Abnormal recognition method and device for laser SLAM positioning, electronic equipment and computer program product
By acquiring and analyzing point cloud matching feature information and multiple positioning data of autonomous driving vehicles, combining errors and variances, identifying laser SLAM positioning abnormalities, solving the problem of accuracy reduction in laser SLAM positioning in open or fewer features scenes, and improving positioning Accuracy and stability.
Patent Information
- Application Number
- CN202411326614.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2024-11-12
AI Technical Summary
Laser SLAM positioning is prone to degradation problems in open or few scenarios, resulting in high matching reliability of point cloud matching algorithm output but large actual positioning deviation. Directly performing subsequent fusion positioning based on the confidence output of the algorithm will lead to a decrease in accuracy.
By obtaining point cloud matching feature information and a variety of positioning data of autonomous driving vehicles, the effectiveness of the positioning data is determined, and combined with the error and variance of point cloud matching feature information and positioning data, abnormal identification of laser SLAM positioning is adopted, and a variety of methods are used to identify the positioning data. Redundant judgment strategies improve identification accuracy.
The online recognition of laser SLAM positioning abnormalities is realized, which improves the accuracy and applicability of abnormal recognition. It is suitable for various positioning scenarios. The confidence of laser positioning data is dynamically adjusted to improve the positioning accuracy and stability of autonomous driving vehicles.
Smart Images

Figure CN118938251A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of laser positioning anomaly recognition, and in particular to a laser SLAM positioning anomaly recognition method, device, electronic equipment, and computer program product. Background Art
[0002] Laser SLAM (Simultaneous Localization and Mapping) is a technology that uses laser sensors to enable robots or autonomous systems to simultaneously locate and build maps in unknown environments. In the field of autonomous driving, laser SLAM positioning technology plays a vital role and is used in conjunction with a variety of sensors to improve the accuracy and robustness of autonomous driving systems.
[0003] Laser SLAM may degrade in some open or feature-less scenes, or when the environment changes, the point cloud matching algorithm may output a high matching confidence, but the actual positioning deviation may be large. Therefore, if subsequent fusion positioning is performed directly based on the confidence output by the algorithm, the accuracy of the fusion positioning will decrease. Summary of the invention
[0004] The embodiments of the present application provide a method, device, electronic device, and computer program product for abnormality recognition in laser SLAM positioning to improve the accuracy of abnormality recognition in laser SLAM positioning.
[0005] The present application embodiment adopts the following technical solutions:
[0006] In a first aspect, an embodiment of the present application provides a method for identifying anomalies in laser SLAM positioning, wherein the method for identifying anomalies in laser SLAM positioning includes:
[0007] Obtain point cloud matching feature information of the autonomous driving vehicle;
[0008] Acquire multiple positioning data of the autonomous vehicle and determine validity of the multiple positioning data, wherein the multiple positioning data are output by multiple positioning sources on the autonomous vehicle;
[0009] According to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data, the laser SLAM positioning is identified as abnormal, and an abnormal identification result of the laser SLAM positioning is obtained.
[0010] Optionally, the point cloud matching feature information includes eigenvalues of an optimized information matrix obtained based on a point cloud matching algorithm, and the abnormality recognition of laser SLAM positioning is performed based on the point cloud matching feature information and the validity of the multiple positioning data, and the abnormality recognition result of laser SLAM positioning is obtained, including:
[0011] If the eigenvalue of the optimized information matrix is less than the preset eigenvalue threshold, the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state;
[0012] Otherwise, according to the various positioning data and the validity of the positioning data, the laser SLAM positioning is identified as abnormal, and the abnormal identification result of the laser SLAM positioning is obtained.
[0013] Optionally, the performing abnormality identification on the laser SLAM positioning according to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data to obtain the abnormality identification result of the laser SLAM positioning includes:
[0014] When the abnormal recognition result of the laser SLAM positioning cannot be determined as an abnormal state according to the point cloud matching feature information, obtaining a plurality of positioning data within a preset time period of the cache;
[0015] Calculate the positioning error and positioning variance between the various positioning data according to the various positioning data within a preset time period;
[0016] According to the positioning errors and positioning variances between various positioning data and the validity of various positioning data, the laser SLAM positioning is identified abnormally to obtain the abnormal identification result of the laser SLAM positioning.
[0017] Optionally, the positioning data includes RTK positioning data, laser positioning data, visual positioning data and fused positioning data, and the laser SLAM positioning is identified based on the positioning errors and positioning variances between the multiple positioning data and the validity of the multiple positioning data, and the abnormal identification result of the laser SLAM positioning is obtained, including:
[0018] If both the RTK positioning data and the visual positioning data are in valid states, then according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fusion positioning data, and the positioning variance between the visual positioning data and the RTK positioning data, abnormal identification is performed on the laser SLAM positioning to obtain an abnormal identification result of the laser SLAM positioning;
[0019] If the RTK positioning data is in a valid state and the visual positioning data is in an invalid state, then according to the continuous count of the RTK positioning data being in a valid state, the positioning error between the RTK positioning data and the fused positioning data, and the positioning variance between the laser positioning data and the fused positioning data, the laser SLAM positioning is identified as abnormal, and an abnormal identification result of the laser SLAM positioning is obtained;
[0020] If both the RTK positioning data and the visual positioning data are in an invalid state, the laser SLAM positioning is identified as abnormal based on the positioning variance and positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the initial positioning reliability of the laser positioning data to obtain an abnormal identification result of the laser SLAM positioning.
[0021] Optionally, the performing abnormality identification on the laser SLAM positioning according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fusion positioning data, and the positioning variance between the visual positioning data and the RTK positioning data comprises:
[0022] Determine whether a positioning error between the visual positioning data and the RTK positioning data is less than a positioning error between the laser positioning data and the visual positioning data, or whether a positioning error between the visual positioning data and the RTK positioning data is less than a first preset error threshold and a positioning error between the laser positioning data and the visual positioning data is less than a second preset error threshold;
[0023] If the positioning error between the visual positioning data and the RTK positioning data is smaller than the positioning error between the laser positioning data and the visual positioning data, and / or the positioning error between the visual positioning data and the RTK positioning data is smaller than a first preset error threshold and the positioning error between the laser positioning data and the visual positioning data is smaller than a second preset error threshold, then according to the positioning variance between the laser positioning data and the fused positioning data and the positioning variance between the visual positioning data and the RTK positioning data, abnormality identification is performed on the laser SLAM positioning;
[0024] If the positioning error between the visual positioning data and the RTK positioning data is not less than the positioning error between the laser positioning data and the visual positioning data, and the positioning error between the visual positioning data and the RTK positioning data is not less than a first preset error threshold, and the positioning error between the laser positioning data and the visual positioning data is not less than a second preset error threshold, then the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state.
[0025] Optionally, the laser SLAM positioning is identified as abnormal according to the continuous count of the RTK positioning data being in a valid state, the positioning error between the RTK positioning data and the fused positioning data, and the positioning variance between the laser positioning data and the fused positioning data, and the abnormal identification result of the laser SLAM positioning is obtained, which includes:
[0026] If the continuous count of the RTK positioning data being in a valid state is greater than a first preset number threshold, and the positioning error between the RTK positioning data and the fused positioning data is less than a third preset error threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than a first preset variance threshold, then an abnormal count is performed on the laser SLAM positioning;
[0027] If the cumulative abnormal count of the laser SLAM positioning reaches a second preset number threshold, or the positioning variance between the laser positioning data and the fused positioning data is greater than a second preset variance threshold, then the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state, wherein the second preset variance threshold is greater than the first preset variance threshold.
[0028] Optionally, the laser SLAM positioning is subjected to abnormal identification according to the positioning variance and positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the initial positioning reliability of the laser positioning data, and the abnormal identification result of the laser SLAM positioning is obtained, including:
[0029] If the positioning variance between the RTK positioning data and the fused positioning data is less than the third preset variance threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than the fourth preset variance threshold, and the initial positioning confidence of the laser positioning data is in the preset confidence interval, and the positioning error between the RTK positioning data and the fused positioning data is less than the fourth preset error threshold, then the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state.
[0030] Optionally, after performing abnormality identification on the laser SLAM positioning according to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data to obtain the abnormality identification result of the laser SLAM positioning, the abnormality identification method of the laser SLAM positioning further includes:
[0031] If the abnormal recognition result of the laser SLAM positioning is an abnormal state, the initial positioning confidence of the laser positioning data is reduced;
[0032] Otherwise, the initial positioning confidence of the laser positioning data is retained.
[0033] In a second aspect, an embodiment of the present application further provides an abnormality recognition device for laser SLAM positioning, wherein the abnormality recognition device for laser SLAM positioning includes:
[0034] A first acquisition unit, used to acquire point cloud matching feature information of the autonomous driving vehicle;
[0035] A second acquisition unit, configured to acquire a plurality of positioning data of the autonomous driving vehicle and determine validity of the plurality of positioning data, wherein the plurality of positioning data are output by a plurality of positioning sources on the autonomous driving vehicle;
[0036] The identification unit is used to identify abnormalities in laser SLAM positioning according to the point cloud matching feature information, multiple positioning data and the validity of the multiple positioning data, so as to obtain an abnormality identification result of the laser SLAM positioning.
[0037] In a third aspect, an embodiment of the present application further provides an electronic device, including:
[0038] Processor; and
[0039] A memory arranged to store computer executable instructions which, when executed, cause the processor to perform any of the methods described above.
[0040] In a fourth aspect, an embodiment of the present application further provides a computer program product, comprising a computer program or instructions, wherein the computer program or instructions implement any of the aforementioned methods when executed by a processor.
[0041] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: the abnormal identification method of laser SLAM positioning in the embodiments of the present application first obtains the point cloud matching feature information of the autonomous driving vehicle; then obtains multiple positioning data of the autonomous driving vehicle and determines the validity of the multiple positioning data, and the multiple positioning data are output by multiple positioning sources on the autonomous driving vehicle; finally, according to the point cloud matching feature information, multiple positioning data and the validity of multiple positioning data, the laser SLAM positioning is abnormally identified to obtain the abnormal identification result of the laser SLAM positioning. The abnormal identification method of laser SLAM positioning in the embodiments of the present application can perform online identification of the abnormal situation of laser SLAM positioning through the validity of the positioning data of the point cloud matching feature information and multiple positioning data sources, and improves the accuracy of abnormal identification through multiple redundant judgment strategies, and can be applied to the abnormal identification of laser SLAM positioning in various positioning scenarios. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0043] Figure 1 A schematic diagram of a process flow of an abnormality recognition method for laser SLAM positioning in an embodiment of the present application;
[0044] Figure 2This is a schematic diagram of the structure of an abnormality recognition device for laser SLAM positioning in an embodiment of the present application;
[0045] Figure 3 This is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION
[0046] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.
[0047] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.
[0048] The present application embodiment provides a method for identifying abnormalities in laser SLAM positioning, such as Figure 1 As shown, a schematic flow chart of an abnormality recognition method for laser SLAM positioning in an embodiment of the present application is provided, and the abnormality recognition method for laser SLAM positioning at least includes the following steps S110 to S130:
[0049] Step S110, obtaining point cloud matching feature information of the autonomous driving vehicle.
[0050] When performing anomaly recognition of laser SLAM positioning, the embodiment of the present application needs to first obtain the current point cloud matching feature information of the autonomous driving vehicle. The point cloud matching feature information reflects the number of environmental features currently matched. The point cloud matching feature information can be used as a preliminary screening condition for anomaly recognition of laser SLAM positioning.
[0051] Step S120, obtaining multiple positioning data of the autonomous driving vehicle and determining the validity of the multiple positioning data, wherein the multiple positioning data are output by multiple positioning sources on the autonomous driving vehicle.
[0052] The embodiments of the present application also need to acquire multiple positioning data generated on the autonomous driving vehicle in real time, such as RTK positioning data output by the RTK positioning device, laser positioning data output by the laser SLAM positioning module, visual positioning data output by the visual positioning module, and fusion positioning information output by the fusion positioning module. Since the frequencies of the positioning data output by different positioning sources are different, the time synchronization method in the prior art can be adopted to perform time synchronization processing on the multiple positioning data generated by multiple positioning sources to ensure the accuracy of subsequent calculations.
[0053] In addition to obtaining positioning data, it is also necessary to obtain the validity of each positioning data, which can be directly output by the positioning source. For example, the RTK positioning device can output the positioning status of the RTK positioning data. The differential state indicates that the accuracy of the RTK positioning data is high and the positioning data is valid. The validity reflects whether each positioning data is credible, and then different anomaly identification strategies are adopted according to the validity of each positioning data.
[0054] Step S130, performing abnormality recognition on the laser SLAM positioning according to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data, to obtain an abnormality recognition result of the laser SLAM positioning.
[0055] Combining the point cloud matching feature information obtained in the previous steps, multiple positioning data and the validity of multiple positioning data, different redundant recognition strategies are adopted to identify abnormalities in laser SLAM positioning, so as to accurately determine whether the current laser positioning data is abnormal.
[0056] The laser SLAM positioning anomaly recognition method of the embodiment of the present application can perform online identification of laser SLAM positioning anomalies through point cloud matching feature information and the validity of positioning data from multiple positioning data sources. It improves the accuracy of anomaly recognition through multiple redundant judgment strategies and can be applied to anomaly recognition of laser SLAM positioning in various positioning scenarios.
[0057] In some embodiments of the present application, the point cloud matching feature information includes the eigenvalue of an optimized information matrix obtained based on a point cloud matching algorithm, and the laser SLAM positioning is identified as abnormal based on the point cloud matching feature information and the validity of the multiple positioning data to obtain the abnormal identification result of the laser SLAM positioning, including: if the eigenvalue of the optimized information matrix is less than a preset eigenvalue threshold, the abnormal identification result of the laser SLAM positioning is determined to be an abnormal state; otherwise, the laser SLAM positioning is identified as abnormal based on the multiple positioning data and the validity of the positioning data to obtain the abnormal identification result of the laser SLAM positioning.
[0058] The point cloud matching feature information of the embodiment of the present application mainly includes the eigenvalues of the optimized information matrix obtained based on the point cloud matching algorithm. For example, the point cloud matching algorithm such as NDT or ICP can be used to match and optimize the point cloud data collected by the laser radar, so as to obtain the optimized information matrix, where the information matrix may include the Hessian matrix and the Jacobian matrix, etc. The eigenvalues of the optimized information matrix are calculated. For example, the idea of the LOAM algorithm can be used to call functions such as matrix addition and multiplication processing to calculate the eigenvalues, so as to obtain the eigenvalues of the optimized information matrix, which reflects the number of point cloud features in the current environment.
[0059] Compare the eigenvalue of the optimized information matrix with the set eigenvalue threshold. If the eigenvalue of the optimized information matrix is less than the set eigenvalue threshold, it means that there is little feature information in the current environment, and it can be considered that degradation has occurred. Therefore, the laser SLAM positioning can be directly determined to be abnormal. If it is not less than the set eigenvalue threshold, it only means that there is a lot of feature information in the current environment, but it does not mean that the laser SLAM positioning is necessarily normal, because it will also involve the influence of factors such as environmental changes. Therefore, it is possible to further combine multiple positioning data and the effectiveness of positioning data to identify abnormalities in laser SLAM positioning.
[0060] In some embodiments of the present application, the laser SLAM positioning is identified as abnormal based on the point cloud matching feature information, multiple positioning data, and the validity of the multiple positioning data to obtain the abnormal identification result of the laser SLAM positioning, including: when it cannot be determined that the abnormal identification result of the laser SLAM positioning is an abnormal state according to the point cloud matching feature information, obtaining multiple positioning data within a preset time period of the cache; calculating the positioning error and positioning variance between the multiple positioning data according to the multiple positioning data within the preset time period; and the laser SLAM positioning is identified as abnormal based on the positioning error and positioning variance between the multiple positioning data and the validity of the multiple positioning data to obtain the abnormal identification result of the laser SLAM positioning.
[0061] When performing anomaly identification of laser SLAM positioning, the embodiment of the present application, if the laser SLAM positioning anomaly cannot be directly determined based on the point cloud matching feature information, the positioning data generated by multiple positioning sources can be further used for redundant judgment. In order to ensure the accuracy and reliability of the judgment, multiple positioning data within a certain period of time, such as 3 seconds, can be cached in real time and time synchronization processing can be performed. Then, the positioning error and positioning variance between the multiple positioning data are calculated. The positioning error can directly reflect the deviation between the two positioning data, and the positioning variance can reflect the consistency of the positioning error change trend of the two positioning data.
[0062] Based on the calculation results of the positioning error and positioning variance between the above-mentioned multiple positioning data and the validity of multiple positioning data, different recognition strategies are adopted to perform anomaly recognition of laser SLAM positioning and obtain the anomaly recognition results of laser SLAM positioning.
[0063] In some embodiments of the present application, the positioning data includes RTK positioning data, laser positioning data, visual positioning data and fused positioning data, and the laser SLAM positioning is identified as abnormal according to the positioning errors and positioning variances between the multiple positioning data and the validity of the multiple positioning data, and the abnormal identification result of the laser SLAM positioning is obtained, including: if the RTK positioning data and the visual positioning data are both valid, then according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the positioning variance between the visual positioning data and the RTK positioning data, the laser SLAM positioning is identified as abnormal to obtain the laser SLAM positioning. Abnormal identification result; if the RTK positioning data is in a valid state and the visual positioning data is in an invalid state, then according to the continuous count of the RTK positioning data in a valid state, the positioning error between the RTK positioning data and the fused positioning data, and the positioning variance between the laser positioning data and the fused positioning data, the laser SLAM positioning is identified as abnormal to obtain the abnormal identification result of laser SLAM positioning; if both the RTK positioning data and the visual positioning data are in an invalid state, then according to the positioning variance and positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the initial positioning reliability of the laser positioning data, the laser SLAM positioning is identified as abnormal to obtain the abnormal identification result of laser SLAM positioning.
[0064] The positioning data of the embodiment of the present application mainly includes RTK positioning data, laser positioning data, visual positioning data and fused positioning data. The fused positioning data refers to the result obtained after fusing multiple positioning data such as RTK positioning data and visual positioning data. Since fusion has not been performed at the current moment, the fused positioning result of the previous frame can be obtained here. When performing laser SLAM positioning anomaly identification, the embodiment of the present application mainly includes the following situations:
[0065] 1) Both RTK positioning data and visual positioning data are valid: that is, the positioning state of RTK positioning data is in differential state, and the positioning confidence output by the visual positioning module is also higher than the threshold requirement, then both RTK positioning data and visual positioning data can be used to identify abnormal laser SLAM positioning. Specifically, the abnormal identification judgment conditions in this case can be constructed based on the positioning error between visual positioning data and RTK positioning data, the positioning error between laser positioning data and visual positioning data, the positioning variance between laser positioning data and fusion positioning data, and the positioning variance between visual positioning data and RTK positioning data;
[0066] 2) RTK positioning data is valid and visual positioning data is invalid: that is, the positioning state of the RTK positioning data is in a differential state, but the positioning confidence output by the visual positioning module does not meet the threshold requirement. In this case, the visual positioning data can be ignored, and abnormal identification can be performed based on the RTK positioning data and the fused positioning data. Specifically, the abnormal identification judgment conditions in this case can be constructed based on the continuous count of the RTK positioning data in a valid state, the positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and other information;
[0067] 3) Both RTK positioning data and visual positioning data are invalid: the positioning state of the RTK positioning data is a non-differential state, and the positioning reliability output by the visual positioning module does not meet the threshold requirement. In this extreme case, both the RTK positioning data and the visual positioning data are unreliable. In order to ensure the stability of the fusion positioning, relatively stricter judgment conditions can be set. Only when such strict judgment conditions are met can the laser SLAM positioning be considered abnormal. Specifically, the abnormal recognition judgment conditions in this case can be constructed based on the positioning variance and positioning error between the RTK positioning data and the fusion positioning data, the positioning variance between the laser positioning data and the fusion positioning data, and the initial positioning reliability of the laser positioning data.
[0068] In some embodiments of the present application, the abnormal recognition of the laser SLAM positioning according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fusion positioning data, and the positioning variance between the visual positioning data and the RTK positioning data includes: determining whether the positioning error between the visual positioning data and the RTK positioning data is less than the positioning error between the laser positioning data and the visual positioning data, or the positioning error between the visual positioning data and the RTK positioning data is less than a first preset error threshold and the positioning error between the laser positioning data and the visual positioning data is less than a second preset error threshold; if the positioning error between the visual positioning data and the RTK positioning data is less than the positioning error between the laser positioning data and the visual positioning data, and / or the positioning error between the visual positioning data and the RTK positioning data is less than a first preset error threshold and the positioning error between the laser positioning data and the visual positioning data is less than a second preset error threshold, then the abnormal recognition of the laser SLAM positioning is performed according to the positioning variance between the laser positioning data and the fusion positioning data and the positioning variance between the visual positioning data and the RTK positioning data; if the positioning error between the visual positioning data and the RTK positioning data is not less than the positioning error between the laser positioning data and the visual positioning data, and the positioning error between the visual positioning data and the RTK positioning data is not less than a first preset error threshold, and the positioning error between the laser positioning data and the visual positioning data is not less than a second preset error threshold, then it is determined that the abnormal recognition result of the laser SLAM positioning is an abnormal state.
[0069] In the case where both the RTK positioning data and the visual positioning data are valid, the abnormal recognition judgment condition constructed in the embodiments of the present application can be expressed in the following form:
[0070] dp_val_rtk - d0 < dp_val_slam, (1)
[0071] dp_val_rtk < d1 and dp_val_slam < d2, (2)
[0072] Among them, dp_val_rtk represents the positioning error between the visual positioning data and the RTK positioning data, dp_val_slam represents the positioning error between the visual positioning data and the laser positioning data, d0 is an empirical value representing the error introduced by the fluctuation of the visual positioning algorithm itself, d1 is the first preset error threshold, and d2 is the second preset error threshold.
[0073] When at least one of the above conditions (1) and (2) is met, it means that the current laser positioning data meets the judgment conditions set based on the positioning error between the positioning data, and the abnormality cannot be directly identified. Therefore, it is necessary to further combine other conditions for judgment, that is, the following condition (3).
[0074] When both conditions (1) and (2) above are not met, it means that the error between the laser positioning data and the visual positioning data is large, while the visual positioning data and RTK positioning data are relatively accurate. Therefore, in this case, the laser SLAM positioning can be directly considered abnormal.
[0075] When at least one of the above conditions (1) and (2) is met, further judgment is made through the following condition (3):
[0076] slam_loc_var>v1 and val_rtk_var<v2,v1> v2, (3)
[0077] Among them, slam_loc_var represents the positioning variance between laser positioning data and fusion positioning data, val_rtk_var represents the positioning variance between visual positioning data and RTK positioning data, and v1 and v2 are different variance thresholds.
[0078] On the basis that at least one of the above conditions (1) and (2) is met, if condition (3) is further met, it means that the error between the visual positioning data and the RTK positioning data is small, and the changing trend of the positioning error is also consistent, while the changing trend of the positioning error between the laser positioning data and the fusion positioning data is inconsistent. Since the fusion positioning data combines the visual positioning data and the RTK positioning data, it means that the current laser positioning data deviates from other normal positioning data. Therefore, the laser SLAM positioning can be considered abnormal at this time.
[0079] It should be noted that in actual applications, the positioning errors and positioning variances between the specific positioning data sources are used as the basis for identifying laser positioning data anomalies. Technical personnel in this field can flexibly adjust according to actual needs. They are not listed here one by one. The key is to improve the accuracy of laser SLAM positioning anomaly identification through mutual constraint verification between multiple positioning data.
[0080] In some embodiments of the present application, the abnormal recognition of the laser SLAM positioning is performed based on the continuous count of the RTK positioning data being in a valid state, the positioning error between the RTK positioning data and the fused positioning data, and the positioning variance between the laser positioning data and the fused positioning data. The obtained abnormal recognition result of the laser SLAM positioning includes: if the continuous count of the RTK positioning data being in a valid state is greater than a first preset number threshold, and the positioning error between the RTK positioning data and the fused positioning data is less than a third preset error threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than a first preset variance threshold, then an abnormal count is performed on the laser SLAM positioning once; if the cumulative abnormal count of the laser SLAM positioning reaches a second preset number threshold, or the positioning variance between the laser positioning data and the fused positioning data is greater than a second preset variance threshold, then it is determined that the abnormal recognition result of the laser SLAM positioning is an abnormal state, where the second preset variance threshold is greater than the first preset variance threshold.
[0081] In the case where the RTK positioning data is valid and the visual positioning data is invalid, the abnormal recognition judgment condition constructed in the embodiments of the present application can be expressed in the following form:
[0082] sensors_true_cnt[0]>n1, and dp_rtk_loc<d3, and slam_loc_var>v3, (4)
[0083] where sensors_true_cnt represents the count of the RTK positioning data being continuously normal, dp_rtk_loc represents the positioning error between the RTK positioning data and the fused positioning data, slam_loc_var represents the positioning variance between the laser positioning data and the fused positioning data, n1 is the first preset number threshold, d3 is the third preset error threshold, and v3 is the first preset variance threshold.
[0084] If condition (4) is satisfied, it indicates that the positioning state of the RTK is stable and the positioning error is also within an acceptable range. However, at this time, the positioning variance between the laser positioning data and the fused positioning data exceeds a certain threshold, indicating that the change in the positioning error of the laser positioning data may be unstable. At this time, an abnormal count of the laser SLAM positioning can be performed first, slam_42_error_cnt++, and it is determined whether the laser SLAM positioning is really abnormal through continuous judgment.
[0085] slam_42_error_cnt>n2, or slam_loc_var>v4, where v4>v3, (5)
[0086] Based on condition (5), if the continuous count of abnormal laser SLAM positioning exceeds a certain threshold, or the positioning variance between the laser positioning data and the fused positioning data is significantly greater than the acceptable upper limit, it can be determined that the laser SLAM positioning is abnormal.
[0087] In some embodiments of the present application, the abnormal recognition of laser SLAM positioning based on the positioning variance and positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the initial positioning confidence of the laser positioning data, to obtain the abnormal recognition result of laser SLAM positioning includes: if the positioning variance between the RTK positioning data and the fused positioning data is less than the third preset variance threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than the fourth preset variance threshold, and the initial positioning confidence of the laser positioning data is within the preset confidence interval, and the positioning error between the RTK positioning data and the fused positioning data is less than the fourth preset error threshold, then it is determined that the abnormal recognition result of the laser SLAM positioning is an abnormal state.
[0088] In the case where both the RTK positioning data and the visual positioning data are invalid, the abnormal recognition judgment condition constructed in the embodiments of the present application can be expressed in the following form:
[0089] rtk_loc_var < v5, and slam_loc_var > v6, and lidar_slam_hl_flag!= 2, and
[0090] dp_rtk_loc < d4, (6)
[0091] Wherein, rtk_loc_var represents the positioning variance between the RTK positioning data and the fused positioning data, slam_loc_var represents the positioning variance between the laser positioning data and the fused positioning data, lidar_slam_hl_flag! represents the initial matching confidence output by the laser SLAM, dp_rtk_loc represents the positioning error between the RTK positioning data and the fused positioning data, v5 is the third preset variance threshold, v6 is the fourth preset variance threshold, and d4 is the fourth preset error threshold.
[0092] The initial matching confidence of the laser positioning data being 2 means that the current laser positioning data is not of high confidence. That is, abnormal recognition is only performed in the case of non-high confidence. Because in the case where other positioning data such as RTK positioning data and visual positioning data are all invalid, if the laser positioning data is currently of high confidence, in order to ensure the stability of the fused positioning, the positioning result of the laser SLAM can be first selected to be trusted and abnormal recognition is no longer performed.
[0093] It should be noted that although both the RTK positioning data and the visual positioning data are invalid in this case, the changing trends of the positioning errors of the RTK positioning data and the fused positioning data are relatively consistent in a short period of time. Therefore, the laser positioning anomaly can still be judged based on the positioning error and positioning variance between the RTK positioning data and the fused positioning data.
[0094] In some embodiments of the present application, after performing abnormality identification on the laser SLAM positioning based on the point cloud matching feature information, multiple positioning data, and the validity of the multiple positioning data to obtain the abnormality identification result of the laser SLAM positioning, the abnormality identification method of the laser SLAM positioning also includes: if the abnormality identification result of the laser SLAM positioning is an abnormal state, reducing the initial positioning reliability of the laser positioning data; otherwise, retaining the initial positioning reliability of the laser positioning data.
[0095] If the laser SLAM positioning is identified as an abnormal state based on the above embodiment, it means that the current laser positioning data is unreliable, and the matching confidence of the initial output of the laser positioning data can be reduced. The specific reduction can be flexibly set according to actual needs and is not specifically limited here. The embodiment of the present application improves the positioning accuracy and positioning stability of the autonomous driving vehicle by dynamically adjusting the confidence of the laser positioning data through real-time online recognition of the abnormal state of laser positioning.
[0096] The embodiment of the present application also provides a laser SLAM positioning abnormality recognition device 200, such as Figure 2 As shown, a schematic diagram of the structure of an abnormality recognition device for laser SLAM positioning in an embodiment of the present application is provided, wherein the abnormality recognition device 200 for laser SLAM positioning includes: a first acquisition unit 210, a second acquisition unit 220 and an identification unit 230, wherein:
[0097] A first acquisition unit 210 is used to acquire point cloud matching feature information of the autonomous driving vehicle;
[0098] A second acquisition unit 220, configured to acquire a plurality of positioning data of the autonomous driving vehicle and determine validity of the plurality of positioning data, wherein the plurality of positioning data are output by a plurality of positioning sources on the autonomous driving vehicle;
[0099] The identification unit 230 is used to perform abnormality identification on the laser SLAM positioning according to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data to obtain an abnormality identification result of the laser SLAM positioning.
[0100] In some embodiments of the present application, the point cloud matching feature information includes the eigenvalue of an optimized information matrix obtained based on a point cloud matching algorithm, and the identification unit 230 is specifically used to: if the eigenvalue of the optimized information matrix is less than a preset eigenvalue threshold, then determine that the abnormal identification result of the laser SLAM positioning is an abnormal state; otherwise, perform abnormal identification on the laser SLAM positioning based on multiple positioning data and the validity of the positioning data to obtain an abnormal identification result of the laser SLAM positioning.
[0101] In some embodiments of the present application, the identification unit 230 is specifically used to: when it is impossible to determine that the abnormal identification result of the laser SLAM positioning is an abnormal state according to the point cloud matching feature information, obtain multiple positioning data within a preset time period of the cache; calculate the positioning error and positioning variance between the multiple positioning data according to the multiple positioning data within the preset time period; perform abnormal identification of the laser SLAM positioning according to the positioning error and positioning variance between the multiple positioning data and the validity of the multiple positioning data, and obtain the abnormal identification result of the laser SLAM positioning.
[0102] In some embodiments of the present application, the positioning data includes RTK positioning data, laser positioning data, visual positioning data and fused positioning data, and the identification unit 230 is specifically used to: if the RTK positioning data and the visual positioning data are both in a valid state, then according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the positioning variance between the visual positioning data and the RTK positioning data, perform abnormal identification on the laser SLAM positioning to obtain an abnormal identification result of the laser SLAM positioning; if the RTK positioning data is in a valid state and the visual positioning data is in a valid state, the identification unit 230 is specifically used to: if the RTK positioning data and the visual positioning data are both in a valid state, then according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the positioning variance between the visual positioning data and the RTK positioning data, perform abnormal identification on the laser SLAM positioning to obtain an abnormal identification result of the laser SLAM positioning; If the RTK positioning data is in an invalid state, then according to the continuous count of the RTK positioning data in the valid state, the positioning error between the RTK positioning data and the fused positioning data, and the positioning variance between the laser positioning data and the fused positioning data, the laser SLAM positioning is identified as abnormal, and the abnormal identification result of laser SLAM positioning is obtained; if the RTK positioning data and the visual positioning data are both in an invalid state, then according to the positioning variance and positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the initial positioning reliability of the laser positioning data, the laser SLAM positioning is identified as abnormal, and the abnormal identification result of laser SLAM positioning is obtained.
[0103] In some embodiments of the present application, the identification unit 230 is specifically used to: determine whether the positioning error between the visual positioning data and the RTK positioning data is smaller than the positioning error between the laser positioning data and the visual positioning data, or, the positioning error between the visual positioning data and the RTK positioning data is smaller than a first preset error threshold and the positioning error between the laser positioning data and the visual positioning data is smaller than a second preset error threshold; if the positioning error between the visual positioning data and the RTK positioning data is smaller than the positioning error between the laser positioning data and the visual positioning data, and / or, the positioning error between the visual positioning data and the RTK positioning data is smaller than a first preset error threshold and the laser positioning data is smaller than a second preset error threshold If the positioning error between the laser positioning data and the visual positioning data is less than the second preset error threshold, the laser SLAM positioning is identified as abnormal according to the positioning variance between the laser positioning data and the fusion positioning data and the positioning variance between the visual positioning data and the RTK positioning data; if the positioning error between the visual positioning data and the RTK positioning data is not less than the positioning error between the laser positioning data and the visual positioning data, and the positioning error between the visual positioning data and the RTK positioning data is not less than the first preset error threshold, and the positioning error between the laser positioning data and the visual positioning data is not less than the second preset error threshold, then the abnormal identification result of the laser SLAM positioning is determined to be an abnormal state.
[0104] In some embodiments of the present application, the identification unit 230 is specifically used to: if the continuous count of the RTK positioning data in a valid state is greater than a first preset number threshold, and the positioning error between the RTK positioning data and the fused positioning data is less than a third preset error threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than the first preset variance threshold, then an abnormality count is performed on the laser SLAM positioning; if the cumulative abnormality count of the laser SLAM positioning reaches a second preset number threshold, or the positioning variance between the laser positioning data and the fused positioning data is greater than the second preset variance threshold, then the abnormal identification result of the laser SLAM positioning is determined to be an abnormal state, wherein the second preset variance threshold is greater than the first preset variance threshold.
[0105] In some embodiments of the present application, the identification unit 230 is specifically used to: if the positioning variance between the RTK positioning data and the fused positioning data is less than a third preset variance threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than a fourth preset variance threshold, and the initial positioning confidence of the laser positioning data is in a preset confidence interval, and the positioning error between the RTK positioning data and the fused positioning data is less than a fourth preset error threshold, then determine that the abnormal identification result of the laser SLAM positioning is an abnormal state.
[0106] In some embodiments of the present application, the laser SLAM positioning anomaly identification device also includes: a confidence updating unit, which is used to perform anomaly identification on the laser SLAM positioning based on the point cloud matching feature information, multiple positioning data, and the validity of the multiple positioning data, and after obtaining the anomaly identification result of the laser SLAM positioning, if the anomaly identification result of the laser SLAM positioning is an abnormal state, then reduce the initial positioning confidence of the laser positioning data; otherwise, retain the initial positioning confidence of the laser positioning data.
[0107] It can be understood that the above-mentioned laser SLAM positioning anomaly recognition device can implement the various steps of the laser SLAM positioning anomaly recognition method provided in the aforementioned embodiment, and the relevant explanations about the laser SLAM positioning anomaly recognition method are applicable to the laser SLAM positioning anomaly recognition device, which will not be repeated here.
[0108] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 3 At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.
[0109] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0110] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.
[0111] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming an abnormality recognition device for laser SLAM positioning at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:
[0112] Obtain point cloud matching feature information of the autonomous driving vehicle;
[0113] Acquire multiple positioning data of the autonomous vehicle and determine validity of the multiple positioning data, wherein the multiple positioning data are output by multiple positioning sources on the autonomous vehicle;
[0114] According to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data, the laser SLAM positioning is identified as abnormal, and an abnormal identification result of the laser SLAM positioning is obtained.
[0115] The above application Figure 1The method performed by the abnormal identification device for laser SLAM positioning disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.
[0116] The present application also provides a computer program product, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by an electronic device including multiple application programs, enable the electronic device to execute Figure 1 The method performed by the abnormality recognition device for laser SLAM positioning in the illustrated embodiment is specifically used to perform:
[0117] Obtain point cloud matching feature information of the autonomous driving vehicle;
[0118] Acquire multiple positioning data of the autonomous vehicle and determine validity of the multiple positioning data, wherein the multiple positioning data are output by multiple positioning sources on the autonomous vehicle;
[0119] According to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data, the laser SLAM positioning is identified as abnormal, and an abnormal identification result of the laser SLAM positioning is obtained.
[0120] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0121] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0122] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0124] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0125] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0126] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0127] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0128] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.
[0129] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims
1. A laser SLAM positioning anomaly recognition method, wherein: The abnormality recognition method of laser SLAM positioning includes: Obtain point cloud matching feature information of the autonomous driving vehicle; Acquire multiple positioning data of the autonomous vehicle and determine validity of the multiple positioning data, wherein the multiple positioning data are output by multiple positioning sources on the autonomous vehicle; According to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data, the laser SLAM positioning is identified as abnormal, and an abnormal identification result of the laser SLAM positioning is obtained.
2. The method for identifying abnormalities in laser SLAM positioning according to claim 1, wherein: The point cloud matching feature information includes the eigenvalue of the optimized information matrix obtained based on the point cloud matching algorithm. The laser SLAM positioning is identified for abnormality based on the point cloud matching feature information and the validity of the multiple positioning data. The abnormality identification result of the laser SLAM positioning includes: If the eigenvalue of the optimized information matrix is less than the preset eigenvalue threshold, the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state; Otherwise, according to the various positioning data and the validity of the positioning data, the laser SLAM positioning is identified as abnormal, and the abnormal identification result of the laser SLAM positioning is obtained.
3. The abnormality identification method of laser SLAM positioning as claimed in claim 1, wherein: According to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data, the laser SLAM positioning is subjected to abnormal identification, and the abnormal identification result of the laser SLAM positioning is obtained, which includes: When the abnormal recognition result of the laser SLAM positioning cannot be determined as an abnormal state according to the point cloud matching feature information, obtaining a plurality of positioning data within a preset time period of the cache; Calculate the positioning error and positioning variance between the various positioning data according to the various positioning data within a preset time period; According to the positioning errors and positioning variances between various positioning data and the validity of various positioning data, the laser SLAM positioning is identified abnormally to obtain the abnormal identification result of the laser SLAM positioning.
4. The abnormality identification method of laser SLAM positioning as claimed in claim 3, wherein: The positioning data includes RTK positioning data, laser positioning data, visual positioning data and fusion positioning data. The laser SLAM positioning is identified based on the positioning errors and positioning variances between the various positioning data and the validity of the various positioning data. The abnormal identification results of the laser SLAM positioning include: If both the RTK positioning data and the visual positioning data are in valid states, then according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fusion positioning data, and the positioning variance between the visual positioning data and the RTK positioning data, abnormal identification is performed on the laser SLAM positioning to obtain an abnormal identification result of the laser SLAM positioning; If the RTK positioning data is in a valid state and the visual positioning data is in an invalid state, then according to the continuous count of the RTK positioning data being in a valid state, the positioning error between the RTK positioning data and the fused positioning data, and the positioning variance between the laser positioning data and the fused positioning data, the laser SLAM positioning is identified as abnormal, and an abnormal identification result of the laser SLAM positioning is obtained; If both the RTK positioning data and the visual positioning data are in an invalid state, the laser SLAM positioning is identified as abnormal based on the positioning variance and positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the initial positioning reliability of the laser positioning data to obtain an abnormal identification result of the laser SLAM positioning.
5. The abnormality identification method of laser SLAM positioning as claimed in claim 4, wherein: The performing abnormal identification on the laser SLAM positioning according to the positioning error between the visual positioning data and the RTK positioning data, the positioning error between the laser positioning data and the visual positioning data, the positioning variance between the laser positioning data and the fusion positioning data, and the positioning variance between the visual positioning data and the RTK positioning data comprises: Determine whether a positioning error between the visual positioning data and the RTK positioning data is less than a positioning error between the laser positioning data and the visual positioning data, or whether a positioning error between the visual positioning data and the RTK positioning data is less than a first preset error threshold and a positioning error between the laser positioning data and the visual positioning data is less than a second preset error threshold; If the positioning error between the visual positioning data and the RTK positioning data is smaller than the positioning error between the laser positioning data and the visual positioning data, and / or the positioning error between the visual positioning data and the RTK positioning data is smaller than a first preset error threshold and the positioning error between the laser positioning data and the visual positioning data is smaller than a second preset error threshold, then according to the positioning variance between the laser positioning data and the fused positioning data and the positioning variance between the visual positioning data and the RTK positioning data, abnormality identification is performed on the laser SLAM positioning; If the positioning error between the visual positioning data and the RTK positioning data is not less than the positioning error between the laser positioning data and the visual positioning data, and the positioning error between the visual positioning data and the RTK positioning data is not less than a first preset error threshold, and the positioning error between the laser positioning data and the visual positioning data is not less than a second preset error threshold, then the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state.
6. The abnormality identification method of laser SLAM positioning as claimed in claim 4, wherein: According to the continuous count of the RTK positioning data being in a valid state, the positioning error between the RTK positioning data and the fused positioning data, and the positioning variance between the laser positioning data and the fused positioning data, the laser SLAM positioning is subjected to abnormal identification, and the abnormal identification result of the laser SLAM positioning is obtained, which includes: If the continuous count of the RTK positioning data being in a valid state is greater than a first preset number threshold, and the positioning error between the RTK positioning data and the fused positioning data is less than a third preset error threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than a first preset variance threshold, then an abnormal count is performed on the laser SLAM positioning; If the cumulative abnormal count of the laser SLAM positioning reaches a second preset number threshold, or the positioning variance between the laser positioning data and the fused positioning data is greater than a second preset variance threshold, then the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state, wherein the second preset variance threshold is greater than the first preset variance threshold.
7. The abnormality identification method of laser SLAM positioning as claimed in claim 4, wherein: The abnormality identification of laser SLAM positioning is performed according to the positioning variance and positioning error between the RTK positioning data and the fused positioning data, the positioning variance between the laser positioning data and the fused positioning data, and the initial positioning reliability of the laser positioning data, and the abnormality identification result of the laser SLAM positioning is obtained, including: If the positioning variance between the RTK positioning data and the fused positioning data is less than the third preset variance threshold, and the positioning variance between the laser positioning data and the fused positioning data is greater than the fourth preset variance threshold, and the initial positioning confidence of the laser positioning data is in the preset confidence interval, and the positioning error between the RTK positioning data and the fused positioning data is less than the fourth preset error threshold, then the abnormal recognition result of the laser SLAM positioning is determined to be an abnormal state.
8. The abnormality identification method of laser SLAM positioning as claimed in claim 1, wherein: After performing abnormality recognition on the laser SLAM positioning according to the point cloud matching feature information, the multiple positioning data and the validity of the multiple positioning data to obtain the abnormality recognition result of the laser SLAM positioning, the abnormality recognition method of the laser SLAM positioning further includes: If the abnormal recognition result of the laser SLAM positioning is an abnormal state, the initial positioning confidence of the laser positioning data is reduced; Otherwise, the initial positioning confidence of the laser positioning data is retained.
9. A laser SLAM positioning anomaly recognition device, wherein: The abnormality recognition device of the laser SLAM positioning includes: A first acquisition unit, used to acquire point cloud matching feature information of the autonomous driving vehicle; A second acquisition unit, configured to acquire a plurality of positioning data of the autonomous driving vehicle and determine validity of the plurality of positioning data, wherein the plurality of positioning data are output by a plurality of positioning sources on the autonomous driving vehicle; The identification unit is used to identify abnormalities in laser SLAM positioning according to the point cloud matching feature information, multiple positioning data and the validity of the multiple positioning data, so as to obtain an abnormality identification result of the laser SLAM positioning.
10. An electronic device comprising: processor; as well as A memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes the abnormality identification method for laser SLAM positioning according to any one of claims 1 to 8.
11. A computer program product, comprising a computer program or an instruction, wherein when the computer program or the instruction is executed by a processor, the abnormality recognition method of laser SLAM positioning according to any one of claims 1 to 8 is implemented.
