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point, and the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

[0135] Optionally, the correction processing module 440 can further include:

[0136] a second correction processing unit, configured to, in a case where the target abnormal type is a second abnormal type, correct the position information of the target abnormal trajectory point to a midpoint position corresponding to the second direction of a fourth connecting line based on the position information of the first adjacent trajectory point and the position information of the second adjacent trajectory point, to obtain the target position sequence information.

[0137] The fourth connecting line is a connecting line between the first adjacent trajectory point and the second adjacent trajectory point, the first adjacent trajectory point is a previous trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point, and the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

[0138] Optionally, the correction processing module 440 can further include:

[0139] a third correction processing unit, configured to, in a case where the target abnormal type is a third abnormal type, correct a roll angle corresponding to the target abnormal trajectory point based on the position information of the second adjacent trajectory point, to obtain the target position sequence information.

[0140] The second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

[0141] Optionally, in a case where the target abnormal trajectory point includes a plurality of abnormal trajectory points, the correction processing module 440 can further include:

[0142] a traversal unit, configured to traverse the plurality of abnormal trajectory points based on a time sequence order corresponding to the plurality of abnormal trajectory points, and in a case where any abnormal trajectory point is traversed, perform the following process:

[0143] a fourth correction processing unit, configured to perform position correction processing on a currently traversed trajectory point based on an adjacent trajectory point of the currently traversed trajectory point and a target abnormal type corresponding to the currently traversed trajectory point, to obtain current position sequence information.

[0144] a position sequence information updating unit, configured to, in a case where the plurality of abnormal trajectory points are traversed, take the current position sequence information as the target position sequence information.

[0145] Optionally, after the correction processing module 440, the apparatus can further include:

[0146] a target point cloud data generation module, configured to generate target point cloud data based on the target position sequence information.

[0147] A map generation module is configured to generate a map of the target region based on the target point cloud data.

[0148] As to the apparatus in the above embodiments, the specific manners in which various modules perform operations have been described in detail in the embodiments of the method, and thus will not be described in detail here.

[0149] Figure 5 is a block diagram of an apparatus for position data correction processing according to an example embodiment. The apparatus can be a terminal, and its internal structure can be as shown in Figure 5 The apparatus includes a processor, a memory, a network interface, a display screen, and an input device connected through a system bus. The processor of the apparatus is configured to provide computing and control capabilities. The memory of the apparatus includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The network interface of the apparatus is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement a position data correction processing method. The display screen of the apparatus can be a liquid crystal display screen or an electronic ink display screen. The input device of the apparatus can be a touch layer overlaid on the display screen, or a key, trackball, or touchpad provided on the housing of the apparatus, or an external keyboard, touchpad, or mouse.

[0150] Figure 6 is a block diagram of an apparatus for position data correction processing according to an example embodiment. The apparatus can be a server, and its internal structure can be as shown in Figure 6 The apparatus includes a processor, a memory, and a network interface connected through a system bus. The processor of the apparatus is configured to provide computing and control capabilities. The memory of the apparatus includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The network interface of the apparatus is configured to communicate with external terminals through a network connection. The computer program is executed by the processor to implement a position data correction processing method.

[0151] Those skilled in the art can understand that Figure 5 or Figure 6 The structure shown in the above embodiments is only a block diagram of part of the structure related to the disclosed solution, and does not constitute a limitation on the apparatus to which the disclosed solution is applied. A specific apparatus can include more or fewer components than those shown in the diagram, or combine certain components, or have a different arrangement of components.

[0152] In an example embodiment, there is also provided an apparatus comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute the instructions to implement the method for position data correction processing as disclosed in the embodiments.

[0153] In an example embodiment, there is also provided a computer readable storage medium storing instructions which, when executed by a processor of an electronic device, cause the electronic device to perform the method for position data correction processing as disclosed in the embodiments.

[0154] In an example embodiment, there is also provided a computer program product comprising instructions which, when executed on a computer, cause the computer to perform the method for position data correction processing as disclosed in the embodiments.

[0155] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. Any reference to memory, storage, databases, or other media in the embodiments disclosed herein can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM), or external cache memory. As an illustration but not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0156] Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the present application cover any and all variations of the application disclosed herein including modifications and adaptations thereof by those skilled in the art. It is intended to include the present application as can be claimed with respect to any and all combinations or subcombinations of the elements and / or features of the application. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application indicated by the following claims.

[0157] It should be understood that the present disclosure is not limited to the precise construction that has been described above and shown in the accompanying drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the present disclosure. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method of correction processing of position data, characterized by, The method comprises the following steps: acquiring position sequence information of at least one to-be-tested trajectory point in a target area; performing multi-dimensional anomaly detection on the at least one to-be-tested trajectory point based on the position sequence information, to obtain an anomaly detection result of the at least one to-be-tested trajectory point, wherein the anomaly detection result represents an anomaly type to which the at least one to-be-tested trajectory point belongs, and the anomaly detection result comprises at least two of a first anomaly detection result corresponding to a first anomaly type, a second anomaly detection result corresponding to a second anomaly type, and a third anomaly detection result corresponding to a third anomaly type, wherein the first anomaly detection result is a first included angle between a first line corresponding to the at least one to-be-tested trajectory point and a second line corresponding to the at least one to-be-tested trajectory point, the first line corresponding to any to-be-tested trajectory point can be a line segment between the any to-be-tested trajectory point and a first adjacent trajectory point corresponding to the any to-be-tested trajectory point, the first adjacent trajectory point is a previous trajectory point in adjacent trajectory points corresponding to the any to-be-tested trajectory point, the second line corresponding to any to-be-tested trajectory point can be a line segment between the any to-be-tested trajectory point and a second adjacent trajectory point corresponding to the at least one to-be-tested trajectory point, the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the at least one to-be-tested trajectory point, the second anomaly detection result is a second included angle between the first line and a first direction, and the third anomaly detection result is angle difference information between a roll angle corresponding to the at least one to-be-tested trajectory point and a roll angle corresponding to the second adjacent trajectory point, and the position sequence information comprises coordinate information and the roll angle; determining a target abnormal trajectory point based on the anomaly detection result; performing position correction processing on the target abnormal trajectory point in the position sequence information based on adjacent trajectory points of the target abnormal trajectory point and a target anomaly type corresponding to the target abnormal trajectory point, to obtain target position sequence information. The method further comprises the following steps:

2. The method of claim 1, wherein determining the first included angle based on coordinate information of the at least one to-be-tested trajectory point, coordinate information of the first adjacent trajectory point, and coordinate information of the second adjacent trajectory point; determining the second included angle based on the coordinate information of the at least one to-be-tested trajectory point and the coordinate information of the first adjacent trajectory point; and determining the angle difference information based on the roll angle of the at least one to-be-tested trajectory point and the roll angle of the second adjacent trajectory point. The method further comprises the following steps: in a case where the target anomaly type is the first anomaly type, correcting position information of the target abnormal trajectory point to a midpoint position corresponding to the third line in the first direction based on position information of the first adjacent trajectory point and position information of the second adjacent trajectory point, to obtain the target position sequence information. The third connecting line is a connecting line between the first adjacent trajectory point and the second adjacent trajectory point, the first adjacent trajectory point is a previous trajectory point in adjacent trajectory points corresponding to the target abnormal trajectory point, and the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

3. The method of claim 1, wherein The position correction processing of the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point obtains target position sequence information. In a case where the target abnormal type is a second abnormal type, the position information of the target abnormal trajectory point is corrected to a midpoint position corresponding to a fourth connecting line in a second direction based on position information of a first adjacent trajectory point and position information of a second adjacent trajectory point, to obtain the target position sequence information. The fourth connecting line is a connecting line between the first adjacent trajectory point and the second adjacent trajectory point, the first adjacent trajectory point is a previous trajectory point in adjacent trajectory points corresponding to the target abnormal trajectory point, and the second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the target abnormal trajectory point.

4. The method of claim 1, wherein The position correction processing of the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point obtains target position sequence information. In a case where the target abnormal type is a third abnormal type, a roll angle corresponding to the target abnormal trajectory point is corrected based on position information of a second adjacent trajectory point, to obtain the target position sequence information. The second adjacent trajectory point is a subsequent trajectory point in adjacent trajectory points corresponding to the target abnormal trajectory point.

5. The method of claim 1, wherein In a case where the target abnormal trajectory point includes a plurality of abnormal trajectory points, the position correction processing of the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point obtains target position sequence information. Based on a time sequence order corresponding to the plurality of abnormal trajectory points, the plurality of abnormal trajectory points are traversed, and in a case where any abnormal trajectory point is traversed, the following process is performed: The current position sequence information is obtained by performing position correction processing on the current traversal trajectory point based on an adjacent trajectory point of the current traversal trajectory point and a target abnormal type corresponding to the current traversal trajectory point. In a case where the plurality of abnormal trajectory points are traversed, the current position sequence information is taken as the target position sequence information.

6. The method of claim 1, wherein After the position correction processing of the target abnormal trajectory point in the position sequence information based on the adjacent trajectory point of the target abnormal trajectory point and the target abnormal type corresponding to the target abnormal trajectory point obtains target position sequence information, the method further includes: Target point cloud data is generated based on the target position sequence information. A map of the target area is generated based on the target point cloud data.

7. A position data correction processing apparatus characterized by comprising: includes: The position sequence information acquisition module is configured to acquire position sequence information of at least one to-be-tested trajectory point in the target area. The anomaly detection result generation module is configured to perform multi-dimensional anomaly detection on the at least one to-be-tested trajectory point based on the position sequence information, and obtain an anomaly detection result of the at least one to-be-tested trajectory point. The anomaly detection result represents a probability that the at least one to-be-tested trajectory point belongs to a plurality of abnormal trajectory points. The anomaly detection result includes at least two of a first anomaly detection result corresponding to a first abnormal type, a second anomaly detection result corresponding to a second abnormal type, and a third anomaly detection result corresponding to a third abnormal type. The first anomaly detection result is a first included angle between a first line corresponding to the at least one to-be-tested trajectory point and a second line corresponding to the at least one to-be-tested trajectory point. The first line corresponding to any to-be-tested trajectory point can be a line between the any to-be-tested trajectory point and a first adjacent trajectory point corresponding to the any to-be-tested trajectory point. The first adjacent trajectory point is a previous trajectory point in adjacent trajectory points corresponding to the any to-be-tested trajectory point. The second line corresponding to any to-be-tested trajectory point can be a line between the any to-be-tested trajectory point and a second adjacent trajectory point corresponding to the at least one to-be-tested trajectory point. The second adjacent trajectory point is a subsequent trajectory point in the adjacent trajectory points corresponding to the at least one to-be-tested trajectory point. The second anomaly detection result is a second included angle between the first line and a first direction. The third anomaly detection result is angle difference information between a roll angle corresponding to the at least one to-be-tested trajectory point and a roll angle corresponding to the second adjacent trajectory point. The position sequence information includes coordinate information and the roll angle. The target abnormal trajectory point determination module is configured to determine a target abnormal trajectory point based on the anomaly detection result. The correction processing module is configured to perform position correction processing on the target abnormal trajectory point in the position sequence information based on adjacent trajectory points of the target abnormal trajectory point and a target abnormal type corresponding to the target abnormal trajectory point, and obtain target position sequence information. The anomaly detection result generation module includes a first included angle determination unit configured to determine the first included angle based on coordinate information of the at least one to-be-tested trajectory point, coordinate information of the first adjacent trajectory point, and coordinate information of the second adjacent trajectory point; a second included angle determination unit configured to determine the second included angle based on the coordinate information of the at least one to-be-tested trajectory point and the coordinate information of the first adjacent trajectory point; and an angle difference information determination unit configured to determine the angle difference information based on a roll angle of the at least one to-be-tested trajectory point and a roll angle of the second adjacent trajectory point.

8. An apparatus for position data correction processing, characterized by comprising: The processor; The memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the position data correction processing method according to any one of claims 1 to 6. ​ 9. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is enabled to perform the position data correction processing method according to any one of claims 1 to 6.

Images