Fusion positioning method, device and electronic device for autonomous driving vehicle

Abstract

The present application discloses a fusion positioning method, device and electronic device for an autonomous driving vehicle. The method includes: obtaining high-precision positioning data and the positioning state of the high-precision positioning data, the matching score of the lidar positioning data and the lidar positioning data, and the positioning state of the high-precision positioning data is obtained based on a preset positioning state prediction model; determining a fusion positioning strategy according to the positioning state of the high-precision positioning data and the matching score of the lidar positioning data, and the fusion positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the lidar positioning data; performing fusion processing on the high-precision positioning data and the lidar positioning data according to the fusion positioning strategy to obtain a fusion positioning result. The present application adaptively determines different fusion positioning strategies for different positioning states of high-precision positioning data and the matching results of lidar data, can be applicable to the positioning requirements of most scenarios, and improves the fusion positioning accuracy of autonomous driving vehicles.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular, to a method, apparatus, and electronic device for integrated positioning of autonomous vehicles. Background Art

[0002] In the autonomous driving scenario, it is often necessary to fuse the data of multiple sensors to obtain the integrated positioning information in order to ensure the positioning accuracy of autonomous vehicles in various complex road conditions.

[0003] However, in different driving road conditions and driving scenarios, the positioning accuracy and positioning effect between sensors are different. The method of directly switching according to the positioning accuracy of each sensor in different scenarios to achieve integrated positioning obviously cannot achieve the optimal positioning effect and cannot meet the positioning requirements of the autonomous driving scenario.

[0004] In addition, due to factors such as calibration and time delay between each sensor, there are large errors in the positioning data obtained between different sensors during integrated positioning. Directly performing integrated positioning will instead cause greater errors, which will seriously affect the positioning accuracy and safety of autonomous vehicles. Summary of the Invention

[0005] Embodiments of this application provide a method, apparatus, and electronic device for integrated positioning of autonomous vehicles to improve the integrated positioning accuracy of autonomous vehicles.

[0006] Embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, embodiments of this application provide a method for integrated positioning of autonomous vehicles, where the method includes:

[0008] Obtain high-precision positioning data and the positioning status of the high-precision positioning data, as well as lidar positioning data and the matching score of the lidar positioning data, where the positioning status of the high-precision positioning data is predicted based on a preset positioning status prediction model;

[0009] Determine an integrated positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data, where the integrated positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the lidar positioning data;

[0010] Perform fusion processing on the high-precision positioning data and the lidar positioning data according to the integrated positioning strategy to obtain the integrated positioning result of the autonomous vehicle.

[0011] Optionally, the high-precision positioning data includes inertial navigation RTK positioning data. Determining the fusion positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data includes:

[0012] If the positioning status of the inertial navigation RTK positioning data is an untrusted status, determine a first fusion positioning strategy according to the matching score of the lidar positioning data;

[0013] If the positioning status of the inertial navigation RTK positioning data is an uncertain status, determine a second fusion positioning strategy according to the matching score of the lidar positioning data;

[0014] If the positioning status of the inertial navigation RTK positioning data is a trusted status, determine a third fusion positioning strategy according to the matching score of the lidar positioning data.

[0015] Optionally, if the positioning status of the inertial navigation RTK positioning data is an untrusted status, determining that the fusion positioning strategy is the first fusion positioning strategy includes:

[0016] If the matching score of the lidar positioning data is not less than a preset matching score threshold, determine a first initial weight of the lidar positioning data;

[0017] Based on the matching score of the lidar positioning data, adjust the first initial weight of the lidar positioning data using a first weight adjustment strategy to obtain a first fusion weight of the lidar positioning data;

[0018] Determine a first fusion weight of the inertial navigation RTK positioning data according to the first fusion weight of the lidar positioning data.

[0019] Optionally, if the positioning status of the inertial navigation RTK positioning data is an uncertain status, determining a second fusion positioning strategy according to the matching score of the lidar positioning data includes:

[0020] If the matching score of the lidar positioning data is not less than a preset matching score threshold, determine a second initial weight of the lidar positioning data;

[0021] Based on the matching score of the lidar positioning data, adjust the second initial weight of the lidar positioning data using a second weight adjustment strategy to obtain a second fusion weight of the lidar positioning data;

[0022] Determine a second fusion weight of the inertial navigation RTK positioning data according to the second fusion weight of the lidar positioning data.

[0023] Optionally, if the positioning status of the inertial navigation RTK positioning data is a reliable status, determining a third fusion positioning strategy according to the matching score of the lidar positioning data includes:

[0024] Determine the initial weight of the inertial navigation RTK positioning data and the differential status of the inertial navigation RTK positioning data;

[0025] According to the differential status of the inertial navigation RTK positioning data, use a third weight adjustment strategy to adjust the initial weight of the inertial navigation RTK positioning data to obtain the third fusion weight of the inertial navigation RTK positioning data;

[0026] Determine the third fusion weight of the lidar positioning data according to the third fusion weight of the inertial navigation RTK positioning data.

[0027] Optionally, the differential status of the inertial navigation RTK positioning data includes a fixed solution status and a non-fixed solution status. The adjusting the initial weight of the inertial navigation RTK positioning data according to the differential status of the inertial navigation RTK positioning data by using a third weight adjustment strategy to obtain the third fusion weight of the inertial navigation RTK positioning data includes:

[0028] According to the differential status of the inertial navigation RTK positioning data, determine the continuous count of the inertial navigation RTK positioning data in the non-fixed solution status;

[0029] Adjust the initial weight of the inertial navigation RTK positioning data according to the continuous count of the inertial navigation RTK positioning data in the non-fixed solution status to obtain the third fusion weight of the inertial navigation RTK positioning data.

[0030] Optionally, if the positioning status of the inertial navigation RTK positioning data is a reliable status, determining a third fusion positioning strategy according to the matching score of the lidar positioning data includes:

[0031] Determine the differential status of the inertial navigation RTK positioning data;

[0032] In the case where the differential status of the inertial navigation RTK positioning data is the fixed solution status, use a preset multipath identification strategy to perform multipath identification;

[0033] According to the multipath identification result and the matching score of the lidar positioning data, determine the fourth fusion weight of the inertial navigation RTK positioning data and the fourth fusion weight of the lidar positioning data.

[0034] In a second aspect, an embodiment of the present application further provides a fusion positioning device for an autonomous driving vehicle, where the device includes:

[0035] An acquisition unit, used to acquire high-precision positioning data and a positioning state of the high-precision positioning data, and laser radar positioning data and a matching score of the laser radar positioning data, wherein the positioning state of the high-precision positioning data is predicted based on a preset positioning state prediction model;

[0036] a determination unit, configured to determine a fusion positioning strategy according to a positioning state of the high-precision positioning data and a matching score of the laser radar positioning data, wherein the fusion positioning strategy includes a fusion weight of the high-precision positioning data and a fusion weight of the laser radar positioning data;

[0037] The fusion positioning unit is used to fuse the high-precision positioning data and the lidar positioning data according to the fusion positioning strategy to obtain a fusion positioning result of the autonomous driving vehicle.

[0038] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0039] Processor; and

[0040] A memory arranged to store computer executable instructions which, when executed, cause the processor to perform any of the methods described above.

[0041] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple application programs, the electronic device executes any of the aforementioned methods.

[0042] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: the fusion positioning method of the autonomous driving vehicle in the embodiments of the present application first obtains the high-precision positioning data and the positioning status of the high-precision positioning data, as well as the matching score of the laser radar positioning data and the laser radar positioning data, and the positioning status of the high-precision positioning data is predicted based on the preset positioning status prediction model; then, according to the positioning status of the high-precision positioning data and the matching score of the laser radar positioning data, a fusion positioning strategy is determined, and the fusion positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the laser radar positioning data; finally, according to the fusion positioning strategy, the high-precision positioning data and the laser radar positioning data are fused and processed to obtain the fusion positioning result of the autonomous driving vehicle. The fusion positioning method of the autonomous driving vehicle in the embodiments of the present application adaptively determines different fusion positioning strategies for different positioning states of the high-precision positioning data and the matching results of the laser radar data, which can be applied to the positioning requirements of most scenarios and improves the fusion positioning accuracy of the autonomous driving vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation to the present application. In the drawings:

[0044] Figure 1 is a schematic flowchart of a fusion positioning method for an autonomous driving vehicle in an embodiment of the present application;

[0045] Figure 2 is a schematic structural diagram of a fusion positioning device for an autonomous driving vehicle in an embodiment of the present application;

[0046] Figure 3 is a schematic structural diagram of an electronic device in an embodiment of the present application. Specific embodiments

[0047] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0048] The following will describe in detail the technical solutions provided by each embodiment of the present application in conjunction with the drawings.

[0049] An embodiment of the present application provides a fusion positioning method for an autonomous driving vehicle. As Figure 1 shown, there is provided a schematic flowchart of a fusion positioning method for an autonomous driving vehicle in an embodiment of the present application. The method at least includes the following steps S110 to step S130:

[0050] Step S110, obtaining high-precision positioning data and the positioning state of the high-precision positioning data, as well as lidar positioning data and the matching score of the lidar positioning data. The positioning state of the high-precision positioning data is predicted based on a preset positioning state prediction model.

[0051] When performing fusion positioning of an autonomous driving vehicle in an embodiment of the present application, it is necessary to first obtain high-precision positioning data collected by a high-precision positioning device. The high-precision positioning device may be, for example, an inertial navigation RTK (Real-time kinematic) positioning device, which generally has a high positioning accuracy. However, since the inertial navigation RTK positioning data depends on external factors such as the quality of satellite positioning signals, in this embodiment of the present application, a trained preset positioning state prediction model is used to predict the positioning state of the high-precision positioning data, so as to evaluate the positioning effect of the high-precision positioning data.

[0052] In addition to the above-mentioned high-precision positioning data, the positioning data fused and processed in the embodiments of this application also includes lidar positioning data obtained by matching with a pre-constructed lidar SLAM map. In addition, the lidar SLAM matching algorithm also outputs a matching score for the lidar positioning data, which can reflect its positioning accuracy to a certain extent. Of course, specifically how to perform lidar SLAM matching can be determined by those skilled in the art according to existing matching algorithms, and no specific limitation is made here.

[0053] Step S120: Determine a fusion positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data. The fusion positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the lidar positioning data.

[0054] Since the positioning status of the high-precision positioning data reflects the positioning accuracy of the high-precision positioning data, and the matching score of the lidar positioning data also reflects the positioning accuracy of the lidar positioning data, this provides strong support for the adaptive determination of the fusion positioning strategy. The fusion positioning strategy can be understood as weighing the importance degree, that is, the weight size, occupied by positioning data from different sources during the fusion process. According to the relative change between the positioning accuracy of the high-precision positioning data and the positioning accuracy of the lidar positioning data, the weight size of the high-precision positioning data and the weight size of the lidar positioning data in the fusion stage can be adaptively determined, so as to meet the positioning requirements of different positioning scenarios and improve the fusion positioning accuracy.

[0055] Step S130: Perform fusion processing on the high-precision positioning data and the lidar positioning data according to the fusion positioning strategy to obtain the fusion positioning result of the autonomous driving vehicle.

[0056] Based on the fusion positioning strategy obtained from the foregoing steps, the high-precision positioning data and the lidar data can be fused according to their respective corresponding fusion weight sizes, so as to obtain the fusion positioning result of the autonomous driving vehicle.

[0057] The fusion positioning method of the autonomous driving vehicle in the embodiments of this application adaptively determines different fusion positioning strategies for different positioning statuses of the high-precision positioning data and the matching results of the lidar data, can meet the positioning requirements of most scenarios, and improves the fusion positioning accuracy of the autonomous driving vehicle.

[0058] For the convenience of understanding the embodiments of the present application, a training method of the preset positioning state prediction model is further introduced here: First, inertial RTK data and the data of the corresponding high-precision inertial device are obtained. The inertial RTK data includes inertial RTK positioning data, the absolute time of the inertial RTK data, the RTK differential state, the Horizontal Dilution Of Precision (HDOP), and the number of satellites. The high-precision inertial device is equipped with a high-precision closed-loop fiber optic gyro and an acceleration sensor, and its positioning accuracy can reach the centimeter level or even the millimeter level, which has higher positioning accuracy compared with general inertial devices. Therefore, it can be regarded as a true value device. Then, the positioning error of the inertial RTK positioning data is determined according to the data of the high-precision inertial device; the confidence of the inertial RTK positioning data is determined according to the positioning error of the inertial RTK positioning data. The greater the positioning error, the smaller the confidence given. Of course, how to assign values specifically can be implemented in combination with the prior art. Finally, the positioning state prediction model is trained using the inertial RTK information and the confidence of the inertial RTK positioning information. Here, the structure of the Long Short-Term Memory (LSTM) network can be used for training to obtain the final preset positioning state prediction model.

[0059] After obtaining the preset positioning state prediction model, by inputting the current RTK differential state, HDOP, and the number of satellites into the preset positioning state prediction model, the positioning state of the inertial RTK positioning data can be obtained. The specific positioning state types that the model can output for the inertial RTK positioning data can be predefined in the training process. For example, it can include multiple positioning states such as 0 = credible state, 1 = non-credible state, 2 = uncertain state, etc.

[0060] In some embodiments of the present application, the high-precision positioning data includes inertial RTK positioning data. Determining the fusion positioning strategy according to the positioning state of the high-precision positioning data and the matching score of the lidar positioning data includes: If the positioning state of the inertial RTK positioning data is a non-credible state, then determine the first fusion positioning strategy according to the matching score of the lidar positioning data; if the positioning state of the inertial RTK positioning data is an uncertain state, then determine the second fusion positioning strategy according to the matching score of the lidar positioning data; if the positioning state of the inertial RTK positioning data is a credible state, then determine the third fusion positioning strategy according to the matching score of the lidar positioning data.

[0061] When determining the fusion positioning strategy in the embodiments of the present application, the positioning states of different inertial RTK positioning data are first distinguished. For example, it may include three types: 0 = reliable state, 1 = unreliable state, and 2 = uncertain state. The reliable state indicates that the inertial RTK positioning data predicted based on the preset positioning state prediction model has a high credibility, that is, the positioning accuracy is high; the unreliable state indicates that the currently predicted inertial RTK positioning data has a low credibility, that is, the positioning accuracy is low; and the uncertain state indicates that the credibility of the currently predicted inertial RTK positioning data is between the reliable state and the unreliable state, and the positioning accuracy is also between the two.

[0062] Therefore, it can be seen that the positioning accuracies corresponding to the positioning states of different inertial RTK positioning data are different, so the importance degrees actually occupied in the fusion stage are also different. Here, different fusion positioning strategies can be further adopted in combination with the specific scenario and the positioning accuracy of the lidar positioning data.

[0063] In some embodiments of the present application, determining the fusion positioning strategy according to the positioning state of the high-precision positioning data and the matching score of the lidar positioning data includes: if the positioning state of the inertial RTK positioning data is the unreliable state or the uncertain state, determining whether the matching score of the lidar positioning data is not less than a preset matching score threshold; if so, determining the initial weight of the lidar positioning data; based on the matching score of the lidar positioning data, using a preset weight adjustment strategy to adjust the initial weight of the lidar positioning data to obtain the fusion weight of the lidar positioning data; and determining the fusion weight of the inertial RTK positioning data according to the fusion weight of the lidar positioning data.

[0064] For the positioning states of untrusted state and uncertain state, both of these states indicate that the positioning effect of the current inertial navigation RTK positioning data is not good enough. Then, the matching score of the lidar positioning data can be further compared with a preset matching score threshold to determine whether the current lidar positioning data meets the requirements of subsequent fusion positioning. If the matching score of the lidar positioning data is greater than the preset matching score threshold, it means that the lidar positioning data can be used for subsequent fusion positioning. Then, at this time, a relatively large initial weight can be assigned to the lidar positioning data first, and then, combined with the actual size of its matching score, the initial weight corresponding to the lidar positioning data is adjusted by using a preset weight adjustment strategy. The core of the preset weight adjustment strategy is that the higher the matching score of the lidar positioning data, the higher the fusion weight corresponding to the lidar positioning data. Since the sources of the fusion data are mainly the lidar positioning data and the inertial navigation RTK positioning data, the sum of the fusion weights of the two can be set to 1. After the fusion weight of the lidar positioning data is determined, the fusion weight of the inertial navigation RTK positioning data is also determined accordingly.

[0065] The sizes of the above-mentioned preset matching score threshold and the initial weight can both be flexibly set according to the actual situation and are not specifically limited herein.

[0066] In some embodiments of the present application, the determining that the fusion positioning strategy is the first fusion positioning strategy if the positioning state of the inertial navigation RTK positioning data is an untrusted state includes: if the matching score of the lidar positioning data is not less than the preset matching score threshold, determining the first initial weight of the lidar positioning data; based on the matching score of the lidar positioning data, adjusting the first initial weight of the lidar positioning data by using a first weight adjustment strategy to obtain the first fusion weight of the lidar positioning data; determining the first fusion weight of the inertial navigation RTK positioning data according to the first fusion weight of the lidar positioning data.

[0067] For the case where the positioning status of the inertial navigation RTK positioning data is in an untrusted state, compared with the other two positioning statuses, the credibility and positioning accuracy of its positioning result are both lower. Therefore, when determining the first fusion positioning strategy in this case, the availability of the lidar positioning data can be judged first. If the matching score slam_core of the lidar positioning data is not less than the preset matching score threshold, it means that the lidar positioning data can be used for subsequent fusion positioning and has a certain positioning accuracy. Then, when the positioning accuracy of the inertial navigation RTK positioning data is very low, a relatively large first initial weight can be given to the lidar positioning data, and then, combined with the actual matching score of the lidar positioning data, the first initial weight of the lidar positioning data is adjusted using the first weight adjustment strategy to obtain the final first fusion weight K_LIO of the lidar positioning data. Finally, the first fusion weight K_GPS of the inertial navigation RTK positioning data is calculated based on the first fusion weight K_LIO of the lidar positioning data.

[0068] To facilitate the understanding of the embodiments of the present application, it is assumed here that the preset matching score threshold is 1.8 and the first initial weight of the lidar positioning data is 0.9. Then the specific implementation logic of the embodiments of the present application can be expressed as:

[0069]

[0070] In the actual scenario, the range of the matching score slam_core of the lidar positioning data is generally 0 to 4. When slam_core is 3, the matching effect is already good enough. Therefore, substituting slam_core = 3, K_LIO = 1 and K_GPS = 0 can be calculated. At this time, it shows that the positioning accuracy of the lidar positioning data is already very high, so the corresponding fusion weight is the largest at this time. When slam_core = 1.8, it means that the positioning accuracy of the lidar data just meets the requirements of fusion positioning at this time. Then it still has a relatively high fusion weight K_LIO = 0.9, that is, the first initial weight. At this time, K_GPS = 0.1. That is, in the case where the inertial navigation RTK positioning data is in an untrusted state, the adjustment range of the first fusion weight of the lidar positioning data is {0.9, 1}, and the adjustment range of the first fusion weight of the inertial navigation RTK positioning data is {0, 0.1}.

[0071] In some embodiments of the present application, when the positioning status of the inertial navigation RTK positioning data is an uncertain status, determining the second fusion positioning strategy according to the matching score of the lidar positioning data includes: if the matching score of the lidar positioning data is not less than a preset matching score threshold, determining the second initial weight of the lidar positioning data; based on the matching score of the lidar positioning data, using a second weight adjustment strategy to adjust the second initial weight of the lidar positioning data to obtain the second fusion weight of the lidar positioning data; determining the second fusion weight of the inertial navigation RTK positioning data according to the second fusion weight of the lidar positioning data.

[0072] For the case where the positioning status of the inertial navigation RTK positioning data is an uncertain status, the credibility and positioning accuracy of its positioning result are between the credible status and the non-credible status. Therefore, the implementation logic of the second fusion positioning strategy is basically the same as that of the first fusion positioning strategy in the credible status, and the main difference lies in that the second initial weight of the lidar positioning data and the corresponding second weight adjustment strategy are different.

[0073] For the convenience of understanding the embodiments of the present application, it is also assumed here that the preset matching score threshold is 1.8. Since the positioning accuracy of the inertial navigation RTK positioning data in this case is better than that in the non-credible status, the second initial weight given to the lidar positioning data can be smaller than the first initial weight in the foregoing embodiments. For example, it can be set to 0.8. Then, the specific implementation logic of the embodiments of the present application can be expressed as:

[0074]

[0075] It can be seen that when slam_core = 3 is substituted, K_LIO = 1 and K_GPS = 0 are calculated. At this time, it shows that the positioning accuracy of the lidar positioning data is very high, so the corresponding fusion weight is the largest. When slam_core = 1.8, it shows that the positioning accuracy of the lidar data just meets the requirements of the fusion positioning. Then it still has a relatively high fusion weight K_LIO = 0.8, that is, the second initial weight. At this time, K_GPS = 0.2. That is, when the inertial navigation RTK positioning data is in an uncertain status, the adjustment range of the second fusion weight of the lidar positioning data is {0.8, 1}, and the adjustment range of the second fusion weight of the inertial navigation RTK positioning data is {0, 0.2}.

[0076] Of course, it should be noted that the size of the above preset matching score threshold and the sizes of the first initial weight and the second initial weight of the lidar positioning data can be flexibly adjusted according to the actual situation, and no specific limitation is made here.

[0077] In some embodiments of the present application, if the positioning status of the inertial navigation RTK positioning data is a trusted status, determining the third fusion positioning strategy according to the matching score of the lidar positioning data includes: determining the initial weight of the inertial navigation RTK positioning data and the differential status of the inertial navigation RTK positioning data; according to the differential status of the inertial navigation RTK positioning data, using a third weight adjustment strategy to adjust the initial weight of the inertial navigation RTK positioning data to obtain the third fusion weight of the inertial navigation RTK positioning data; determining the third fusion weight of the lidar positioning data according to the third fusion weight of the inertial navigation RTK positioning data.

[0078] Different from the previous two positioning statuses, when the positioning status of the inertial navigation RTK positioning data is a trusted status, it indicates that the positioning accuracy of the inertial navigation RTK positioning data predicted by the preset positioning status prediction model is relatively high. Then, a relatively large initial weight can be assigned to the inertial navigation RTK positioning data. In addition, to further improve the fusion positioning accuracy, the differential status of the current inertial navigation RTK positioning data can also be determined, such as whether it is a fixed solution or a non-fixed solution. Furthermore, according to the differential status of the inertial navigation RTK positioning data, a third weight adjustment strategy can be used to adjust the initial weight of the inertial navigation RTK positioning data, thereby further improving the accuracy of judging the positioning accuracy of the inertial navigation RTK positioning data.

[0079] It should be noted that the "positioning status of the inertial navigation RTK positioning data" and the "differential status of the inertial navigation RTK positioning data" in the embodiments of the present application are two different concepts. The former is the positioning effect of the inertial navigation RTK positioning data predicted based on the self-defined positioning status prediction model in the foregoing embodiments, while the latter is a single RTK differential status, such as a fixed solution, a single-point solution, or a floating-point solution, etc. Since a single RTK differential status usually cannot meet the needs of autonomous driving, for example, in complex urban roads, there may still be a large positioning error when the RTK differential status is a fixed solution. Therefore, it is necessary to evaluate the accuracy of the positioning result from a relatively more comprehensive and accurate dimension, that is, the positioning status of the inertial navigation RTK positioning data predicted by the preset positioning status prediction model.

[0080] Although the positioning status of the inertial navigation RTK positioning data can more accurately reflect the positioning accuracy of the positioning result compared with a single RTK positioning differential status, it should also be noted that in the scenario of fusion positioning, it is still necessary to consider how to make good use of the positioning status and differential status of the inertial navigation RTK positioning data, which is also one of the key points of the present application.

[0081] In some embodiments of the present application, the differential state of the inertial navigation RTK positioning data includes a fixed solution state and a non-fixed solution state. Adjusting the initial weight of the inertial navigation RTK positioning data by using a third weight adjustment strategy according to the differential state of the inertial navigation RTK positioning data to obtain the third fusion weight of the inertial navigation RTK positioning data includes: determining the continuous count of the inertial navigation RTK positioning data in the non-fixed solution state according to the differential state of the inertial navigation RTK positioning data; adjusting the initial weight of the inertial navigation RTK positioning data according to the continuous count of the inertial navigation RTK positioning data in the non-fixed solution state to obtain the third fusion weight of the inertial navigation RTK positioning data.

[0082] As mentioned above, the differential state of the inertial navigation RTK positioning data can be divided into two states: fixed solution and non-fixed solution. Generally, the fixed solution can achieve centimeter-level positioning accuracy, while the positioning accuracy of the non-fixed solution is relatively low, generally between decimeter level and meter level, which obviously cannot meet the requirements of autonomous driving.

[0083] When the positioning state of the inertial navigation RTK positioning data is in a credible state, since the prediction of the positioning state is predicted by a preset positioning state prediction model, there may be misidentification. Therefore, the differential state of the corresponding inertial navigation RTK positioning data may not necessarily be a fixed solution, but may also be a non-fixed solution. Therefore, it is possible to first determine whether the differential state of the current inertial navigation RTK positioning data is a non-fixed solution. If it is a non-fixed solution, the duration or continuous count of this non-fixed solution state can be further calculated. The longer the duration or the more the count of the non-fixed solution, the less accurate the credible state of the inertial navigation RTK positioning data predicted by the current preset positioning state prediction model, and the lower the positioning accuracy of the inertial navigation RTK positioning data. Based on this, the initial weight of the inertial navigation RTK positioning data can be adjusted to balance the fusion weights corresponding to different fusion data in this case.

[0084] For the convenience of understanding the embodiments of the present application, it is assumed here that the initial weight of the inertial navigation RTK positioning data is 0.9. Then the specific implementation logic of the embodiments of the present application can be expressed as:

[0085]

[0086]

[0087] Among them, gps_state is the differential state of the inertial RTK positioning data. 4 represents a fixed solution. sat_recovery_cnt represents the continuous count of non-fixed solution states. When the continuous count of non-fixed solution states reaches 300, the initial weight of the inertial RTK positioning data will start to be attenuated and adjusted until the continuous count reaches 700. For example, substituting sat_recovery_cnt = 300, K_GPS = 0.9 can be calculated. Substituting sat_recovery_cnt = 700, K_GPS = 0.1 can be calculated, indicating that when the non-fixed solution state appears for more than 3 seconds, the weight attenuation starts, and the switch is completed until the 7th second.

[0088] That is to say, when the inertial RTK positioning data is in a reliable state and the differential state shows continuous non-fixed solutions, the adjustment range of the third fusion weight of the inertial RTK positioning data is {0.1, 0.9}, and the adjustment range of the second fusion weight of the lidar positioning data is also {0.1, 0.9}.

[0089] Of course, it should be noted that the initial weight of the above inertial RTK positioning data and the adjustment strategy for the initial weight can be flexibly adjusted according to the actual situation and are not specifically limited here.

[0090] In some embodiments of the present application, if the positioning state of the inertial RTK positioning data is a reliable state, then determining the third fusion positioning strategy according to the matching score of the lidar positioning data includes: determining the differential state of the inertial RTK positioning data; when the differential state of the inertial RTK positioning data is in a fixed solution state, using a preset multipath recognition strategy to perform multipath recognition; according to the multipath recognition result and the matching score of the lidar positioning data, determining the fourth fusion weight of the inertial RTK positioning data and the fourth fusion weight of the lidar positioning data.

[0091] When the positioning status of the inertial navigation RTK positioning data is in a trustworthy state, in addition to the situations in the foregoing embodiments that may cause the trustworthy state of the inertial navigation RTK positioning data to be inaccurate, there is also a situation where the "multipath effect" occurs. In satellite positioning measurements such as GNSS (Global Navigation Satellite System), if the satellite signals (reflected waves) reflected by reflectors near the measured station enter the receiver antenna, they will interfere with the signals directly from the satellites (direct waves), thereby causing the observed values to deviate from the true values and generating the so-called "multipath error". This interference delay effect caused by the propagation of multipath signals is called the "multipath effect". When a positioning device relying on satellite positioning signals has a multipath effect, the introduction of multipath errors will affect the positioning accuracy of autonomous vehicles.

[0092] Based on this, in the embodiments of the present application, the differential status of the inertial navigation RTK positioning data can also be determined first. When the differential status is a fixed solution, although it indicates that the theoretical positioning accuracy of the inertial navigation RTK positioning data is relatively high at this time, if the inertial navigation RTK positioning device has a multipath effect, the actual positioning error will increase greatly. Therefore, it is possible to further identify whether there is a multipath effect. For the differential status being a non-fixed solution status, it already indicates that the accuracy of the satellite positioning data is insufficient, and thus there is no actual need to identify the multipath effect anymore.

[0093] Therefore, the multipath recognition strategy adopted in the embodiments of the present application can be as follows: when the differential status corresponding to the satellite positioning data at the current moment is in a fixed solution state, determine the current height difference according to the satellite positioning data at the current moment; obtain the current driving speed of the autonomous vehicle, and determine the maximum height difference according to the current driving speed and the pitch angle. Determine the first multipath identifier flag_dh according to the current driving speed vb, the maximum height difference max_dh, and the current height difference dh; determine the first antenna speed vh1 according to the current height difference, and determine the second multipath identifier flag_vel according to the first antenna speed vh1 and the second antenna speed vh0 output by the inertial navigation device; obtain the current number of satellites GPS num, and determine the third multipath identifier flag_num according to the current number of satellites GPSnum and the satellite number threshold; determine the multipath recognition result according to the first multipath identifier flag_dh, the second multipath identifier flag_vel, and the third multipath identifier flag_num. The multipath recognition strategy for autonomous driving in the embodiments of the present application defines three different multipath recognition identifiers to redundantly recognize the multipath effect in the current scenario, greatly improving the accuracy of multipath effect recognition, and further improving the positioning robustness and safety of autonomous vehicles.

[0094] For ease of understanding, the above multi-path recognition strategy can be expressed in the following form:

[0095] If vb > 0.5 and (dh - max_dh) > 0.0, then flag_dh = 1;

[0096] Otherwise, flag_dh ≠ 1;

[0097] If abs((abs(vh1) - abs(vh0)) > 0.2, then flag_vel = 1;

[0098] Otherwise, flag_vel ≠ 1;

[0099] If GPS num < 12, flag_num = 1;

[0100] Otherwise, flag_num ≠ 1;

[0101] Among them, flag_dh = 1, flag_vel = 1, and flag_num = 1 all indicate entering the multi-path identification. 0.5 is the empirical value of the preset driving speed threshold, 0.2 is the empirical value of the preset antenna speed difference threshold, 12 is the preset satellite number threshold. Those skilled in the art can flexibly adjust their magnitudes according to the actual situation. abs() is the absolute value function.

[0102] When any one or more of flag_dh, flag_vel, and flag_num is equal to 1, it is considered that the multi-path effect is recognized. Otherwise, it is considered that there is no multi-path identification. This way can improve the accuracy of multi-path recognition on the one hand and avoid frequent switching between multi-path scenarios and normal scenarios on the other hand, thus ensuring the positioning robustness and safety of autonomous vehicles.

[0103] When the multi-path effect is recognized, it indicates that the current inertial RTK positioning data may not be accurate enough. Therefore, at this time, the matching score of the lidar positioning data can be further combined to re-determine the fourth fusion weight of the inertial RTK positioning data and the fourth fusion weight of the lidar positioning data.

[0104] In some embodiments of the present application, determining the fourth fusion weight of the inertial RTK positioning data and the fourth fusion weight of the lidar positioning data according to the multipath recognition result and the matching score of the lidar positioning data includes: when the multipath recognition result is entering the multipath scenario, determining the third initial weight of the lidar positioning data; according to the count of entering the multipath scenario and the matching score of the lidar positioning data, using the fourth weight adjustment strategy to adjust the third initial weight of the lidar positioning data to obtain the fourth fusion weight of the lidar positioning data; determining the fourth fusion weight of the inertial RTK positioning data according to the fourth fusion weight of the lidar positioning data.

[0105] To facilitate the understanding of the embodiments of the present application, the specific implementation logic of the embodiments of the present application is further illustrated here:

[0106]

[0107] Among them, multipath_flag represents the multipath identifier, multipath_flag = 1 represents entering the multipath identifier, 1.8 is the preset matching score threshold, which can be flexibly adjusted according to the actual situation, multipath_cnt represents the count of entering the multipath identifier, and here a timeout of 3s can be reserved for the subsequent multipath exit scenario to ensure the stability of the fusion positioning. Of course, the size of the timeout can also be flexibly adjusted according to actual requirements.

[0108] It can be seen that as multipath_flag increases, the value of K_LIO also increases. When multipath_flag = 4, that is, the multipath identifier has been recognized continuously for 4 times, indicating that the inertial RTK positioning data is greatly affected by multipath errors at this time and is difficult to be used for subsequent fusion positioning. Therefore, the calculated K_LIO = 1 and K_GPS = 0 at this time.

[0109] In some embodiments of the present application, determining the fourth fusion weight of the inertial RTK positioning data and the fourth fusion weight of the lidar positioning data according to the multipath recognition result and the matching score of the lidar positioning data includes: when the multipath recognition result is exiting the multipath scenario, determining the fourth initial weight of the lidar positioning data; according to the count of exiting the multipath scenario and the matching score of the lidar positioning data, using the fifth weight adjustment strategy to adjust the fourth initial weight of the lidar positioning data to obtain the fourth fusion weight of the lidar positioning data; determining the fourth fusion weight of the inertial RTK positioning data according to the fourth fusion weight of the lidar positioning data.

[0110] To facilitate the understanding of the embodiments of the present application, the specific implementation logic of the embodiments of the present application is further illustrated as follows:

[0111]

[0112]

[0113] It can be seen that in the multi-path exit scenario, it is not a complete exit, but the value of K_LIO gradually decays. When the multi-path exit scenario is just recognized, still taking the maximum value of multipath_flag as 4 and the timeout as 3 as an example, at this time multipath_cnt_max = multipath_cnt = 4 + 3 = 7, and K_LIO = 1 can be calculated. When multipath_cnt_max = 7 and multipath_cnt = 0, K_LIO = 0.8 can be calculated.

[0114] In an embodiment of the present application, when fusing the high-precision positioning data and lidar positioning data according to the fusion positioning strategy, the following logic can be adopted to implement:

[0115] double utm_x_cal_gps = utm_x_cal;

[0116] double utm_y_cal_gps = utm_y_cal;

[0117] double utm_x_cal_fusion = 0;

[0118] double utm_y_cal_fusion = 0;

[0119] utm_x_cal += lidar_dp.x();

[0120] utm_y_cal += lidar_dp.y();

[0121] utm_x_cal_fusion = utm_x_cal * K_LIO + utm_x_cal_gps * K_GPS;

[0122] utm_y_cal_fusion = utm_y_cal * K_LIO + utm_y_cal_gps * K_GPS;

[0123] utm_x_cal = utm_x_cal_fusion;

[0124] utm_y_cal = utm_y_cal_fusion;

[0125] Among them, utm_x_cal and utm_y_cal are the observed values in the x - direction and y - direction obtained after fusion. Finally, the extended Kalman filter fusion is performed using the observed values in the x - direction and y - direction after fusion to obtain the final fusion positioning result.

[0126] An embodiment of the present application also provides a fusion positioning device 200 for an autonomous driving vehicle, as Figure 2 shown in the structural schematic diagram of a fusion positioning device for an autonomous driving vehicle in an embodiment of the present application. The device 200 includes: an acquisition unit 210, a determination unit 220, and a fusion positioning unit 230, where:

[0127] The acquisition unit 210 is configured to acquire high - precision positioning data and the positioning status of the high - precision positioning data, as well as lidar positioning data and the matching score of the lidar positioning data. The positioning status of the high - precision positioning data is predicted based on a preset positioning status prediction model;

[0128] The determination unit 220 is configured to determine a fusion positioning strategy according to the positioning status of the high - precision positioning data and the matching score of the lidar positioning data. The fusion positioning strategy includes the fusion weight of the high - precision positioning data and the fusion weight of the lidar positioning data;

[0129] The fusion positioning unit 230 is configured to perform fusion processing on the high - precision positioning data and the lidar positioning data according to the fusion positioning strategy to obtain the fusion positioning result of the autonomous driving vehicle.

[0130] In some embodiments of the present application, the high - precision positioning data includes inertial navigation RTK positioning data. The determination unit 220 is specifically configured to: if the positioning status of the inertial navigation RTK positioning data is an untrusted state, determine a first fusion positioning strategy according to the matching score of the lidar positioning data; if the positioning status of the inertial navigation RTK positioning data is an uncertain state, determine a second fusion positioning strategy according to the matching score of the lidar positioning data; if the positioning status of the inertial navigation RTK positioning data is a trusted state, determine a third fusion positioning strategy according to the matching score of the lidar positioning data.

[0131] In some embodiments of the present application, the determining unit 220 is specifically configured to: if the matching score of the lidar positioning data is not less than a preset matching score threshold, determine a first initial weight of the lidar positioning data; based on the matching score of the lidar positioning data, use a first weight adjustment strategy to adjust the first initial weight of the lidar positioning data to obtain a first fusion weight of the lidar positioning data; determine a first fusion weight of the inertial RTK positioning data according to the first fusion weight of the lidar positioning data.

[0132] In some embodiments of the present application, the determining unit 220 is specifically configured to: if the matching score of the lidar positioning data is not less than a preset matching score threshold, determine a second initial weight of the lidar positioning data; based on the matching score of the lidar positioning data, use a second weight adjustment strategy to adjust the second initial weight of the lidar positioning data to obtain a second fusion weight of the lidar positioning data; determine a second fusion weight of the inertial RTK positioning data according to the second fusion weight of the lidar positioning data.

[0133] In some embodiments of the present application, the determining unit 220 is specifically configured to: determine an initial weight of the inertial RTK positioning data and a differential state of the inertial RTK positioning data; according to the differential state of the inertial RTK positioning data, use a third weight adjustment strategy to adjust the initial weight of the inertial RTK positioning data to obtain a third fusion weight of the inertial RTK positioning data; determine a third fusion weight of the lidar positioning data according to the third fusion weight of the inertial RTK positioning data.

[0134] In some embodiments of the present application, the differential state of the inertial RTK positioning data includes a fixed solution state and a non-fixed solution state. The determining unit 220 is specifically configured to: according to the differential state of the inertial RTK positioning data, determine a continuous count of the inertial RTK positioning data being in the non-fixed solution state; according to the continuous count of the inertial RTK positioning data being in the non-fixed solution state, adjust the initial weight of the inertial RTK positioning data to obtain a third fusion weight of the inertial RTK positioning data.

[0135] In some embodiments of the present application, the determining unit 220 is specifically configured to: determine the differential state of the inertial RTK positioning data; in the case where the differential state of the inertial RTK positioning data is the fixed solution state, perform multipath identification using a preset multipath identification strategy; determine a fourth fusion weight of the inertial RTK positioning data and a fourth fusion weight of the lidar positioning data according to the multipath identification result and the matching score of the lidar positioning data.

[0136] It can be understood that the above-mentioned fusion positioning device for autonomous vehicles can implement each step of the fusion positioning method for autonomous vehicles provided in the foregoing embodiments. The relevant explanations regarding the fusion positioning method for autonomous vehicles are applicable to the fusion positioning device for autonomous vehicles, and will not be elaborated here.

[0137] Figure 3 It is a schematic structural diagram of an electronic device according to an embodiment of the present application. Please refer to 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. Among them, the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory, such as at least one disk memory, etc. Of course, the electronic device may also include other hardware required for other services.

[0138] The processor, network interface, and memory can be interconnected through an internal bus, and the internal bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 3 only a bidirectional arrow is used in

[0139] The memory is used to store programs. Specifically, the program may include program code, and the program code includes computer operation instructions. The memory may include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0140] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a fusion positioning device for autonomous vehicles at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0141] Obtain high-precision positioning data and the positioning status of the high-precision positioning data, as well as lidar positioning data and the matching score of the lidar positioning data, where the positioning status of the high-precision positioning data is predicted based on a preset positioning status prediction model;

[0142] Determine a fusion positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data, where the fusion positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the lidar positioning data;

[0143] According to the fusion positioning strategy, perform fusion processing on the high-precision positioning data and the lidar positioning data to obtain a fusion positioning result of the autonomous driving vehicle.

[0144] The above as in this application Figure 1 The method executed by the fusion positioning device of the autonomous driving vehicle disclosed in the embodiments shown above can be applied to 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 the integrated logic circuit in the hardware of the processor or the instructions in the form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may 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 gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by the combination of the hardware and software modules in the decoding processor. The software module may be located in a mature storage medium 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 combines its hardware to complete the steps of the above method.

[0145] The electronic device can also execute Figure 1 the method executed by the fusion positioning device of the autonomous driving vehicle in Figure 1 and implement the functions of the fusion positioning device of the autonomous driving vehicle in the embodiments shown above. The embodiments of the present application will not be elaborated here.

[0146] Embodiments of the present application also propose a computer-readable storage medium storing one or more programs including instructions that, when executed by an electronic device including a plurality of application programs, enable the electronic device to execute Figure 1 the method executed by the fusion positioning device of the autonomous driving vehicle in the illustrated embodiment, and specifically used to execute:

[0147] Obtain high-precision positioning data and the positioning status of the high-precision positioning data, as well as lidar positioning data and the matching score of the lidar positioning data, where the positioning status of the high-precision positioning data is predicted based on a preset positioning status prediction model;

[0148] Determine a fusion positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data, where the fusion positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the lidar positioning data;

[0149] Fuse the high-precision positioning data and the lidar positioning data according to the fusion positioning strategy to obtain the fusion positioning result of the autonomous driving vehicle.

[0150] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can 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.

[0151] The present invention is described with reference to the 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 flow and / or block in the flowchart and / or block diagram, and the combination of flows 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 the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more of the flows or multiple flows and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0152] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction means that implements the function specified in one or more of the blocks and / or processes. Figure 1 one or more of the processes and / or blocks Figure 1 specified in one or more of the blocks and / or processes.

[0153] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the function specified in one or more of the blocks and / or processes. Figure 1 one or more of the processes and / or blocks Figure 1 specified in one or more of the blocks and / or processes.

[0154] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0155] Memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory such as read only memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

[0156] Computer-readable media includes both permanent and non-permanent, removable and non-removable media implemented by any method or technology for storage of information such as 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 technologies, compact disc read only memory (CD-ROM), digital versatile discs (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 transitory media such as modulated data signals and carrier waves.

[0157] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising said element.

[0158] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0159] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A fusion positioning method for an autonomous vehicle, wherein, The method includes: Obtaining high-precision positioning data and the positioning status of the high-precision positioning data, as well as lidar positioning data and the matching score of the lidar positioning data, where the positioning status of the high-precision positioning data is predicted based on a preset positioning status prediction model; Determining a fusion positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data, where the fusion positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the lidar positioning data; Performing fusion processing on the high-precision positioning data and the lidar positioning data according to the fusion positioning strategy to obtain the fusion positioning result of the autonomous vehicle; The high-precision positioning data includes inertial navigation RTK positioning data, and determining the fusion positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data includes: If the positioning status of the inertial navigation RTK positioning data is a credible status, determining a third fusion positioning strategy according to the matching score of the lidar positioning data; The if the positioning status of the inertial navigation RTK positioning data is a credible status, determining a third fusion positioning strategy according to the matching score of the lidar positioning data includes: Determining the initial weight of the inertial navigation RTK positioning data and the differential status of the inertial navigation RTK positioning data; Adjusting the initial weight of the inertial navigation RTK positioning data according to the differential status of the inertial navigation RTK positioning data by using a third weight adjustment strategy to obtain the third fusion weight of the inertial navigation RTK positioning data; Determining the third fusion weight of the lidar positioning data according to the third fusion weight of the inertial navigation RTK positioning data.

2. The method according to claim 1, wherein Determining the fusion positioning strategy according to the positioning status of the high-precision positioning data and the matching score of the lidar positioning data includes: If the positioning status of the inertial navigation RTK positioning data is an uncredible status, determining a first fusion positioning strategy according to the matching score of the lidar positioning data; If the positioning status of the inertial navigation RTK positioning data is an uncertain status, determining a second fusion positioning strategy according to the matching score of the lidar positioning data.

3. The method according to claim 2, wherein The if the positioning status of the inertial navigation RTK positioning data is an uncredible status, determining the fusion positioning strategy as the first fusion positioning strategy includes: If the matching score of the lidar positioning data is not less than a preset matching score threshold, determining the first initial weight of the lidar positioning data; Adjusting the first initial weight of the lidar positioning data based on the matching score of the lidar positioning data by using a first weight adjustment strategy to obtain the first fusion weight of the lidar positioning data; Determining the first fusion weight of the inertial navigation RTK positioning data according to the first fusion weight of the lidar positioning data.

4. The method according to claim 2, wherein The if the positioning status of the inertial navigation RTK positioning data is an uncertain status, determining a second fusion positioning strategy according to the matching score of the lidar positioning data includes: If the matching score of the lidar positioning data is not less than the preset matching score threshold, determine the second initial weight of the lidar positioning data; Based on the matching score of the lidar positioning data, use the second weight adjustment strategy to adjust the second initial weight of the lidar positioning data to obtain the second fusion weight of the lidar positioning data; Determine the second fusion weight of the inertial navigation RTK positioning data according to the second fusion weight of the lidar positioning data.

5. The method according to claim 1, wherein, The differential state of the inertial navigation RTK positioning data includes a fixed solution state and a non-fixed solution state. The adjusting the initial weight of the inertial navigation RTK positioning data by using a third weight adjustment strategy according to the differential state of the inertial navigation RTK positioning data to obtain the third fusion weight of the inertial navigation RTK positioning data includes: According to the differential state of the inertial navigation RTK positioning data, determine the continuous count of the inertial navigation RTK positioning data in the non-fixed solution state; According to the continuous count of the inertial navigation RTK positioning data in the non-fixed solution state, adjust the initial weight of the inertial navigation RTK positioning data to obtain the third fusion weight of the inertial navigation RTK positioning data.

6. The method according to claim 2, wherein The if the positioning state of the inertial navigation RTK positioning data is a credible state, determining a third fusion positioning strategy according to the matching score of the lidar positioning data includes: Determine the differential state of the inertial navigation RTK positioning data; In the case where the differential state of the inertial navigation RTK positioning data is in a fixed solution state, use a preset multipath identification strategy to perform multipath identification; According to the multipath identification result and the matching score of the lidar positioning data, determine the fourth fusion weight of the inertial navigation RTK positioning data and the fourth fusion weight of the lidar positioning data.

7. An integrated positioning device for an autonomous vehicle, wherein, The device includes: An acquisition unit, configured to acquire high-precision positioning data and the positioning state of the high-precision positioning data, as well as lidar positioning data and the matching score of the lidar positioning data, where the positioning state of the high-precision positioning data is predicted based on a preset positioning state prediction model; A determination unit, configured to determine a fusion positioning strategy according to the positioning state of the high-precision positioning data and the matching score of the lidar positioning data, where the fusion positioning strategy includes the fusion weight of the high-precision positioning data and the fusion weight of the lidar positioning data; A fusion positioning unit, configured to perform fusion processing on the high-precision positioning data and the lidar positioning data according to the fusion positioning strategy to obtain a fusion positioning result of the autonomous vehicle; The high-precision positioning data includes inertial navigation RTK positioning data, and the determination unit is specifically configured to: If the positioning state of the inertial navigation RTK positioning data is a credible state, determine a third fusion positioning strategy according to the matching score of the lidar positioning data; The determination unit is specifically configured to: Determine the initial weight of the inertial navigation RTK positioning data and the differential state of the inertial navigation RTK positioning data; According to the differential state of the inertial RTK positioning data, the initial weight of the inertial RTK positioning data is adjusted by using a third weight adjustment strategy to obtain the third fusion weight of the inertial RTK positioning data; Determine the third fusion weight of the lidar positioning data according to the third fusion weight of the inertial RTK positioning data.

8. An electronic device, comprising: A processor; And A memory arranged to store computer-executable instructions, which when executed cause the processor to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, the computer-readable storage medium stores one or more programs, which when executed by an electronic device including a plurality of application programs, cause the electronic device to execute the method according to any one of claims 1 to 6.

Images