Positioning information processing method, apparatus, device, and medium

Abstract

The application relates to a positioning information processing method, device, equipment and medium, belongs to the field of artificial intelligence, and further relates to electronic map technology. The method comprises the following steps: acquiring satellite information corresponding to a target object; the satellite information is provided by a target satellite for positioning the target object; information of the target object is collected by a sensor arranged on the target object to obtain sensing information of the target object; satellite features are obtained by performing feature extraction on the satellite information; sensing features are obtained by performing feature extraction on the sensing information; and the positioning accuracy of the target satellite is predicted according to the satellite features and the sensing features. The method can improve the accuracy of the positioning accuracy.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more particularly to electronic map technology, and especially to a method, apparatus, device and medium for processing location information. Background Technology

[0002] Satellite positioning systems are technologies that use satellites to accurately locate targets. Satellite positioning technology enables the simultaneous observation of multiple satellites at any given time and any point on Earth, facilitating navigation and positioning functions. It can be widely applied in various scenarios requiring positioning, such as guiding aircraft, ships, vehicles, and individuals safely and accurately along selected routes to their destinations on time. However, in many situations, satellite positioning cannot provide accurate positioning. For example, cloudy days, thick cloud cover, and lightning can affect satellite signals; surrounding buildings and dense forests can also interfere with signal reception; and locations such as basements, tunnels, and under overpasses can also affect signal reception. In these cases, the positioning information provided by the satellite may be inaccurate. Therefore, we need to determine the accuracy of satellite positioning to assess its reliability.

[0003] In traditional technologies, the accuracy information received from satellites is usually directly determined as the final positioning accuracy. However, the accuracy information directly provided by satellites usually cannot reflect the true positioning accuracy of the satellites, resulting in low accuracy of the obtained satellite positioning accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a positioning information processing method, device, equipment, and medium that can improve the accuracy of satellite positioning in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a location information processing method, the method comprising:

[0006] Obtain satellite information corresponding to the target object; the satellite information is provided by the target satellite that locates the target object;

[0007] Information about the target object is collected by sensors installed on the target object to obtain sensing information about the target object;

[0008] Feature extraction is performed on the satellite information to obtain satellite features;

[0009] Feature extraction is performed on the sensing information to obtain sensing features;

[0010] Based on the satellite characteristics and the sensing characteristics, the positioning accuracy of the target satellite is predicted.

[0011] Secondly, this application provides a positioning information processing device, the device comprising:

[0012] An acquisition module is used to acquire satellite information corresponding to a target object; the satellite information is provided by a target satellite that locates the target object; information about the target object is collected by sensors installed on the target object to obtain sensing information about the target object; feature extraction is performed on the satellite information to obtain satellite features; feature extraction is performed on the sensing information to obtain sensing features;

[0013] The prediction module is used to predict the positioning accuracy of the target satellite based on the satellite characteristics and the sensing characteristics.

[0014] In one embodiment, the apparatus further includes:

[0015] The comparison module is used to initially fuse the sensor positioning information and the satellite positioning information in the satellite information to obtain the initial fused positioning information of the target object; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information; the initial fused positioning information and the satellite positioning information are compared and analyzed to obtain positioning difference features;

[0016] The prediction module is also used to predict the positioning accuracy of the target satellite based on the satellite features, the sensing features, and the positioning difference features.

[0017] In one embodiment, the initial fused positioning information includes initial fused latitude and longitude information; the satellite positioning information includes satellite positioning latitude and longitude information; the positioning difference features include latitude and longitude difference features; the comparison module is further configured to perform latitude and longitude difference comparison analysis between the initial fused latitude and longitude information and the satellite positioning latitude and longitude information to obtain latitude and longitude difference features.

[0018] In one embodiment, the initial fused positioning information includes initial fused velocity information; the satellite positioning information includes satellite positioning velocity information; the positioning difference feature includes velocity difference feature; the comparison module is further configured to determine velocity difference information between the initial fused velocity information and the satellite positioning velocity information; and to determine the velocity difference feature based on the ratio of the velocity difference information to the initial fused velocity information.

[0019] In one embodiment, the satellite information includes satellite azimuth, satellite elevation, and satellite signal-to-noise ratio; the satellite features include satellite distribution features; the number of target satellites is multiple; the acquisition module is further configured to, for each target satellite, determine the position information of the target satellite based on its azimuth and elevation, and determine the target area where the target satellite is located based on the position information; for each target area, determine the satellite distribution sub-features corresponding to the target area based on the satellite signal-to-noise ratio and position information corresponding to satellites of the same constellation in the target area; the satellites of the same constellation are the target satellites located in the target area and belonging to the same constellation; and determine the satellite distribution features based on the satellite distribution sub-features corresponding to each target area.

[0020] In one embodiment, the target area includes target quadrants; the acquisition module is further configured to, for each target satellite, project the target satellite onto a target plane according to the satellite azimuth and elevation angles of the target satellite to obtain the position information of the target satellite on the target plane; the target plane includes multiple quadrants; and determine the target quadrant in the target plane where the target satellite is located based on the position information of the target satellite on the target plane.

[0021] In one embodiment, the number of target satellites is multiple; the satellite information includes at least one of satellite azimuth angle, satellite elevation angle, or satellite signal-to-noise ratio (SNR); the satellite features include at least one of satellite azimuth angle features, mean SNR features, median SNR features, or SNR percentage features; wherein, the satellite azimuth angle features are determined based on the satellite azimuth angles corresponding to each of the target satellites; the mean SNR features are determined based on the mean SNRs corresponding to satellites at large elevation angles; the large elevation angle satellites are target satellites whose elevation angles are greater than a preset elevation angle; the median SNR features are determined by taking the median of the SNRs corresponding to each of the target satellites; the SNR percentage features are determined based on the ratio of the number of target satellites to the total number of satellites, wherein the number of target satellites is the number of target satellites whose SNRs are less than a preset SNR; and the total number of satellites is the number of target satellites.

[0022] In one embodiment, the prediction module is further configured to perform availability prediction on the satellite positioning information based on the satellite features and the sensing features; and if the availability prediction result indicates that the satellite positioning information is available, perform the step of predicting the positioning accuracy of the target satellite based on the satellite features and the sensing features.

[0023] In one embodiment, the availability prediction result is obtained by an availability determination model trained on a first training sample; the first training sample includes multiple sets of positive samples and multiple sets of negative samples; wherein, each set of positive samples includes positive sample satellite information and corresponding positive sample sensor information; the positive sample satellite positioning information in the positive sample satellite information satisfies a long-distance difference condition with the reference positioning information; the positive sample sensor information is sensor information collected within a first target time period; the first target time period is the time period from the last time the satellite information was collected to the time point of the positive sample satellite information collection; each set of negative samples includes negative sample satellite information and corresponding negative sample sensor information; the negative sample satellite positioning information in the negative sample satellite information satisfies a first short-distance difference condition with the reference positioning information; the negative sample sensor information is sensor information collected within a second target time period; the second target time period is the time period from the last time the satellite information was collected to the time point of the negative sample satellite information collection.

[0024] In one embodiment, each group of positive samples further includes positive sample fusion positioning information; the positive sample fusion positioning information is obtained by initial fusion of positive sample sensor positioning information and positive sample satellite positioning information; the positive sample sensor positioning information is obtained by positioning prediction of the target object based on the positive sample sensor information; each group of negative samples further includes negative sample fusion positioning information; the negative sample fusion positioning information is obtained by initial fusion of negative sample sensor positioning information and negative sample satellite positioning information; the negative sample sensor positioning information is obtained by positioning prediction of the target object based on the negative sample sensor information.

[0025] In one embodiment, the positioning accuracy is predicted by a trained accuracy prediction model; the device further includes:

[0026] A training module is used to acquire a second training sample; the second training sample includes multiple sets of target samples; each set of target samples includes sample satellite information and corresponding sample sensor information; the sample satellite positioning information in the sample satellite information satisfies a second near-range difference condition with the reference positioning information; the sample sensor information is sensor information collected within a third target time period; the third target time period is the time period from the last time the satellite information was collected to the time point of the sample satellite information collection; the accuracy prediction model to be trained predicts the accuracy of the target satellite positioning based on the second training sample to obtain the predicted positioning accuracy; the model parameters of the accuracy prediction model to be trained are updated according to the difference between the predicted positioning accuracy and the reference positioning accuracy to iteratively train the accuracy prediction model to be trained; the reference positioning accuracy is determined based on the difference between the sample satellite positioning information and the reference positioning information.

[0027] In one embodiment, each group of target samples further includes sample fusion positioning information; the sample fusion positioning information is obtained by initial fusion of sample sensor positioning information and sample satellite positioning information; the sample sensor positioning information is obtained by positioning prediction of the target object based on the sample sensor information.

[0028] In one embodiment, the positioning accuracy is used to fuse the sensor positioning information and the satellite positioning information in the satellite information; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information.

[0029] In one embodiment, the target object is an autonomous vehicle in an autonomous driving scenario; the sensor is a vehicle sensor installed on the autonomous vehicle; the vehicle sensor includes at least one of a vehicle speed sensor, a vehicle vision sensor, or a vehicle inertial sensor.

[0030] Thirdly, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the various method embodiments of this application.

[0031] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the various method embodiments of this application.

[0032] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the various method embodiments of this application.

[0033] The aforementioned positioning information processing method, apparatus, equipment, medium, and computer program product acquire satellite information corresponding to a target object; this satellite information is provided by the target satellite used to locate the target object; information is collected from the target object using sensors installed on it to obtain sensor information specific to the target object; feature extraction is performed on the satellite information to obtain satellite features; feature extraction is also performed on the sensor information to obtain sensor features; and the positioning accuracy of the target satellite is predicted based on the satellite features and sensor features. Since the determination of satellite positioning accuracy considers both satellite information from the target satellite and sensor information from the sensors, this application achieves more accurate satellite positioning accuracy compared to the traditional method of directly determining the final positioning accuracy from the accuracy information received from the satellite. Attached Figure Description

[0034] Figure 1 This is an application environment diagram of the location information processing method in one embodiment;

[0035] Figure 2 This is a flowchart illustrating a location information processing method in one embodiment;

[0036] Figure 3 This is a schematic diagram illustrating the process of determining satellite distribution characteristics in one embodiment;

[0037] Figure 4 This is a schematic diagram showing the distribution of target satellites in each quadrant of the target plane in one embodiment;

[0038] Figure 5 This is a flowchart illustrating the location information processing method in another embodiment;

[0039] Figure 6 This is a schematic diagram comparing the predicted error, the actual error, and the error given by the satellite in one embodiment of the application.

[0040] Figure 7 This is a flowchart illustrating the location information processing method in yet another embodiment;

[0041] Figure 8 This is a structural block diagram of a positioning information processing device in one embodiment;

[0042] Figure 9 This is a structural block diagram of the positioning information processing device in another embodiment;

[0043] Figure 10 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0045] The location information processing method provided in this application can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on other servers. Terminal 102 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, cloud security, host security and other network security services, CDN, and basic cloud computing services such as big data and artificial intelligence platforms. Terminal 102 and server 104 can be directly or indirectly connected via wired or wireless communication, which is not limited herein.

[0046] Terminal 102 can acquire satellite information corresponding to the target object. This satellite information is provided by the target satellite used to locate the target object, and is obtained by collecting information from sensors installed on the target object. Terminal 102 can extract features from the satellite information to obtain satellite features. Terminal 102 can also extract features from the sensor information to obtain sensor features. Based on the satellite features and sensor features, Terminal 102 can predict the positioning accuracy of the target satellite.

[0047] It is understood that terminal 102 can use the predicted positioning accuracy as a fusion reference parameter to fuse the sensor positioning information and the satellite positioning information from the satellite information to obtain the fused positioning information of the target object. The sensor positioning information is obtained by predicting the positioning of the target object based on sensor information. It is also understood that terminal 102 can send the predicted positioning accuracy to server 104, and server 104 can use the positioning accuracy as a fusion reference parameter to fuse the sensor positioning information and the satellite positioning information from the satellite information to obtain the fused positioning information of the target object, and then send the fused positioning information to terminal 102. This embodiment does not limit this; it is understood that… Figure 1 The application scenarios shown are for illustrative purposes only and are not limited to these.

[0048] It should be noted that some of the positioning information processing methods in this application utilize artificial intelligence technology. For example, predicting the availability of satellite positioning information is achieved using artificial intelligence technology, and the positioning accuracy of the target satellite is also predicted using artificial intelligence technology.

[0049] In one embodiment, such as Figure 2 As shown, a location information processing method is provided. This embodiment applies this method to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps:

[0050] Step 202: Obtain satellite information corresponding to the target object; the satellite information is provided by the target satellite used to locate the target object.

[0051] In this context, the object is a mobile entity on which the terminal can be deployed. For example, the object could be a vehicle or a person. It can be understood that if the object is a vehicle, the terminal can be deployed on the vehicle; if the object is a person, the terminal can be carried by the person. Satellite information is provided by target satellites used to locate the target object.

[0052] Specifically, the number and range of satellites that the terminal can receive are fixed. By using the satellites that the terminal can receive as target satellites, the terminal can obtain satellite information provided by at least one target satellite for locating the target object.

[0053] Step 204: Collect information about the target object by using sensors installed on the target object to obtain sensing information about the target object.

[0054] Specifically, the target object is equipped with sensors and a terminal. The terminal can collect information about the target object through the sensors installed on the target object to obtain sensing information about the target object.

[0055] Step 206: Extract features from satellite information to obtain satellite features.

[0056] Among them, satellite features are features obtained by extracting features from satellite information.

[0057] In one embodiment, satellite information includes at least one of satellite azimuth, satellite elevation, and satellite signal-to-noise ratio. The terminal can extract features from at least one of satellite azimuth, satellite elevation, or satellite signal-to-noise ratio to obtain satellite features.

[0058] In one embodiment, the satellite information further includes satellite positioning information, which may include at least one of satellite positioning latitude and longitude information, satellite positioning velocity information, satellite positioning altitude information, satellite positioning time, satellite positioning accuracy factor, or satellite positioning status. The terminal can directly use the satellite positioning information as a satellite feature.

[0059] Step 208: Extract features from the sensing information to obtain sensing features.

[0060] Among them, sensing features are features obtained by extracting features from sensing information.

[0061] In one embodiment, the terminal can directly use the sensing information collected by the sensor as sensing features.

[0062] In one embodiment, the sensor disposed on the target object includes at least one of a velocity sensor, a vision sensor, or an inertial sensor. It is understood that if the sensor includes a velocity sensor, the sensing information includes velocity information acquired by the velocity sensor. If the sensor includes a vision sensor, the sensing information includes image information acquired by the vision sensor. If the sensor includes an inertial sensor, the sensing information includes acceleration and angular velocity information acquired by the inertial sensor.

[0063] In one embodiment, the terminal can acquire map information, which may include at least one of lane information, lane line information, or lane centerline information. If the sensor includes a vision sensor, the terminal can correct the image information acquired by the vision sensor based on the map information to obtain corrected image information.

[0064] Step 210: Based on satellite characteristics and sensor characteristics, predict the positioning accuracy of the target satellite.

[0065] Among them, positioning accuracy is used to characterize the precision of satellite positioning information in satellite information.

[0066] Specifically, the terminal can predict the positioning accuracy of the target satellite based on satellite and sensor characteristics. The terminal can also predict the positioning of the target object based on sensor information to obtain sensor positioning information. Furthermore, the terminal can use the positioning accuracy as a fusion reference parameter to fuse the sensor positioning information and the satellite positioning information from the satellite data to obtain the fused positioning information of the target object.

[0067] In one embodiment, the predicted positioning accuracy can be used to fuse sensor positioning information and satellite positioning information. The sensor positioning information is obtained by predicting the location of the target object based on sensor data. It can be understood that the fused positioning information serves as the final positioning result for the target object. Thus, based on more accurate satellite positioning accuracy, more accurate fused positioning information for the target object can be obtained, improving the positioning accuracy for the target object.

[0068] In one embodiment, the terminal can use positioning accuracy as the fusion weight of satellite positioning information to perform weighted fusion of the sensor positioning information and the satellite positioning information in the satellite information to obtain the fused positioning information of the target object.

[0069] In the aforementioned positioning information processing method, satellite information corresponding to the target object is acquired; this satellite information is provided by the target satellite used for positioning the target object; sensors installed on the target object collect information from the target object to obtain sensor information for the target object; feature extraction is performed on the satellite information to obtain satellite features. Feature extraction is also performed on the sensor information to obtain sensor features. Based on the satellite features and sensor features, the positioning accuracy of the target satellite is predicted. Since the determination of satellite positioning accuracy considers both satellite information from the target satellite and sensor information from the sensors, compared to the traditional method of directly determining the final positioning accuracy from the accuracy information received from the satellite, this application can obtain more accurate satellite positioning accuracy. Therefore, based on this more accurate satellite positioning accuracy, more accurate fused positioning information for the target object can be obtained, improving the positioning accuracy for the target object.

[0070] In one embodiment, the method further includes: initially fusing the sensor positioning information and the satellite positioning information in the satellite information to obtain the initial fused positioning information of the target object; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information; performing a difference comparison analysis on the initial fused positioning information and the satellite positioning information to obtain positioning difference features; and predicting the positioning accuracy of the target satellite based on the satellite features and the sensor features, including: predicting the positioning accuracy of the target satellite based on the satellite features, the sensor features, and the positioning difference features.

[0071] The initial fused positioning information is the positioning information obtained by initially fusing sensor positioning information and satellite positioning information. It can be understood that the initial fused positioning information is only an intermediate positioning result in the target object positioning process, not the final positioning result. It can also be understood that based on the initial fused positioning information, a more accurate positioning accuracy of the target satellite can be predicted, thus obtaining a more accurate fused positioning information for the target object. This fused positioning information is the final positioning result for the target object. Positioning difference features are features obtained by comparing and analyzing the differences between the initial fused positioning information and the satellite positioning information, used to characterize the magnitude of the difference between the initial fused positioning information and the satellite positioning information.

[0072] Specifically, the terminal can determine the initial fusion weights corresponding to the sensor positioning information and the satellite positioning information respectively, and based on the initial fusion weights, perform initial fusion of the satellite positioning information from the sensor positioning information and the satellite information to obtain the initial fused positioning information of the target object. Furthermore, the terminal can perform difference comparison analysis between the initial fused positioning information and the satellite positioning information to obtain positioning difference features, and predict the positioning accuracy of the target satellite based on the satellite features, sensor features, and positioning difference features.

[0073] In one embodiment, the satellite information includes a satellite positioning accuracy factor, which the terminal can directly use as the initial fusion weight of the satellite positioning information. Based on the initial fusion weight of the satellite positioning information, the terminal can determine the initial fusion weight of the sensor positioning information. Furthermore, based on the initial fusion weights corresponding to the sensor positioning information and the satellite positioning information respectively, the terminal can perform initial fusion of the satellite positioning information from the sensor positioning information and the satellite information to obtain the initial fused positioning information of the target object.

[0074] In one embodiment, the initial fused positioning information includes initial fused acceleration information; the satellite positioning information includes satellite positioning acceleration information; and the positioning difference features include acceleration difference features. The positioning difference features are obtained by performing a difference comparison analysis between the initial fused positioning information and the satellite positioning information, including: performing an acceleration difference comparison analysis between the initial fused acceleration information and the satellite positioning acceleration information to obtain acceleration difference features. Wherein, the initial fused acceleration information is the information in the initial fused positioning information used to represent the acceleration of the target object, the satellite positioning acceleration information is the information in the satellite positioning information used to represent the acceleration of the target object, and the acceleration difference features are the difference features in the positioning difference features used to represent acceleration.

[0075] In one embodiment, the initial fused positioning information includes initial fused angular velocity information; the satellite positioning information includes satellite positioning angular velocity information; and the positioning difference features include angular velocity difference features. The positioning difference features are obtained by performing a difference comparison analysis between the initial fused positioning information and the satellite positioning information, including: performing an angular velocity difference comparison analysis between the initial fused angular velocity information and the satellite positioning angular velocity information to obtain angular velocity difference features. Wherein, the initial fused angular velocity information is the information in the initial fused positioning information used to represent the angular velocity of the target object, the satellite positioning angular velocity information is the information in the satellite positioning information used to represent the angular velocity of the target object, and the angular velocity difference features are the difference features in the positioning difference features used to represent the angular velocity.

[0076] In the above embodiments, the positioning accuracy of the target satellite can be further improved by jointly predicting the positioning accuracy of the target satellite through satellite features, sensor features, and positioning difference features obtained by comparing and analyzing the differences between the initial fused positioning information and the satellite positioning information.

[0077] In one embodiment, the initial fused positioning information includes initial fused latitude and longitude information; the satellite positioning information includes satellite positioning latitude and longitude information; the positioning difference features include latitude and longitude difference features; and the positioning difference analysis of the initial fused positioning information and the satellite positioning information to obtain the positioning difference features includes: performing latitude and longitude difference analysis of the initial fused latitude and longitude information and the satellite positioning latitude and longitude information to obtain the latitude and longitude difference features.

[0078] Among them, the initial fused latitude and longitude information is the information used to represent the latitude and longitude of the target object in the initial fused positioning information; the satellite positioning latitude and longitude information is the information used to represent the latitude and longitude of the target object in the satellite positioning information; and the latitude and longitude difference features are the difference features used to represent the latitude and longitude in the positioning difference features.

[0079] Specifically, the terminal can perform a latitude and longitude difference analysis by comparing the initial fused latitude and longitude information with the satellite positioning latitude and longitude information to obtain latitude and longitude difference characteristics. In other words, the terminal can determine the distance between the location corresponding to the initial fused latitude and longitude information and the location corresponding to the satellite positioning latitude and longitude information based on the initial fused latitude and longitude information and the satellite positioning latitude and longitude information; this distance is the latitude and longitude difference characteristic.

[0080] In one embodiment, if the satellite positioning latitude and longitude at the same time are lat1, lon1, and the initial fused positioning latitude and longitude are lat2, lon2, then the difference in latitude and longitude between the initial fused positioning information and the satellite positioning information, i.e., the distance between the initial fused latitude and longitude and the satellite positioning latitude and longitude, can be calculated using the following formula:

[0081] dist_diff=haversine(lon1,lat1,lon2,lat2);

[0082] Here, haversine() represents the semi-sine function, and dist_diff represents the distance between the initial fused latitude and longitude and the satellite positioning latitude and longitude.

[0083] In the above embodiments, by comparing and analyzing the latitude and longitude differences between the initial fused latitude and longitude information and the satellite positioning latitude and longitude information, latitude and longitude difference features are obtained. Then, by combining satellite features, sensor features, and latitude and longitude difference features, the positioning accuracy of the target satellite can be predicted, which can further improve the accuracy of satellite positioning.

[0084] In one embodiment, the initial fused positioning information includes initial fused velocity information; the satellite positioning information includes satellite positioning velocity information; the positioning difference feature includes velocity difference feature; and the initial fused positioning information and the satellite positioning information are compared and analyzed to obtain the positioning difference feature, including: determining the velocity difference information between the initial fused velocity information and the satellite positioning velocity information; and determining the velocity difference feature based on the ratio of the velocity difference information to the initial fused velocity information.

[0085] Among them, the initial fused velocity information is the information used to represent the velocity of the target object in the initial fused positioning information, the satellite positioning velocity information is the information used to represent the velocity of the target object in the satellite positioning information, and the velocity difference feature is the difference feature used to represent the velocity in the positioning difference feature.

[0086] In one embodiment, the terminal can compare the initial fused velocity information with the satellite positioning velocity information to determine the velocity difference information between the initial fused velocity information and the satellite positioning velocity information, and determine the velocity difference feature based on the ratio of the velocity difference information to the initial fused velocity information.

[0087] In one embodiment, if the satellite positioning velocity is V1 and the initial fusion velocity is V2, then the velocity difference characteristic can be calculated using the following formula:

[0088] diff_V = abs(V1-V2) / V2

[0089] Where abs() represents the absolute value function, (V1-V2) represents the velocity difference information between the initial fused velocity information and the satellite positioning velocity information, (V1-V2) / V2 represents the ratio of the velocity difference information to the initial fused velocity information, and diff_V represents the velocity difference feature.

[0090] In the above embodiments, by determining the velocity difference information between the initial fused velocity information and the satellite positioning velocity information, and by determining the velocity difference feature based on the ratio of the velocity difference information to the initial fused velocity information, the positioning accuracy of the target satellite can be predicted by combining satellite features, sensor features, and velocity difference features, thereby further improving the accuracy of satellite positioning.

[0091] In one embodiment, satellite information includes satellite azimuth, satellite elevation, and satellite signal-to-noise ratio; satellite characteristics include satellite distribution characteristics; the number of target satellites is multiple; such as Figure 3 As shown, feature extraction is performed on satellite information to obtain satellite features, including:

[0092] Step 302: For each target satellite, determine the position information of the target satellite based on its azimuth and elevation angles, and determine the target area where the target satellite is located based on the position information.

[0093] Step 304: For each target area, determine the satellite distribution sub-features corresponding to the target area based on the satellite signal-to-noise ratio and position information of the satellites in the same constellation in the target area; satellites in the same constellation are target satellites located in the target area and belonging to the same constellation.

[0094] Step 306: Determine the satellite distribution characteristics based on the satellite distribution sub-features corresponding to each target area.

[0095] Among them, the satellite distribution sub-features are the distribution characteristics corresponding to each target area separately. The satellite distribution characteristics are the distribution characteristics corresponding to all target areas together. A constellation includes multiple satellites; it can be understood that a constellation is a collection of satellites, a satellite network composed of multiple satellites configured in a preset manner.

[0096] Specifically, for each target satellite, the terminal can determine its position information based on its azimuth and elevation angles, and then determine the target area where the target satellite is located. For each target area, which may include target satellites from multiple constellations, the terminal can determine the satellite distribution sub-features corresponding to the target area and the constellation based on the signal-to-noise ratio and position information of each target satellite belonging to that constellation. Furthermore, for each constellation, the terminal can determine the satellite distribution features corresponding to that constellation based on the satellite distribution sub-features of each target area within that constellation. The terminal can then stitch together the satellite distribution features corresponding to multiple constellations to obtain the satellite distribution features corresponding to the target object.

[0097] For example, the terminal can support receiving signals from constellations including Constellation 1, Constellation 2, and Constellation 3. The number of target areas is 8. The satellite distribution features corresponding to Constellation 1, Constellation 2, and Constellation 3 are each 8-dimensional feature vectors. This means that each target area corresponds to a one-dimensional feature vector. The terminal can concatenate the 8-dimensional satellite distribution feature vectors corresponding to Constellation 1, Constellation 2, and Constellation 3 respectively to obtain a 24-dimensional satellite distribution feature vector corresponding to the target object.

[0098] In one embodiment, the satellite information also includes the satellite identifier of the target satellite. There is a correspondence between the satellite identifier and the constellation; that is, the satellite identifier of each target satellite included in each constellation corresponds to that constellation. The correspondence between satellite identifiers and constellations is as follows:

[0099] prn<=32:GPS

[0100] prn>32&prn<=64:SBAS,

[0101] prn>64&prn<=96:GLONASS

[0102] prn>=193&prn<=195:QZSS,

[0103] prn>=201&prn<=235:BEIDOU,

[0104] Here, prn represents the satellite identifier, and GPS, SBAS, GLONASS, QZSS, and BEIDOU represent different constellations. For example, target satellites with a satellite identifier prn less than or equal to 32 belong to the GPS constellation, while target satellites with a satellite identifier prn greater than 32 and less than or equal to 64 belong to the SBAS constellation.

[0105] In the above embodiments, by determining the target area where each target satellite is located and determining the satellite distribution sub-features corresponding to each target area, and then determining the satellite distribution features based on the satellite distribution sub-features corresponding to each target area, the accuracy of the obtained satellite distribution features can be improved.

[0106] In one embodiment, the target area includes target quadrants; for each target satellite, the position information of the target satellite is determined based on its azimuth and elevation angles, and the target area where the target satellite is located is determined based on the position information, including: for each target satellite, projecting the target satellite onto a target plane based on its azimuth and elevation angles to obtain the position information of the target satellite on the target plane; the target plane includes multiple quadrants; and determining the target quadrant in the target plane where the target satellite is located based on the position information of the target satellite on the target plane.

[0107] Specifically, for each target satellite, the terminal can project the target satellite onto the target plane based on its azimuth and elevation angles to obtain its position information on the target plane. The target plane comprises multiple quadrants, and the terminal can determine the target quadrant within that plane based on the satellite's position information. For each target quadrant, the terminal determines the satellite distribution sub-features corresponding to that quadrant based on the signal-to-noise ratio and position information of each target satellite belonging to the same constellation within that quadrant; and based on these satellite distribution sub-features, it determines the satellite distribution characteristics.

[0108] In one embodiment, such as Figure 4 As shown, the target plane comprises eight quadrants. For each target satellite, the terminal can project the target satellite onto the target plane based on its azimuth and elevation angles to obtain its position information on the target plane. Furthermore, the terminal can determine the target quadrant within the target plane based on this position information. For example, two target satellites are located in quadrants 1, 5, and 8; three target satellites are located in quadrants 2, 3, and 4; and one target satellite is located in quadrants 6 and 7.

[0109] In one embodiment, the position information of the target satellite on the target plane can be understood to include the coordinates of the target satellite on the target plane, i.e., (px, py), which can be calculated using the following formula:

[0110]

[0111] Where θ represents the satellite azimuth angle, Indicates the satellite's pitch angle.

[0112] In one embodiment, the number of target quadrants is 8. For each constellation, such as the BeiDou constellation, and for each target quadrant, such as the first quadrant, the satellite distribution sub-features corresponding to the first quadrant can be calculated using the following formula:

[0113]

[0114] dist = sqrt(px*px + py*py)

[0115] Where sqrt() represents the square root function, dist represents the coordinate processing information calculated based on the coordinates of the target satellite, k represents the number of target satellites belonging to the BeiDou constellation in the first quadrant, and SNR... i This represents the signal-to-noise ratio of the i-th target satellite. This indicates the satellite distribution sub-features of the BeiDou constellation in the first quadrant.

[0116] It is understandable that the terminal can stitch together the satellite distribution features corresponding to the eight quadrants of the BeiDou constellation to obtain the satellite distribution features corresponding to the BeiDou constellation:

[0117]

[0118] In the above embodiments, by projecting each target satellite onto the target plane, the position information of each target satellite on the target plane is obtained. Based on the position information of the target satellites on the target plane, the target quadrant in which each target satellite is located on the target plane is determined. Furthermore, based on the satellite distribution sub-features corresponding to each target quadrant, the satellite distribution features are determined, which can further improve the accuracy of the obtained satellite distribution features.

[0119] In one embodiment, the number of target satellites is multiple; satellite information includes at least one of satellite azimuth angle, satellite elevation angle, or satellite signal-to-noise ratio (SNR); satellite features include at least one of satellite azimuth angle features, mean SNR features, median SNR features, or SNR percentage features; wherein, the satellite azimuth angle features are determined based on the satellite azimuth angle corresponding to each target satellite; the mean SNR features are determined based on the mean SNR corresponding to satellites at large elevation angles; large elevation angle satellites are target satellites whose elevation angle is greater than a preset elevation angle; the median SNR features are determined by taking the median of the SNR corresponding to each target satellite; the SNR percentage features are determined based on the ratio of the number of targets to the total number of satellites; the number of targets is the number of target satellites whose SNR is less than a preset SNR; and the total number of satellites is the number of target satellites.

[0120] In one embodiment, the terminal can determine the satellite azimuth characteristics based on the satellite azimuth angles corresponding to each target satellite. Specifically, the terminal can calculate the standard deviation of the satellite azimuth angles corresponding to each target satellite to obtain the satellite azimuth characteristics.

[0121] For example, if the azimuth angles corresponding to each target satellite are θ1, θ2, θ3, ..., θ n , among them, θ1<<θ2<<θ3<<…<<θ n The satellite azimuth characteristics can be calculated using the following formula:

[0122] diff_azimuth_std=std(diff_angle1,diff_angle2,…,diff_anglen-1),

[0123] diff_angle1 = θ2 - θ1,

[0124] diff_angle2 = θ3 - θ2,

[0125] …

[0126] diff_anglen-1=θ n -θ n-1 .

[0127] Where std() represents the standard deviation function, diff_angle represents the difference in satellite azimuth angles, diff_azimuth_std represents the satellite azimuth characteristic, and n is a constant.

[0128] In one embodiment, the terminal can calculate the average signal-to-noise ratio (SNR) of satellites at various elevation angles and use the calculated average as the SNR average feature.

[0129] For example, if the elevation angles of each target satellite are respectively With a preset elevation angle of 60°, the terminal can identify satellites with elevation angles greater than 60° as high-elevation-angle satellites. The signal-to-noise ratios (SNRs) corresponding to these high-elevation-angle satellites are SNR1, SNR2, SNR3, ..., SNR... m The mean signal-to-noise ratio characteristic can then be calculated using the following formula:

[0130] high_elevation_snr_mean=mean(SNR1, SNR2, SNR3,…,SNR m )

[0131] Where mean() is the mean function, high_elevation_snr_mean is the signal-to-noise ratio mean feature, n and m are constants, and n is greater than or equal to m.

[0132] In one embodiment, the terminal can calculate the median signal-to-noise ratio (SNR) of each target satellite and use the calculated median as the median SNR feature.

[0133] For example, if the signal-to-noise ratios (SNRs) of each target satellite are SNR1, SNR2, SNR3, ..., SNRn, then the median SNR characteristic can be calculated using the following formula:

[0134] median_snr_feature=meidan(SNR1, SNR2, SNR3,…,SNRn)

[0135] Where meidan() represents the median function, median_snr_feature represents the median signal-to-noise ratio feature, and n is a constant.

[0136] In one embodiment, the terminal can determine the ratio of the number of targets to the total number of satellites and use the determined ratio as a signal-to-noise ratio feature.

[0137] In one embodiment, the preset signal-to-noise ratio (SNR) includes a first preset SNR and a second preset SNR. The target quantity includes a first target quantity with a satellite SNR less than the first preset SNR, and a second target quantity with a satellite SNR less than the second preset SNR. The terminal can determine the ratio of the first target quantity to the total number of satellites and use the determined ratio as an SNR proportion feature, and determine the ratio of the second target quantity to the total number of satellites and use the determined ratio as an SNR proportion feature.

[0138] For example, the first preset signal-to-noise ratio (SNR) is 30 dB, the second preset SNR is 40 dB, and the number of targets includes the number of first targets (N1) with a satellite SNR less than 30 dB and the number of second targets (N2) with a satellite SNR less than 40 dB. If the total number of satellites is N, the SNR proportion characteristic can be calculated using the following formula:

[0139] percentage_less_than_30db=N1 / N, percentage_less_than_40db=N2 / N

[0140] Where percentage_less_than_30db represents the signal-to-noise ratio percentage characteristic corresponding to 30db, and percentage_less_than_40db represents the signal-to-noise ratio percentage characteristic corresponding to 40db.

[0141] In the above embodiments, by acquiring various satellite information and obtaining various types of satellite features from the various satellite information, and then predicting the positioning accuracy of the target satellite based on the sensing features and relatively rich satellite features, the prediction accuracy of the target satellite's positioning accuracy can be further improved.

[0142] In one embodiment, the method further includes: predicting the availability of satellite positioning information based on satellite characteristics and sensing characteristics; and, if the availability prediction result indicates that the satellite positioning information is available, performing the step of predicting the positioning accuracy of the target satellite based on satellite characteristics and sensing characteristics.

[0143] Specifically, the terminal can predict the availability of satellite positioning information based on satellite and sensor characteristics, obtaining an availability prediction result. This result includes whether the satellite positioning information is available or unavailable. It can be understood that available satellite positioning information indicates relatively accurate positioning, while unavailable information indicates less accurate positioning. When the availability prediction result indicates that the satellite positioning information is available, the terminal can predict the positioning accuracy of the target satellite based on satellite and sensor characteristics, and use this accuracy as a fusion reference parameter to fuse the satellite positioning information from the sensor and satellite data to obtain the fused positioning information of the target object. When the availability prediction result indicates that the satellite positioning information is unavailable, the terminal can directly determine the sensor positioning information as the fused positioning information of the target object.

[0144] In the above embodiments, the availability of satellite positioning information is first predicted based on satellite characteristics and sensor characteristics to determine whether the satellite positioning information is usable, i.e., to preliminarily determine the positioning accuracy of the satellite positioning information. Furthermore, if the availability prediction result indicates that the satellite positioning information is usable, it indicates that the satellite positioning information is relatively accurate. Then, based on satellite characteristics and sensor characteristics, the positioning accuracy of the target satellite is predicted, which can further improve the prediction accuracy of the target satellite's positioning accuracy.

[0145] In one embodiment, the availability prediction result is obtained by an availability determination model trained on a first training sample; the first training sample includes multiple sets of positive samples and multiple sets of negative samples; wherein, each set of positive samples includes positive sample satellite information and corresponding positive sample sensor information; the positive sample satellite positioning information in the positive sample satellite information satisfies the long-distance difference condition with the reference positioning information; the positive sample sensor information is sensor information collected within a first target time period; the first target time period is the time period from the last time the satellite information was collected to the time point of the positive sample satellite information collection; each set of negative samples includes negative sample satellite information and corresponding negative sample sensor information; the negative sample satellite positioning information in the negative sample satellite information satisfies the first short-distance difference condition with the reference positioning information; the negative sample sensor information is sensor information collected within a second target time period; the second target time period is the time period from the last time the satellite information was collected to the time point of the negative sample satellite information collection.

[0146] Among them, the reference positioning information is the positioning information for the target object provided by the true value device set on the target object.

[0147] In one embodiment, the long-distance difference condition can be that the distance between the positive sample satellite positioning information and the reference positioning information is greater than a first preset distance threshold. For example, if the first preset distance threshold is 2 meters, the long-distance difference condition can be that the distance between the positive sample satellite positioning information and the reference positioning information is greater than 2 meters. The first short-distance difference condition can be that the distance between the positive sample satellite positioning information and the reference positioning information is less than a second preset distance threshold. For example, if the second preset distance threshold is 1 meter, the first short-distance difference condition can be that the distance between the positive sample satellite positioning information and the reference positioning information is less than 1 meter.

[0148] In one embodiment, the long-distance difference condition may also be that the distance between the positive sample satellite positioning information and the reference positioning information falls within a first preset distance range. The first short-distance difference condition may also be that the distance between the positive sample satellite positioning information and the reference positioning information falls within a second preset distance range. Wherein, each distance value in the first preset distance range is greater than each distance value in the second preset distance range.

[0149] In the above embodiments, by training the availability determination model with a first training sample that includes multiple sets of positive samples and multiple sets of negative samples, the availability determination model's accuracy in determining availability can be improved.

[0150] In one embodiment, each group of positive samples further includes positive sample fusion positioning information; the positive sample fusion positioning information is obtained by initial fusion of positive sample sensor positioning information and positive sample satellite positioning information; the positive sample sensor positioning information is obtained by positioning prediction of the target object based on the positive sample sensor information; each group of negative samples further includes negative sample fusion positioning information; the negative sample fusion positioning information is obtained by initial fusion of negative sample sensor positioning information and negative sample satellite positioning information; the negative sample sensor positioning information is obtained by positioning prediction of the target object based on the negative sample sensor information.

[0151] In one embodiment, the availability determination model can be modeled using XGBoost (decision tree) or Random Forest; this embodiment is not limited to either. The parameters for modeling using XGBoost may include at least one of the following: booster, objective, gamma, max_depth, lambda, subsample, colsample_bytree, min_child_weight, silent, eta, seed, nthread, and scale_pos_weight (positive sample weight).

[0152] In one embodiment, after the availability determination model is modeled using XGBoost (decision tree), meaning the model training is complete, the availability determination model can include multiple trees. The terminal can input satellite features, sensor features, and positioning difference features into the trained availability determination model. Each tree will then correspond to a leaf node, and each leaf node will correspond to a score. The terminal can sum the scores from each tree and determine the availability of the satellite positioning information based on the summed score.

[0153] In the above embodiments, by further limiting the positive and negative samples in the first training samples of the usability determination model, that is, each group of positive samples also includes positive sample fusion positioning information, and each group of negative samples also includes negative sample fusion positioning information, the usability determination accuracy of the usability determination model can be further improved.

[0154] In one embodiment, the positioning accuracy is predicted by a trained accuracy prediction model; the method further includes a training step for the accuracy prediction model; the training step for the accuracy prediction model includes: acquiring a second training sample; the second training sample includes multiple sets of target samples; each set of target samples includes sample satellite information and corresponding sample sensor information; the sample satellite positioning information in the sample satellite information satisfies a second near-range difference condition with the reference positioning information; the sample sensor information is sensor information collected within a third target time period; the third target time period is the time period from the last time the satellite information was collected to the time point of the sample satellite information collection; predicting the target satellite positioning accuracy based on the second training sample using the accuracy prediction model to be trained, thereby obtaining the predicted positioning accuracy; updating the model parameters of the accuracy prediction model to be trained according to the difference between the predicted positioning accuracy and the reference positioning accuracy, to iteratively train the accuracy prediction model to be trained; the reference positioning accuracy is determined based on the difference between the sample satellite positioning information and the reference positioning information.

[0155] Specifically, the terminal can acquire a second training sample and input it into the accuracy prediction model to be trained. The accuracy prediction model then predicts the target satellite positioning accuracy based on the second training sample, thus obtaining the predicted positioning accuracy. Furthermore, the terminal can determine a loss value based on the difference between the predicted positioning accuracy and the reference positioning accuracy, and update the model parameters of the accuracy prediction model to be trained based on the loss value. This iterative training of the accuracy prediction model yields a trained accuracy prediction model.

[0156] In one embodiment, the second proximity difference condition can be that the distance between the positive sample satellite positioning information and the reference positioning information is less than a third preset distance threshold. For example, the third preset distance threshold can also be 2 meters, in which case the second proximity difference condition can be that the distance between the positive sample satellite positioning information and the reference positioning information is less than 2 meters.

[0157] In one embodiment, the second proximity difference condition may also be that the distance between the positive sample satellite positioning information and the reference positioning information falls within a third preset distance range.

[0158] In one embodiment, the accuracy prediction model can be modeled using a deep neural network. For example, a deep neural network might have the following four layers: the first layer is a fully connected layer with 40 neurons, using ReLU (Rectified Linear Unified Function) as the activation function. The second and third layers are fully connected layers with 10 neurons each, also using ReLU as the activation function. The fourth layer is the output layer, a fully connected layer with 1 neuron, using a linear function as the activation function. The optimizer used is the Adam (Adaptive Moment Estimator), and the objective function for regression is mean squared error (mean squared error loss function).

[0159] In one embodiment, after the accuracy prediction model is modeled using a deep neural network, i.e., after the accuracy prediction model training is completed, the terminal can input satellite features, sensor features, and positioning difference features into the trained accuracy prediction model. The fully connected layers in the accuracy prediction model extract and combine features from the input satellite features, sensor features, and positioning difference features, and then predict the positioning accuracy of the target satellite based on the combined features.

[0160] In the above embodiments, by training the accuracy prediction model with a second training sample including multiple sets of target samples, and by updating the model parameters of the accuracy prediction model by predicting the difference between the positioning accuracy and the reference positioning accuracy during the training process, the positioning accuracy prediction accuracy of the accuracy prediction model can be improved through iterative training.

[0161] In one embodiment, each set of target samples also includes sample fusion positioning information; the sample fusion positioning information is obtained by initial fusion of sample sensor positioning information and sample satellite positioning information; the sample sensor positioning information is obtained by positioning prediction of the target object based on sample sensor information.

[0162] In the above embodiments, by further limiting the target samples included in the second training samples of the training accuracy prediction model, that is, the target samples also include sample fusion positioning information, the positioning accuracy prediction accuracy of the accuracy prediction model can be further improved.

[0163] In one embodiment, the target object is an autonomous vehicle in an autonomous driving scenario; the sensor is a vehicle sensor installed on the autonomous vehicle; the vehicle sensor includes at least one of a vehicle speed sensor, a vehicle vision sensor, or a vehicle inertial sensor.

[0164] Specifically, the terminal can acquire satellite features obtained by feature extraction from satellite information, where the satellite information is provided by the target satellite for locating the autonomous vehicle. The terminal can also acquire sensor features obtained by feature extraction from sensor information, where the sensor information is collected by vehicle sensors installed on the autonomous vehicle. Furthermore, the terminal can predict the positioning accuracy of the target satellite based on the satellite features and sensor features. This positioning accuracy is used as a fusion reference parameter to fuse the sensor positioning information and the satellite positioning information to obtain the fused positioning information of the autonomous vehicle. The sensor positioning information is obtained by predicting the positioning of the autonomous vehicle based on the sensor information.

[0165] It is understandable that vehicle speed sensors can be used to acquire vehicle speed and direction information. Vehicle vision sensors can be used to acquire image information including lane lines or obstacles. Vehicle inertial sensors can be used to acquire vehicle acceleration and angular velocity information.

[0166] In the above embodiments, applying the positioning information processing method of this application to autonomous driving scenarios can improve the positioning accuracy of autonomous vehicles in autonomous driving scenarios.

[0167] In one embodiment, such as Figure 5 As shown, during the training phase, the terminal can acquire satellite information, sensor information, initial fused positioning information, and ground truth information to generate training samples. Based on the generated samples, it trains an availability determination model and an accuracy prediction model. During the prediction phase, the terminal can acquire satellite information, sensor information, and initial fused positioning information to predict the positioning accuracy of the target satellite. Based on these information, feature extraction is performed, and the availability determination model first determines the availability of the satellite positioning information based on the extracted features. If the satellite positioning information is determined to be available, the accuracy prediction model further predicts the positioning accuracy based on the extracted features. If the satellite positioning information is determined to be unavailable, a prompt indicating that the satellite positioning information is unavailable is directly output.

[0168] In one embodiment, positioning accuracy is typically characterized by positioning error, such as Figure 6 As shown, the positioning error directly provided by the satellite differs significantly from the actual positioning error, indicating that the positioning accuracy provided by the satellite is not very accurate. In contrast, the positioning error predicted in this application differs less from the actual positioning error, indicating that the positioning accuracy predicted in this application is more accurate than the traditional positioning accuracy provided by the satellite.

[0169] like Figure 7 As shown, in one embodiment, a location information processing method is provided. This embodiment applies this method to... Figure 1 Taking terminal 102 as an example, the method specifically includes the following steps:

[0170] Step 702: Train the usability determination model to be trained using the first training samples to obtain a trained usability determination model; wherein, the first training samples include multiple sets of positive samples and multiple sets of negative samples; each set of positive samples includes positive sample satellite information and corresponding positive sample sensor information; the positive sample satellite positioning information in the positive sample satellite information satisfies the long-distance difference condition with the reference positioning information; the positive sample sensor information is sensor information collected within a first target time period; the first target time period is the time period from the last time the satellite information was collected to the time point of the positive sample satellite information collection; each set of negative samples includes negative sample satellite information and corresponding negative sample sensor information; the negative sample satellite positioning information in the negative sample satellite information satisfies the first short-distance difference condition with the reference positioning information; the negative sample sensor information is sensor information collected within a second target time period; the second target time period is the time period from the last time the satellite information was collected to the time point of the negative sample satellite information collection.

[0171] In one embodiment, each group of positive samples further includes positive sample fusion positioning information; the positive sample fusion positioning information is obtained by initial fusion of positive sample sensor positioning information and positive sample satellite positioning information; the positive sample sensor positioning information is obtained by positioning prediction of the target object based on the positive sample sensor information; each group of negative samples further includes negative sample fusion positioning information; the negative sample fusion positioning information is obtained by initial fusion of negative sample sensor positioning information and negative sample satellite positioning information; the negative sample sensor positioning information is obtained by positioning prediction of the target object based on the negative sample sensor information.

[0172] Step 704: Obtain the second training sample; the second training sample includes multiple sets of target samples; each set of target samples includes sample satellite information and corresponding sample sensor information; the sample satellite positioning information in the sample satellite information satisfies the second near-range difference condition with the reference positioning information; the sample sensor information is the sensor information collected within the third target time period; the third target time period is the time period from the last time point of satellite information collection to the time point of sample satellite information collection.

[0173] Step 706: The accuracy of the target satellite positioning is predicted based on the second training sample using the accuracy prediction model to be trained, and the predicted positioning accuracy is obtained.

[0174] Step 708: Based on the difference between the predicted positioning accuracy and the reference positioning accuracy, update the model parameters of the accuracy prediction model to be trained, and iteratively train the accuracy prediction model to be trained to obtain the trained accuracy prediction model; the reference positioning accuracy is determined based on the difference between the sample satellite positioning information and the reference positioning information.

[0175] In one embodiment, each set of target samples also includes sample fusion positioning information; the sample fusion positioning information is obtained by initial fusion of sample sensor positioning information and sample satellite positioning information; the sample sensor positioning information is obtained by positioning prediction of the target object based on sample sensor information.

[0176] Step 710: Obtain satellite information corresponding to the target object; the satellite information is provided by the target satellite used to locate the target object.

[0177] Step 712: Collect information about the target object by using sensors installed on the target object to obtain sensing information about the target object.

[0178] Step 714: Extract features from satellite information to obtain satellite features.

[0179] In one embodiment, satellite information includes satellite azimuth, satellite elevation, and satellite signal-to-noise ratio; satellite features include satellite distribution features; there are multiple target satellites; for each target satellite, the position information of the target satellite is determined based on its azimuth and elevation, and the target area where the target satellite is located is determined based on the position information; for each target area, the satellite distribution sub-features corresponding to the target area are determined based on the satellite signal-to-noise ratio and position information corresponding to satellites of the same constellation in the target area; satellites of the same constellation are target satellites located in the target area and belonging to the same constellation; satellite distribution features are determined based on the satellite distribution sub-features corresponding to each target area.

[0180] In one embodiment, the number of target satellites is multiple; satellite information includes at least one of satellite azimuth angle, satellite elevation angle, or satellite signal-to-noise ratio (SNR); satellite features include at least one of satellite azimuth angle features, mean SNR features, median SNR features, or SNR percentage features; wherein, the satellite azimuth angle features are determined based on the satellite azimuth angle corresponding to each target satellite; the mean SNR features are determined based on the mean SNR corresponding to satellites at large elevation angles; large elevation angle satellites are target satellites whose elevation angle is greater than a preset elevation angle; the median SNR features are determined by taking the median of the SNR corresponding to each target satellite; the SNR percentage features are determined based on the ratio of the number of targets to the total number of satellites; the number of targets is the number of target satellites whose SNR is less than a preset SNR; and the total number of satellites is the number of target satellites.

[0181] Step 716: Extract features from the sensing information to obtain sensing features.

[0182] Step 718: Initially fuse the sensor positioning information and the satellite positioning information in the satellite information to obtain the initial fused positioning information of the target object.

[0183] Step 720: Perform a difference comparison analysis on the initial fused positioning information and satellite positioning information to obtain positioning difference features.

[0184] In one embodiment, the initial fused positioning information includes initial fused latitude and longitude information; the satellite positioning information includes satellite positioning latitude and longitude information; and the positioning difference features include latitude and longitude difference features. The initial fused latitude and longitude information and the satellite positioning latitude and longitude information are compared and analyzed to obtain the latitude and longitude difference features.

[0185] In one embodiment, the initial fused positioning information includes initial fused velocity information; the satellite positioning information includes satellite positioning velocity information; and the positioning difference feature includes velocity difference features. The velocity difference information between the initial fused velocity information and the satellite positioning velocity information is determined; and the velocity difference feature is determined based on the ratio of the velocity difference information to the initial fused velocity information.

[0186] Step 722: Using the trained availability determination model, the availability of satellite positioning information is predicted based on satellite characteristics, sensor characteristics, and positioning difference characteristics.

[0187] Step 724: If the availability prediction result indicates that the satellite positioning information is available, predict the positioning accuracy of the target satellite based on satellite characteristics, sensing characteristics, and positioning difference characteristics using the trained accuracy prediction model.

[0188] This application also provides an application scenario in which the above-described positioning information processing method is applied. Specifically, this positioning information processing method can be applied to the positioning of autonomous vehicles in autonomous driving scenarios. The terminal can train the availability judgment model to be trained using a first training sample to obtain a trained availability judgment model; wherein, the first training sample includes multiple sets of positive samples and multiple sets of negative samples; each set of positive samples includes positive sample satellite information and corresponding positive sample sensor information; the positive sample satellite positioning information in the positive sample satellite information satisfies the long-distance difference condition with the reference positioning information; the positive sample sensor information is sensor information collected within a first target time period; the first target time period is the time period from the last time satellite information was collected to the time point of positive sample satellite information collection; each set of negative samples includes negative sample satellite information and corresponding negative sample sensor information; the negative sample satellite positioning information in the negative sample satellite information satisfies the first short-distance difference condition with the reference positioning information; the negative sample sensor information is sensor information collected within a second target time period; the second target time period is the time period from the last time satellite information was collected to the time point of negative sample satellite information collection. Each group of positive samples also includes positive sample fused positioning information; positive sample fused positioning information is obtained by initial fusion of positive sample sensor positioning information and positive sample satellite positioning information; positive sample sensor positioning information is obtained by predicting the positioning of autonomous vehicles based on positive sample sensor information; each group of negative samples also includes negative sample fused positioning information; negative sample fused positioning information is obtained by initial fusion of negative sample sensor positioning information and negative sample satellite positioning information; negative sample sensor positioning information is obtained by predicting the positioning of autonomous vehicles based on negative sample sensor information.

[0189] The terminal can acquire a second training sample; the second training sample includes multiple sets of target samples; each set of target samples includes sample satellite information and corresponding sample sensor information; the sample satellite positioning information in the sample satellite information satisfies a second near-range difference condition with the reference positioning information; the sample sensor information is sensor information collected within a third target time period; the third target time period is the time period from the last time the satellite information was collected to the time the sample satellite information was collected; the accuracy prediction model to be trained predicts the accuracy of the target satellite positioning based on the second training sample, thus obtaining the predicted positioning accuracy; based on the difference between the predicted positioning accuracy and the reference positioning accuracy, the model parameters of the accuracy prediction model to be trained are updated to iteratively train the accuracy prediction model to be trained, thus obtaining the trained accuracy prediction model; the reference positioning accuracy is determined based on the difference between the sample satellite positioning information and the reference positioning information. Each set of target samples also includes sample fusion positioning information; the sample fusion positioning information is obtained by initial fusion of sample sensor positioning information and sample satellite positioning information; the sample sensor positioning information is obtained by predicting the positioning of the autonomous vehicle based on the sample sensor information.

[0190] The terminal can acquire satellite information corresponding to the target object and extract features from the satellite information to obtain satellite features. The satellite information is provided by the target satellites used for locating the autonomous vehicle. The satellite information includes satellite azimuth, satellite elevation, and satellite signal-to-noise ratio. The satellite features include satellite distribution characteristics. There are multiple target satellites. For each target satellite, the position information of the target satellite is determined based on its azimuth and elevation, and the target area where the target satellite is located is determined based on the position information. For each target area, the satellite distribution sub-features corresponding to the target area are determined based on the satellite signal-to-noise ratio and position information of the satellites in the same constellation in the target area. Satellites in the same constellation are target satellites located in the target area and belonging to the same constellation. The satellite distribution characteristics are determined based on the satellite distribution sub-features corresponding to each target area.

[0191] It is understood that satellite information may also include at least one of satellite azimuth, satellite elevation, or satellite signal-to-noise ratio (SNR); satellite features include at least one of satellite azimuth features, mean SNR features, median SNR features, or SNR percentage features; wherein, satellite azimuth features are determined based on the satellite azimuth corresponding to each target satellite; mean SNR features are determined based on the mean SNR corresponding to satellites at various elevation angles; satellites at large elevation angles are target satellites whose elevation angle is greater than a preset elevation angle; median SNR features are determined by taking the median of the SNR corresponding to each target satellite; SNR percentage features are determined based on the ratio of the number of targets to the total number of satellites; the number of targets is the number of target satellites whose SNR is less than a preset SNR; the total number of satellites is the number of target satellites.

[0192] The terminal can extract features from the sensor information to obtain sensor features; the sensor information is information collected by vehicle sensors installed on the autonomous vehicle; the initial fusion of the sensor positioning information and satellite positioning information is performed to obtain the initial fused positioning information of the autonomous vehicle; the initial fused positioning information and satellite positioning information are compared and analyzed to obtain positioning difference features; the initial fused positioning information includes initial fused latitude and longitude information; the satellite positioning information includes satellite positioning latitude and longitude information; the positioning difference features include latitude and longitude difference features. The initial fused positioning information includes initial fused speed information; the satellite positioning information includes satellite positioning speed information; the positioning difference features include speed difference features. The speed difference information between the initial fused speed information and the satellite positioning speed information is determined; the speed difference feature is determined based on the ratio of the speed difference information to the initial fused speed information.

[0193] The terminal can use a trained availability determination model to predict the availability of satellite positioning information based on satellite characteristics, sensor characteristics, and positioning difference characteristics. If the availability prediction result indicates that the satellite positioning information is available, the terminal can use a trained accuracy prediction model to predict the positioning accuracy of the target satellite based on satellite characteristics, sensor characteristics, and positioning difference characteristics.

[0194] It is understood that the predicted positioning accuracy can be used to fuse sensor positioning information and satellite positioning information; whereby the sensor positioning information is obtained by predicting the positioning of autonomous vehicles based on sensor information. Through the positioning information processing method of this application, more accurate satellite positioning accuracy can be obtained, thereby leading to more accurate fused positioning information for autonomous vehicles and improving the positioning accuracy for autonomous vehicles.

[0195] It is also understandable that the accurate positioning accuracy predicted in this application can be provided to a third party so that the third party can verify whether the positioning accuracy it obtained is accurate.

[0196] This application also provides another application scenario where the aforementioned positioning information processing method is applied. Specifically, this positioning information processing method can be applied to scenarios involving the positioning of assisted driving vehicles. It is understood that positioning of the assisted driving vehicle is also necessary in assisted driving scenarios to determine its location. Through the positioning information processing method of this application, more accurate satellite positioning accuracy can be obtained, thereby enabling the acquisition of more accurate fused positioning information for the assisted driving vehicle, improving the positioning accuracy of the assisted driving vehicle. It is also understood that the positioning information processing method of this application can be applied to navigation scenarios during vehicle driving, or to positioning scenarios for robots other than vehicles. Through the positioning information processing method of this application, more accurate satellite positioning accuracy can be obtained, thereby enabling the acquisition of more accurate fused positioning information for vehicles or other robots, improving the positioning accuracy for vehicles or other robots.

[0197] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially, these steps are not necessarily executed in that order. Unless otherwise expressly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the above embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0198] In one embodiment, such as Figure 8 As shown, a positioning information processing device 800 is provided. This device can be a software module or a hardware module, or a combination of both, integrated into a computer device. Specifically, the device includes:

[0199] The acquisition module 802 is used to acquire satellite information corresponding to the target object; the satellite information is provided by the target satellite that locates the target object; the target object is collected by sensors set on the target object to obtain sensing information for the target object; the satellite information is used to extract features to obtain satellite features; the sensing information is used to extract features to obtain sensing features.

[0200] The prediction module 804 is used to predict the positioning accuracy of the target satellite based on satellite characteristics and sensor characteristics.

[0201] In one embodiment, the apparatus further includes:

[0202] The comparison module is used to initially fuse the sensor positioning information and the satellite positioning information to obtain the initial fused positioning information of the target object; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information; the initial fused positioning information and the satellite positioning information are compared and analyzed to obtain the positioning difference features;

[0203] The prediction module 804 is also used to predict the positioning accuracy of the target satellite based on satellite characteristics, sensor characteristics and positioning difference characteristics.

[0204] In one embodiment, the initial fused positioning information includes initial fused latitude and longitude information; the satellite positioning information includes satellite positioning latitude and longitude information; the positioning difference features include latitude and longitude difference features; the comparison module is also used to perform latitude and longitude difference comparison analysis between the initial fused latitude and longitude information and the satellite positioning latitude and longitude information to obtain latitude and longitude difference features.

[0205] In one embodiment, the initial fused positioning information includes initial fused velocity information; the satellite positioning information includes satellite positioning velocity information; the positioning difference feature includes velocity difference feature; the comparison module is further configured to determine the velocity difference information between the initial fused velocity information and the satellite positioning velocity information; and to determine the velocity difference feature based on the ratio of the velocity difference information to the initial fused velocity information.

[0206] In one embodiment, satellite information includes satellite azimuth, satellite elevation, and satellite signal-to-noise ratio; satellite features include satellite distribution features; the number of target satellites is multiple; the acquisition module 802 is further configured to, for each target satellite, determine the position information of the target satellite based on its azimuth and elevation, and determine the target area where the target satellite is located based on the position information; for each target area, determine the satellite distribution sub-features corresponding to the target area based on the satellite signal-to-noise ratio and position information corresponding to satellites of the same constellation in the target area; satellites of the same constellation are target satellites located in the target area and belonging to the same constellation; and determine the satellite distribution features based on the satellite distribution sub-features corresponding to each target area.

[0207] In one embodiment, the target area includes target quadrants; the acquisition module 802 is further configured to, for each target satellite, project the target satellite onto the target plane according to the satellite azimuth and elevation angles of the target satellite to obtain the position information of the target satellite on the target plane; the target plane includes multiple quadrants; and determine the target quadrant in the target plane where the target satellite is located based on the position information of the target satellite on the target plane.

[0208] In one embodiment, the number of target satellites is multiple; satellite information includes at least one of satellite azimuth angle, satellite elevation angle, or satellite signal-to-noise ratio (SNR); satellite features include at least one of satellite azimuth angle features, mean SNR features, median SNR features, or SNR percentage features; wherein, the satellite azimuth angle features are determined based on the satellite azimuth angle corresponding to each target satellite; the mean SNR features are determined based on the mean SNR corresponding to satellites at large elevation angles; large elevation angle satellites are target satellites whose elevation angle is greater than a preset elevation angle; the median SNR features are determined by taking the median of the SNR corresponding to each target satellite; the SNR percentage features are determined based on the ratio of the number of targets to the total number of satellites; the number of targets is the number of target satellites whose SNR is less than a preset SNR; and the total number of satellites is the number of target satellites.

[0209] In one embodiment, the prediction module 804 is further configured to predict the availability of satellite positioning information based on satellite features and sensing features; and, if the availability prediction result indicates that the satellite positioning information is available, to perform the step of predicting the positioning accuracy of the target satellite based on satellite features and sensing features.

[0210] In one embodiment, the availability prediction result is obtained by an availability determination model trained on a first training sample; the first training sample includes multiple sets of positive samples and multiple sets of negative samples; wherein, each set of positive samples includes positive sample satellite information and corresponding positive sample sensor information; the positive sample satellite positioning information in the positive sample satellite information satisfies the long-distance difference condition with the reference positioning information; the positive sample sensor information is sensor information collected within a first target time period; the first target time period is the time period from the last time the satellite information was collected to the time point of the positive sample satellite information collection; each set of negative samples includes negative sample satellite information and corresponding negative sample sensor information; the negative sample satellite positioning information in the negative sample satellite information satisfies the first short-distance difference condition with the reference positioning information; the negative sample sensor information is sensor information collected within a second target time period; the second target time period is the time period from the last time the satellite information was collected to the time point of the negative sample satellite information collection.

[0211] In one embodiment, each group of positive samples further includes positive sample fusion positioning information; the positive sample fusion positioning information is obtained by initial fusion of positive sample sensor positioning information and positive sample satellite positioning information; the positive sample sensor positioning information is obtained by positioning prediction of the target object based on the positive sample sensor information; each group of negative samples further includes negative sample fusion positioning information; the negative sample fusion positioning information is obtained by initial fusion of negative sample sensor positioning information and negative sample satellite positioning information; the negative sample sensor positioning information is obtained by positioning prediction of the target object based on the negative sample sensor information.

[0212] In one embodiment, the positioning accuracy is predicted by a trained accuracy prediction model; the device further includes:

[0213] The training module is used to acquire a second training sample. The second training sample includes multiple sets of target samples. Each set of target samples includes sample satellite information and corresponding sample sensor information. The sample satellite positioning information and the reference positioning information in the sample satellite information satisfy a second near-range difference condition. The sample sensor information is sensor information collected within a third target time period. The third target time period is the time period between the last time the satellite information was collected and the time point of the sample satellite information collection. The accuracy prediction model to be trained predicts the accuracy of the target satellite positioning based on the second training sample, thus obtaining the predicted positioning accuracy. The model parameters of the accuracy prediction model to be trained are updated according to the difference between the predicted positioning accuracy and the reference positioning accuracy to iteratively train the accuracy prediction model to be trained. The reference positioning accuracy is determined based on the difference between the sample satellite positioning information and the reference positioning information.

[0214] In one embodiment, each set of target samples also includes sample fusion positioning information; the sample fusion positioning information is obtained by initial fusion of sample sensor positioning information and sample satellite positioning information; the sample sensor positioning information is obtained by positioning prediction of the target object based on sample sensor information.

[0215] In one embodiment, the positioning accuracy is used to fuse the satellite positioning information from the sensor positioning information and the satellite information; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information.

[0216] In one embodiment, the target object is an autonomous vehicle in an autonomous driving scenario; the sensor is a vehicle sensor installed on the autonomous vehicle; the vehicle sensor includes at least one of a vehicle speed sensor, a vehicle vision sensor, or a vehicle inertial sensor.

[0217] In one embodiment, reference Figure 9 In addition to the acquisition module 802 and the prediction module 804, the positioning information processing device 900 may also include a comparison module 806 and a training module 808.

[0218] The aforementioned positioning information processing device acquires satellite information corresponding to the target object; the satellite information is provided by the target satellite used for positioning the target object; it collects information about the target object through sensors installed on the target object to obtain sensor information for the target object; it extracts features from the satellite information to obtain satellite features. It also extracts features from the sensor information to obtain sensor features. Based on the satellite features and sensor features, it predicts the positioning accuracy of the target satellite. Since the determination of satellite positioning accuracy considers both satellite information from the target satellite and sensor information from the sensors, compared to the traditional method of directly determining the final positioning accuracy from the accuracy information received from the satellite, this application can obtain more accurate satellite positioning accuracy. Therefore, based on the more accurate satellite positioning accuracy, more accurate fused positioning information for the target object can be obtained, improving the positioning accuracy for the target object.

[0219] Each module in the aforementioned positioning information processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0220] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 10As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a location information processing method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0221] Those skilled in the art will understand that Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0222] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0223] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0224] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0225] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0226] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0227] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0228] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for processing location information, characterized in that, The method includes: Obtain satellite information corresponding to the target object; the satellite information is provided by multiple target satellites that locate the target object, and the satellite information includes satellite positioning information, satellite azimuth angle, satellite elevation angle and satellite signal-to-noise ratio; Information about the target object is collected by sensors installed on the target object to obtain sensing information about the target object; The sensor positioning information and the satellite positioning information are initially fused to obtain the initial fused positioning information of the target object; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information. The initial fused positioning information and the satellite positioning information are compared and analyzed to obtain positioning difference characteristics; Feature extraction is performed on the satellite information to obtain satellite features; the satellite features include satellite distribution features determined according to the satellite distribution sub-features corresponding to each target region. For each target satellite, the target region is determined according to the position information of the target satellite, which is determined according to the satellite azimuth and elevation angles of the target satellite. The satellite distribution sub-features are determined according to the satellite signal-to-noise ratio and position information corresponding to the satellites of the same constellation in the target region. Feature extraction is performed on the sensing information to obtain sensing features; The positioning accuracy of the target satellite is predicted based on the satellite characteristics, the sensing characteristics, and the positioning difference characteristics.

2. The method according to claim 1, characterized in that, The initial fused positioning information includes initial fused latitude and longitude information; the satellite positioning information includes satellite positioning latitude and longitude information; the positioning difference features include latitude and longitude difference features; The step of performing a difference comparison analysis between the initial fused positioning information and the satellite positioning information to obtain positioning difference features includes: The latitude and longitude information of the initial fused system is compared and analyzed with the latitude and longitude information of the satellite positioning system to obtain the latitude and longitude difference characteristics.

3. The method according to claim 1, characterized in that, The initial fused positioning information includes initial fused velocity information; the satellite positioning information includes satellite positioning velocity information; the positioning difference feature includes velocity difference feature; The step of performing a difference comparison analysis between the initial fused positioning information and the satellite positioning information to obtain positioning difference features includes: Determine the velocity difference information between the initial fused velocity information and the satellite positioning velocity information; The velocity difference characteristics are determined based on the ratio of the velocity difference information to the initial fusion velocity information.

4. The method according to claim 1, characterized in that, The step of extracting features from the satellite information to obtain satellite features includes: For each target satellite, the position information of the target satellite is determined based on its azimuth and elevation angles, and the target area where the target satellite is located is determined based on the position information. For each target area, satellite distribution sub-features corresponding to the target area are determined based on the satellite signal-to-noise ratio and position information of the satellites in the same constellation within the target area; the satellites in the same constellation are the target satellites located in the target area and belonging to the same constellation. The satellite distribution characteristics are determined based on the satellite distribution sub-features corresponding to each target region.

5. The method according to claim 4, characterized in that, The target area includes target quadrants; for each target satellite, determining the position information of the target satellite based on its azimuth and elevation angles, and determining the target area where the target satellite is located based on the position information, includes: For each target satellite, the target satellite is projected onto a target plane based on its azimuth and elevation angles to obtain its position information on the target plane; the target plane includes multiple quadrants. Based on the position information of the target satellite on the target plane, determine the target quadrant in the target plane where the target satellite is located.

6. The method according to claim 1, characterized in that, The satellite features include at least one of the following: satellite azimuth feature, mean signal-to-noise ratio feature, median signal-to-noise ratio feature, or signal-to-noise ratio percentage feature; The satellite azimuth feature is determined based on the azimuth angles corresponding to each of the target satellites; the mean signal-to-noise ratio (SNR) feature is determined based on the mean SNR of satellites corresponding to major elevation angles; the major elevation angle satellites are target satellites whose elevation angles are greater than a preset elevation angle; the median SNR feature is determined by taking the median of the SNRs corresponding to each of the target satellites; the SNR ratio ratio is determined based on the ratio of the number of targets to the total number of satellites, where the number of targets is the number of target satellites whose SNR is less than a preset SNR; and the total number of satellites is the number of target satellites.

7. The method according to claim 1, characterized in that, The method further includes: Based on the satellite characteristics and the sensing characteristics, the availability of the satellite positioning information is predicted; If the availability prediction result indicates that the satellite positioning information is available, the step of predicting the positioning accuracy of the target satellite based on the satellite characteristics, the sensing characteristics, and the positioning difference characteristics is performed.

8. The method according to claim 7, characterized in that, The availability prediction result is obtained by the availability determination model trained on the first training sample; the first training sample includes multiple sets of positive samples and multiple sets of negative samples; Each group of positive samples includes positive sample satellite information and corresponding positive sample sensor information; the positive sample satellite positioning information in the positive sample satellite information satisfies the long-distance difference condition with the reference positioning information; the positive sample sensor information is sensor information collected within a first target time period; the first target time period is the time period from the last time the satellite information was collected to the time point of the positive sample satellite information collection. Each set of negative samples includes negative sample satellite information and corresponding negative sample sensor information; the negative sample satellite positioning information in the negative sample satellite information satisfies a first proximity difference condition with the reference positioning information; the negative sample sensor information is sensor information collected within a second target time period; the second target time period is the time period between the last time the satellite information was collected and the time point of the negative sample satellite information collection.

9. The method according to claim 8, characterized in that, Each group of positive samples also includes positive sample fusion positioning information; the positive sample fusion positioning information is obtained by initial fusion of positive sample sensor positioning information and positive sample satellite positioning information; the positive sample sensor positioning information is obtained by positioning prediction of the target object based on the positive sample sensor information; Each group of negative samples also includes negative sample fusion positioning information; the negative sample fusion positioning information is obtained by initial fusion of negative sample sensor positioning information and negative sample satellite positioning information; the negative sample sensor positioning information is obtained by positioning prediction of the target object based on the negative sample sensor information.

10. The method according to claim 1, characterized in that, The positioning accuracy is predicted by a trained accuracy prediction model; the method further includes a training step for the accuracy prediction model; the training step for the accuracy prediction model includes: Acquire a second training sample; the second training sample includes multiple sets of target samples; each set of target samples includes sample satellite information and corresponding sample sensor information; the sample satellite positioning information in the sample satellite information satisfies a second near-range difference condition with the reference positioning information; the sample sensor information is sensor information collected within a third target time period; the third target time period is the time period from the last time the satellite information was collected to the time point of the sample satellite information collection; The accuracy of the target satellite positioning is predicted by the accuracy prediction model to be trained based on the second training sample, and the predicted positioning accuracy is obtained. Based on the difference between the predicted positioning accuracy and the reference positioning accuracy, the model parameters of the accuracy prediction model to be trained are updated to iteratively train the accuracy prediction model to be trained; the reference positioning accuracy is determined based on the difference between the sample satellite positioning information and the reference positioning information.

11. The method according to claim 10, characterized in that, Each set of target samples also includes sample fusion positioning information; the sample fusion positioning information is obtained by initial fusion of sample sensor positioning information and sample satellite positioning information; the sample sensor positioning information is obtained by positioning prediction of the target object based on the sample sensor information.

12. The method according to claim 1, characterized in that, The positioning accuracy is used to fuse the sensor positioning information and the satellite positioning information in the satellite information; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information.

13. The method according to any one of claims 1 to 12, characterized in that, The target object is an autonomous vehicle in an autonomous driving scenario; the sensor is a vehicle sensor installed on the autonomous vehicle; the vehicle sensor includes at least one of a vehicle speed sensor, a vehicle vision sensor, or a vehicle inertial sensor.

14. A positioning information processing device, characterized in that, The device includes: An acquisition module is used to acquire satellite information corresponding to a target object; the satellite information is provided by multiple target satellites that locate the target object, and the satellite information includes satellite positioning information, satellite azimuth angle, satellite elevation angle, and satellite signal-to-noise ratio; the target object is used to collect information by sensors installed on the target object to obtain sensing information for the target object; The comparison module is used to initially fuse the sensor positioning information and the satellite positioning information to obtain the initial fused positioning information of the target object; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information; and the initial fused positioning information and the satellite positioning information are compared and analyzed to obtain positioning difference features. The acquisition module is further configured to perform feature extraction on the satellite information to obtain satellite features; the satellite features include satellite distribution features determined based on satellite distribution sub-features corresponding to each target region. For each target satellite, the target region is determined based on the position information of the target satellite, which is determined based on the satellite azimuth and elevation angles of the target satellite. The satellite distribution sub-features are determined based on the signal-to-noise ratio and position information of satellites in the same constellation within the target region; and the sensor information is further configured to perform feature extraction to obtain sensor features. The prediction module is used to predict the positioning accuracy of the target satellite based on the satellite features, the sensing features, and the positioning difference features.

15. The apparatus according to claim 14, characterized in that, The initial fused positioning information includes initial fused latitude and longitude information; the satellite positioning information includes satellite positioning latitude and longitude information; the positioning difference features include latitude and longitude difference features; The comparison module is also used to perform latitude and longitude difference comparison analysis between the initial fused latitude and longitude information and the satellite positioning latitude and longitude information to obtain latitude and longitude difference features.

16. The apparatus according to claim 14, characterized in that, The initial fused positioning information includes initial fused velocity information; the satellite positioning information includes satellite positioning velocity information; the positioning difference feature includes velocity difference feature; The comparison module is also used to determine the speed difference information between the initial fused speed information and the satellite positioning speed information; and to determine the speed difference characteristics based on the ratio of the speed difference information to the initial fused speed information.

17. The apparatus according to claim 14, characterized in that, The acquisition module is further configured to determine the position information of each target satellite based on its azimuth and elevation angles, and to determine the target area where the target satellite is located based on the position information. For each target area, satellite distribution sub-features are determined based on the satellite signal-to-noise ratio and position information of satellites in the same constellation within the target area; the satellites in the same constellation are the target satellites located in the target area and belonging to the same constellation; satellite distribution features are determined based on the satellite distribution sub-features corresponding to each target area.

18. The apparatus according to claim 17, characterized in that, The target area includes target quadrants; the acquisition module is further configured to, for each target satellite, project the target satellite onto the target plane according to the satellite azimuth and satellite elevation angles of the target satellite, to obtain the position information of the target satellite on the target plane; The target plane includes multiple quadrants; based on the position information of the target satellite on the target plane, the target quadrant in which the target satellite is located in the target plane is determined.

19. The apparatus according to claim 14, characterized in that, The satellite features include at least one of the following: satellite azimuth feature, mean signal-to-noise ratio feature, median signal-to-noise ratio feature, or signal-to-noise ratio percentage feature; The satellite azimuth feature is determined based on the azimuth angles corresponding to each of the target satellites; the mean signal-to-noise ratio (SNR) feature is determined based on the mean SNR of satellites corresponding to major elevation angles; the major elevation angle satellites are target satellites whose elevation angles are greater than a preset elevation angle; the median SNR feature is determined by taking the median of the SNRs corresponding to each of the target satellites; the SNR ratio ratio is determined based on the ratio of the number of targets to the total number of satellites, where the number of targets is the number of target satellites whose SNR is less than a preset SNR; and the total number of satellites is the number of target satellites.

20. The apparatus according to claim 14, characterized in that, The prediction module is further configured to perform availability prediction on the satellite positioning information based on the satellite features and the sensing features; and, if the availability prediction result indicates that the satellite positioning information is available, execute the step of predicting the positioning accuracy of the target satellite based on the satellite features, the sensing features, and the positioning difference features.

21. The apparatus according to claim 20, characterized in that, The availability prediction result is obtained by the availability determination model trained on the first training sample; the first training sample includes multiple sets of positive samples and multiple sets of negative samples; Each group of positive samples includes positive sample satellite information and corresponding positive sample sensor information; the positive sample satellite positioning information in the positive sample satellite information satisfies the long-distance difference condition with the reference positioning information; the positive sample sensor information is sensor information collected within a first target time period; the first target time period is the time period from the last time the satellite information was collected to the time point of the positive sample satellite information collection. Each set of negative samples includes negative sample satellite information and corresponding negative sample sensor information; the negative sample satellite positioning information in the negative sample satellite information satisfies a first proximity difference condition with the reference positioning information; the negative sample sensor information is sensor information collected within a second target time period; the second target time period is the time period between the last time the satellite information was collected and the time point of the negative sample satellite information collection.

22. The apparatus according to claim 21, characterized in that, Each group of positive samples also includes positive sample fusion positioning information; the positive sample fusion positioning information is obtained by initial fusion of positive sample sensor positioning information and positive sample satellite positioning information; the positive sample sensor positioning information is obtained by positioning prediction of the target object based on the positive sample sensor information; Each group of negative samples also includes negative sample fusion positioning information; the negative sample fusion positioning information is obtained by initial fusion of negative sample sensor positioning information and negative sample satellite positioning information; the negative sample sensor positioning information is obtained by positioning prediction of the target object based on the negative sample sensor information.

23. The apparatus according to claim 14, characterized in that, The positioning accuracy is predicted by a trained accuracy prediction model. The device also includes a training module for acquiring a second training sample. The second training sample includes multiple sets of target samples. Each set of target samples includes sample satellite information and corresponding sample sensor information. The sample satellite positioning information and the reference positioning information in the sample satellite information satisfy a second near-range difference condition. The sample sensor information is sensor information collected within a third target time period. The third target time period is the time period from the last time the satellite information was collected to the time of the sample satellite information collection. The accuracy prediction model to be trained predicts the positioning accuracy of the target satellite based on the second training sample to obtain the predicted positioning accuracy. The model parameters of the accuracy prediction model to be trained are updated according to the difference between the predicted positioning accuracy and the reference positioning accuracy to iteratively train the accuracy prediction model to be trained. The reference positioning accuracy is determined based on the difference between the sample satellite positioning information and the reference positioning information.

24. The apparatus according to claim 23, characterized in that, Each set of target samples also includes sample fusion positioning information; the sample fusion positioning information is obtained by initial fusion of sample sensor positioning information and sample satellite positioning information; the sample sensor positioning information is obtained by positioning prediction of the target object based on the sample sensor information.

25. The apparatus according to claim 14, characterized in that, The positioning accuracy is used to fuse the sensor positioning information and the satellite positioning information in the satellite information; the sensor positioning information is obtained by predicting the positioning of the target object based on the sensor information.

26. The apparatus according to any one of claims 14 to 25, characterized in that, The target object is an autonomous vehicle in an autonomous driving scenario; the sensor is a vehicle sensor installed on the autonomous vehicle; the vehicle sensor includes at least one of a vehicle speed sensor, a vehicle vision sensor, or a vehicle inertial sensor.

27. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 13.

28. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 13.

29. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 13.

Images