Scene recognition method and device, medium and vehicle

Abstract

The invention relates to a scene recognition method and device, a medium and a vehicle, and relates to the technical field of navigation, and the scene recognition method comprises the steps: obtaining target observation data of a target satellite; determining target feature information related to the scene according to the target observation data; and determining a scene type of a target observation position corresponding to the target observation data according to the target feature information. According to the invention, the scene type of the target observation position is determined by extracting the target feature information in the target observation data, so that the accuracy of scene recognition can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of navigation technology, and in particular to a scene recognition method, device, medium, and vehicle. Background Art

[0002] Among related technologies, scene recognition methods based on visual positioning and deep learning have high requirements on computing power and large amounts of data. Summary of the Invention

[0003] To overcome the problems existing in the related art, the present disclosure provides a scene recognition method, device, medium and vehicle.

[0004] According to a first aspect of an embodiment of the present disclosure, a scene recognition method is provided, including:

[0005] Acquire target observation data of target satellite;

[0006] Determining target feature information related to the scene based on the target observation data;

[0007] The scene type of the target observation position corresponding to the target observation data is determined according to the target feature information.

[0008] Optionally, the target characteristic information includes the number of carriers in the target observation data that do not experience cycle slips;

[0009] The determining, based on the target feature information, the scene type of the target observation position corresponding to the target observation data includes:

[0010] When the number of carrier waves without cycle slips is less than a preset threshold, it is determined that the target observation position is in a first occlusion scenario.

[0011] Optionally, there are multiple target satellites, and the target characteristic information further includes a maximum observation blind spot angle among multiple observation blind spot angles, and an observation blind spot angle in the true north direction, wherein the observation blind spot angle refers to an observation blind spot angle between two adjacent target satellites in a target satellite sequence, and the target satellite sequence is obtained by sorting the target satellites in ascending order according to their azimuth angles;

[0012] The determining, based on the target feature information, the scene type of the target observation position corresponding to the target observation data includes:

[0013] When the number of carriers that have not experienced cycle slips is greater than or equal to the preset number threshold, and at least one of the maximum observation blind spot angle and the north direction observation blind spot angle meets the preset conditions, it is determined that the target observation position is in the second occlusion scene, wherein the occlusion degree of the first occlusion scene is greater than the occlusion degree of the second occlusion scene.

[0014] Optionally, whether the maximum observation blind spot angle meets a preset condition is determined by:

[0015] When the maximum observation blind spot angle is greater than or equal to the first angle threshold, it is determined that the maximum observation blind spot angle meets the preset condition.

[0016] Optionally, whether the north direction observation blind spot angle meets the preset condition is determined by the following method:

[0017] Obtaining a target latitude value of the target observation position;

[0018] Determining a target second angle threshold according to the target latitude value, a correspondence between the latitude value and the second angle threshold;

[0019] When the north direction observation blind spot angle is greater than or equal to the target second angle threshold, it is determined that the north direction observation blind spot angle meets the preset condition.

[0020] Optionally, the latitude value is positively correlated with the second angle threshold.

[0021] Optionally, determining, based on the target feature information, the scene type of the target observation position corresponding to the target observation data includes:

[0022] When the number of carriers that have not experienced cycle slips is greater than or equal to the preset number threshold, and the maximum observation blind spot angle and the north direction observation blind spot angle do not meet the preset conditions, it is determined that the target observation position is in the third occlusion scene, wherein the occlusion degree of the second occlusion scene is greater than the occlusion degree of the third occlusion scene.

[0023] Optionally, acquiring target observation data of a target satellite includes:

[0024] Obtain observation data from multiple satellites;

[0025] For each satellite, if the observation data of the satellite includes a carrier phase observation value, the satellite is determined as the target satellite, and the observation data of the satellite is determined as the target observation data.

[0026] According to a second aspect of an embodiment of the present disclosure, there is provided a scene recognition device, including:

[0027] an acquisition module, configured to acquire target observation data of a target satellite;

[0028] A first determining module is configured to determine target feature information related to the scene based on the target observation data;

[0029] The second determining module is configured to determine the scene type of the target observation position corresponding to the target observation data according to the target feature information.

[0030] According to a third aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the method provided by the first aspect of the embodiment of the present disclosure is implemented.

[0031] According to a fourth aspect of an embodiment of the present disclosure, a vehicle is provided, comprising:

[0032] a storage device for storing a computer program;

[0033] A processing device is used to execute the computer program to implement the method provided by the first aspect of the embodiment of the present disclosure.

[0034] The technical solutions provided by the embodiments of the present disclosure may have the following beneficial effects:

[0035] The present disclosure determines target feature information related to the scene through target observation data of the target satellite, and determines the scene type of the target observation position corresponding to the target observation data based on the target feature information. In this way, the scene type of the target observation position is determined by extracting the target feature information from the target observation data, which does not require large computing power and data volume, and can improve the accuracy of scene recognition.

[0036] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0038] Figure 1 The figure is a flowchart of a scene recognition method according to an exemplary embodiment.

[0039] Figure 2 is a schematic diagram showing a first occlusion scene according to an exemplary embodiment.

[0040] Figure 3 is a schematic diagram showing a second occlusion scene according to an exemplary embodiment.

[0041] Figure 4 is a schematic diagram showing a third occlusion scenario according to an exemplary embodiment.

[0042] Figure 5 The figure is a schematic diagram showing the distribution of azimuth and elevation angles of a satellite sequence according to an exemplary embodiment.

[0043] Figure 6 The figure is a block diagram of a scene recognition device according to an exemplary embodiment.

[0044] Figure 7 is a block diagram of a vehicle according to an exemplary embodiment. DETAILED DESCRIPTION

[0045] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

[0046] When a vehicle is driving on an urban road, if there is an overpass with a high-precision map above, visual positioning may locate the overpass above, resulting in positioning error. The original GNSS (Global Navigation Satellite System) signal is an electromagnetic wave signal from a satellite. Some of the features it contains, such as the signal-to-noise ratio, the number of visible satellites, the geometric distribution of satellites, etc., are closely related to the environment in which the receiver is located and can provide scene-related information. Therefore, by extracting scene-related features from the received GNSS original signal, the scene in which the receiver is located can be analyzed and deduced, that is, scene recognition can be achieved. Combined with existing scene recognition methods, it can further improve the accuracy and robustness of positioning.

[0047] Reference Figure 1 , Figure 1 is a flow chart showing a scene recognition method according to an exemplary embodiment. Figure 1 As shown, the scene recognition method is used in a vehicle and includes the following steps.

[0048] In step S101 , target observation data of a target satellite is acquired.

[0049] In step S102, target feature information related to the scene is determined based on the target observation data.

[0050] In step S103, the scene type of the target observation position corresponding to the target observation data is determined according to the target feature information.

[0051] For example, a satellite refers to a natural celestial body that orbits a planet in a periodic closed orbit. The target satellites are some of the total number of satellites, and the target observation data are the observation data included in the raw GNSS signals acquired by the target satellites. A receiver mounted on a vehicle can receive observation data from multiple satellites, and target observation data of the target satellite can be obtained from the observation data from the multiple satellites. For example, data preprocessing can be performed on the observation data from the multiple satellites, and observation data that meets preset requirements can be selected as the target observation data of the target satellite.

[0052] For example, the target observation position is the location of a receiver that receives the target observation data. Here, the receiver can be set on a vehicle. The target feature information related to the scene may include the number of carrier phase observation values, the number of visible satellites, the distribution location of visible satellites, etc. The target feature information can be used to determine the environmental characteristics of the location of the target observation position. Using the target observation data, feature information related to the scene involved in the observation data can be obtained. Based on the target feature information, the specific scene type of the target observation position can be determined. For example, a model, algorithm or preset judgment rule can be used to determine the scene type corresponding to the target feature information, and finally the specific scene type of the target observation position can be obtained.

[0053] The present disclosure uses target observation data from a target satellite to determine target feature information related to a scene. Based on this target feature information, the present disclosure determines the scene type of the target observation location corresponding to the target observation data. This improves the accuracy of scene recognition by extracting target feature information from the target observation data. Furthermore, compared to other existing scene recognition methods, this method requires less computing power, less data, and is less expensive.

[0054] As an optional embodiment, the target characteristic information includes the number of carrier waves in the target observation data that do not experience cycle slips;

[0055] According to the target feature information, the scene type of the target observation location corresponding to the target observation data is determined, including:

[0056] When the number of carrier waves without cycle slips is less than a preset threshold, it is determined that the target observation position is in the first occlusion scenario.

[0057] For example, the target observation data includes carrier phase observations, which represent the phase angle of the satellite signal. The difference between the carrier phase observations and the phase angle at the time of signal transmission can be used to calculate the distance the signal has propagated. Each carrier cycle corresponds to a certain wavelength, namely the wavelength of the signal. When the receiver encounters a signal interruption or anomaly during the observation process, the carrier phase observations may undergo discontinuous jumps, a phenomenon called cycle slips. In one case, when the satellite signal is blocked by buildings, trees, or other objects, the receiver may lose signal tracking, resulting in discontinuity in the carrier phase observations and cycle slips in the carrier phase observations. In another case, the satellite signal travels through multiple paths during propagation, causing the receiver to receive both the original signal and the reflected signal. The resulting multipath effect may also cause jumps in the carrier phase observations.

[0058] For example, the number of carriers that do not cycle slip is the number of carrier phase observation values ​​that do not experience the cycle slip phenomenon. The preset number threshold can be 5-9, and further can be 7. When the number of carriers that do not cycle slip is less than the preset number threshold, the observation signal sent by the target satellite may be largely blocked by buildings, trees or other objects, so it can be determined that the target observation position is in the first blocking scene. Among them, the first blocking scene can be a scene in which the ratio of the area of ​​the blocking object above the vehicle to the entire picture in the picture obtained by the camera in front of the vehicle is greater than or equal to the first proportion threshold. For example, the first proportion threshold can be one-half. You can refer to Figure 2 , Figure 2 A schematic diagram of a vehicle in a first obstruction scene obtained by a camera in front of the vehicle is shown, wherein the vehicle is on the road directly below an overpass, and fences are set up on both sides of the road. The overpass blocks most of the view directly above the vehicle.

[0059] As an optional embodiment, there are multiple target satellites, and the target characteristic information further includes a maximum observation blind spot angle among the multiple observation blind spot angles, and an observation blind spot angle in the true north direction. The observation blind spot angle refers to the angle of the observation blind spot between two adjacent target satellites in the target satellite sequence. The target satellite sequence is obtained by sorting the target satellites in ascending order according to their azimuth angles.

[0060] According to the target feature information, the scene type of the target observation location corresponding to the target observation data is determined, including:

[0061] When the number of carriers without cycle slips is greater than or equal to a preset number threshold, and at least one of the maximum observation blind spot angle and the north direction observation blind spot angle meets the preset conditions, it is determined that the target observation position is in the second occlusion scene, wherein the occlusion degree of the first occlusion scene is greater than the occlusion degree of the second occlusion scene.

[0062] For example, the azimuth angle represents the horizontal angle between the north direction line of a certain point and the target direction line in a clockwise direction. The azimuth angle of the target satellite is the angle between the north direction line of the target observation position and the direction line corresponding to the target satellite and the target observation position in a clockwise direction. The target satellite sequence is obtained by sorting the target satellites in ascending order according to their azimuth angles. Figure 5 , Figure 5 The numbers in the outermost circle represent the azimuth, which ranges from 0 to 360 degrees. The numbers corresponding to each dotted circle represent the altitude. Figure 5 The range of the mid-altitude angle is 0 to 90 degrees, N is the north direction line, and satellites 1, 2, 3 and 4 are the target satellite sequences sorted from small to large in azimuth angle. The observation blind area represents the fan-shaped area between two adjacent target satellites in the target satellite sequence, and the observation blind area angle represents the angle of the observation blind area between two adjacent target satellites in the target satellite sequence. Figure 5 The angle of the sector area between two adjacent target satellites is θ1, θ2, θ3 and θ4. The north direction observation blind area angle is the angle of the observation blind area including the north direction line in multiple observation blind area angles. The maximum observation blind area angle is the maximum value of multiple observation blind area angles except the north direction observation blind area. Figure 5 The observation blind area angle in the central north direction is θ4, and the maximum observation blind area angle is θ3.

[0063] For example, when the number of carriers that do not cycle slip is less than a preset number threshold, the observation signal sent by the target satellite may be less obscured by buildings, trees or other objects, or the observation signal sent by the target satellite may not be obscured. In this case, the scene type of the target observation position can be determined based on the maximum observation blind spot angle and the observation blind spot angle in the north direction. Wherein, when at least one of the maximum observation blind spot angle and the observation blind spot angle in the north direction meets the preset conditions, it can be determined that the target observation position is in the second obstruction scene, and the obstruction degree of the first obstruction scene is greater than the obstruction degree of the second obstruction scene. Wherein, the second obstruction scene can be a scene in which the ratio of the area of ​​the obstruction above the vehicle to the entire picture captured by the camera in front of the vehicle is less than the first proportion threshold and greater than or equal to the second proportion threshold, for example, the first proportion threshold is one-half, and the second proportion threshold is one-quarter. You can refer to Figure 3 , Figure 3 A schematic diagram of a vehicle in a second obstruction scene obtained by a camera in front of the vehicle is shown, wherein the vehicle is on the road below the side of an overpass, and a fence is set up on one side of the road. The overpass blocks part of the view above the side where the vehicle is located, while the view above the other side is relatively open.

[0064] As an optional embodiment, whether the maximum observation blind spot angle meets the preset conditions is determined in the following manner:

[0065] When the maximum observation blind spot angle is greater than or equal to the first angle threshold, it is determined that the maximum observation blind spot angle meets the preset condition.

[0066] For example, the first angle threshold may be 60 to 120 degrees, and may be further set to 90 degrees. When the maximum observation blind spot angle is greater than or equal to 90 degrees, the maximum observation blind spot angle meets the preset condition, and the observation signal sent by the target satellite may be less obstructed by buildings, trees, or other objects. Therefore, it can be determined that the target observation position is in the second obstruction scenario.

[0067] As an optional embodiment, whether the blind spot angle in the north direction meets the preset conditions is determined in the following manner:

[0068] Get the target latitude value of the target observation position;

[0069] Determining the target second angle threshold according to the target latitude value, the correspondence between the latitude value and the second angle threshold;

[0070] When the blind spot angle of the observation in the due north direction is greater than or equal to the target second angle threshold, it is determined that the blind spot angle of the observation in the due north direction meets the preset condition.

[0071] For example, whether the blind spot angle of observation in the due north direction meets the preset conditions is related to the latitude value of the target observation position, and the target latitude value represents the latitude of the target observation position. According to the target latitude value, the correspondence between the latitude value and the second angle threshold, the target second angle threshold used to judge whether the blind spot angle of observation in the due north direction meets the preset conditions can be determined. Among them, in the correspondence between the latitude value and the second angle threshold, the larger the latitude value, the larger the second angle threshold can be. The higher the latitude, the sparser the satellites in the due north direction, so the second angle threshold is generally greater than the first angle threshold. When the target observation position is in a low-latitude area, the lower the latitude, the denser the satellites in the due north direction, and the corresponding second angle threshold can be close to the first angle threshold, so the second angle threshold is greater than or equal to the first angle threshold.

[0072] In one feasible implementation, the scene type of the target observation position is determined based on the latitude value of the target observation position. When the latitude value is less than or equal to the first threshold value, the target observation position is in a low-latitude area. Therefore, when the observation blind spot angle in the north direction is greater than or equal to the second threshold value, or the maximum observation blind spot angle is greater than or equal to the second threshold value, the scene type of the target observation position is determined to be the second occlusion scene. For example, when the latitude value is less than or equal to 39.5 degrees, and when the observation blind spot angle in the north direction is greater than or equal to 90 degrees, or the maximum observation blind spot angle is greater than or equal to 90 degrees, the scene type of the target observation position is determined to be the second occlusion scene.

[0073] When the latitude value is greater than the first threshold and less than or equal to the third threshold, the target observation location is in the mid-latitude region. Therefore, when the observation blind spot angle in the due north direction is greater than or equal to the fourth threshold, or the maximum observation blind spot angle is greater than or equal to the second threshold, the scene type of the target observation location is determined to be the second occlusion scene. For example, when the latitude value is greater than 39.5 degrees and less than or equal to 43 degrees, and when the observation blind spot angle in the due north direction is greater than or equal to 100 degrees, or the maximum observation blind spot angle is greater than or equal to 90 degrees, the scene type of the target observation location is determined to be the second occlusion scene.

[0074] When the latitude value is greater than the third threshold, the target observation location is in a high-latitude region. At this time, the geometric distribution of satellites in the north direction is the sparsest. Considering the accuracy of the measurement, it can be set that when the observation blind spot angle in the north direction is greater than or equal to the fifth threshold, or the maximum observation blind spot angle is greater than or equal to the second threshold, the scene type of the target observation location is determined to be the second occlusion scene. For example, when the latitude value is greater than 43 degrees, and when the observation blind spot angle in the north direction is greater than or equal to 105 degrees, or the maximum observation blind spot angle is greater than or equal to 90 degrees, the scene type of the target observation location is determined to be the second occlusion scene.

[0075] As an optional embodiment, the latitude value is positively correlated with the second angle threshold.

[0076] For example, the higher the latitude value, the sparser the satellites in the north direction. Therefore, in order to avoid measurement errors, the higher the latitude value, the larger the second angle threshold corresponding to the blind spot angle of the north direction observation can be set. Therefore, the latitude value can be positively correlated with the second angle threshold. Among them, a second angle threshold can be set for each latitude value. For example, when the latitude value is 40 degrees, the second angle threshold is 101 degrees; when the latitude value is 41 degrees, the second angle threshold is 103 degrees. In addition, a second angle threshold can be corresponding to each latitude value interval. For example, when the latitude value is in the range of 39.5 degrees to 43 degrees, the second angle threshold is 100 degrees; when the latitude value is in the range of 43 degrees to 57 degrees, the second angle threshold is 103 degrees.

[0077] In the scene recognition method disclosed herein, based on the characteristic that the satellite azimuth distribution changes with the latitude value, that is, the geometric distribution of satellites in the north direction gradually becomes sparse, resulting in the characteristic that the satellite observation blind spot in the north direction in open scenes naturally becomes larger. Therefore, the target second angle threshold determined according to the target latitude value and the relationship between the latitude value and the second angle threshold is more suitable for actual scene judgment and has better robustness.

[0078] As an optional embodiment, determining the scene type of the target observation position corresponding to the target observation data according to the target feature information includes:

[0079] When the number of carriers without cycle slips is greater than or equal to the preset number threshold, and the maximum observation blind spot angle and the observation blind spot angle in the north direction do not meet the preset conditions, it is determined that the target observation position is in the third occlusion scene, where the occlusion degree of the second occlusion scene is greater than the occlusion degree of the third occlusion scene.

[0080] For example, when the number of carriers that do not cycle slip is less than a preset number threshold, the observation signal sent by the target satellite may be less obstructed by buildings, trees or other objects, or the observation signal sent by the target satellite is not obstructed. In this case, the scene type of the target observation position can be further determined based on the maximum observation blind spot angle and the observation blind spot angle in the north direction. When the maximum observation blind spot angle and the observation blind spot angle in the north direction meet the preset conditions, the target observation position is determined to be in the second obstruction scene. When the maximum observation blind spot angle and the observation blind spot angle in the north direction do not meet the preset conditions, the target observation position is determined to be in the third obstruction scene, and the obstruction degree of the second obstruction scene is greater than the obstruction degree of the third obstruction scene. The third obstruction scene can be a scene in which the ratio of the area of ​​the obstruction above the vehicle to the entire picture captured by the camera in front of the vehicle is less than the second proportion threshold, for example, the second proportion threshold is one quarter. You can refer to Figure 4 , Figure 4A schematic diagram of a vehicle in a third occlusion scenario obtained by a camera in front of the vehicle is shown, wherein there is an overpass in front of the vehicle that is transverse to the direction of travel of the vehicle, but there are no fences on both sides of the road. At this time, the vehicle's upper field of view is not blocked and the field of view is relatively wide.

[0081] As an optional embodiment, obtaining target observation data of a target satellite includes:

[0082] Obtain observation data from multiple satellites;

[0083] For each satellite, if the observation data of the satellite includes a carrier phase observation value, the satellite is determined as a target satellite, and the observation data of the satellite is determined as target observation data.

[0084] For example, a receiver's position can be calculated by receiving signals transmitted by satellites at two different frequencies. These frequencies are called L1 and L2. The L1 frequency is 1575.42 MHz, and the L2 frequency is 1227.60 MHz. Since these two frequency signals propagate at different speeds through the atmosphere, the receiver's position can be calculated by measuring the phase difference between them. This technique is widely used in the Global Positioning System and other satellite navigation systems.

[0085] Among them, the satellite's observation data may include low-precision pseudo-range observation values ​​and high-precision carrier phase observation values. If the satellite's observation data includes carrier phase observation values, the satellite can be determined as a target satellite, and the satellite's observation data can be determined as target observation data. If the satellite's observation data does not include carrier phase observation values, it can be considered that the observation data obtained by this satellite is poor, for example, the observation data is deteriorated after being affected by NLOS (Non Line of Sight) signals and multipath signals. Therefore, here, the satellite whose observation data includes carrier phase observation values ​​is determined as the target satellite, and the satellite's observation data is determined as the target observation data, which can reduce the interference of error signals such as NLOS signals and multipath signals on scene recognition and increase the accuracy of scene recognition.

[0086] As a specific implementation method, refer to Figure 2-5, taking the scene types including fully occluded scenes, semi-occluded scenes and open scenes as an example, the scene recognition method may include: obtaining observation data from multiple satellites, and performing data preprocessing on the observation data of multiple satellites, wherein satellites without high-precision carrier phase observation values ​​are eliminated, and coarse positioning is performed through standard single point positioning (SPP) and other means to obtain the target latitude value of the target observation position where the receiver is located. Determine the number N of carrier phase observation values ​​without cycle slips. If N<7, it is determined that the target observation position where the receiver is located is in a fully occluded scene. If N≥7, proceed to the next step. Calculate the azimuth angles of the remaining satellites after eliminating the satellites without high-precision carrier phase observation values, and sort the satellites from small to large according to the azimuth angles of the satellites to obtain the target satellite sequence, and determine the angle θ between two adjacent satellites in the target satellite sequence, such as Figure 5 As shown, the sector-shaped area between the two satellites is the observation blind spot, and θ is the observation blind spot angle. Determine the azimuth angle between the last satellite and the first satellite in the target satellite sequence, that is, the observation blind spot angle in the north direction including the north direction line. Based on the latitude value of the receiver, determine whether the target observation position of the receiver is in a semi-obstructed scene or an open scene. If N ≥ 7 and any of the following conditions are met, it is judged as a semi-obstructed scene: latitude ≤ 39.5 degrees, the observation blind spot angle in the north direction is greater than 90 degrees, or the maximum observation blind spot angle is greater than 90 degrees; 43 degrees ≥ latitude > 39.5 degrees, the observation blind spot angle in the north direction is greater than 100 degrees, or the maximum observation blind spot angle is greater than 90 degrees; latitude > 43 degrees, the observation blind spot angle in the north direction is greater than 105 degrees, or the maximum observation blind spot angle is greater than 90 degrees. If N ≥ 7 and the semi-occluded scene conditions are not met, it is judged as an open scene, that is: latitude ≤ 39.5 degrees, the observation blind spot angle in the north direction is less than 90 degrees and the maximum observation blind spot angle is less than 90 degrees; 43 degrees ≥ latitude > 39.5 degrees, the observation blind spot angle in the north direction is less than 100 degrees and the maximum observation blind spot angle is less than 90 degrees; latitude > 43 degrees, the observation blind spot angle in the north direction is less than 105 degrees and the maximum observation blind spot angle is less than 90 degrees.

[0087] The present invention uses GNSS raw signals for scene recognition. Compared with other scene recognition methods based on vision and deep learning, it has the advantages of low computing power requirements, small amount of data required, and low cost. Moreover, when using GNSS signals for scene recognition, by eliminating satellites without high-precision carrier phase observation values ​​and using the observation data of the remaining satellites for judgment, the interference of NLOS signals and multipath signals on scene recognition is reduced. At the same time, based on the characteristic that the distribution of satellite azimuth angles changes with latitude, that is, the geometric distribution of satellites in the north direction gradually becomes sparse, resulting in a larger blind spot for satellite observation in the north direction in open scenes, the adaptability of the judgment method when the receiver is at different latitudes is set, and it has good robustness.

[0088] Figure 6 FIG. 1 is a block diagram of a scene recognition device according to an exemplary embodiment. Figure 6 The scene recognition device includes an acquisition module 601, a first determination module 602 and a second determination module 603.

[0089] The acquisition module 601 is configured to acquire target observation data of a target satellite;

[0090] The first determining module 602 is configured to determine target feature information related to the scene based on the target observation data;

[0091] The second determining module 603 is configured to determine the scene type of the target observation position corresponding to the target observation data according to the target feature information.

[0092] As an optional embodiment, the target characteristic information includes the number of carrier waves in the target observation data that do not experience cycle slips;

[0093] The second determining module 603 includes:

[0094] The first determining submodule is configured to determine that the target observation position is in a first occlusion scenario when the number of carriers without cycle slips is less than a preset threshold.

[0095] As an optional embodiment, there are multiple target satellites, and the target characteristic information further includes a maximum observation blind spot angle among the multiple observation blind spot angles, and an observation blind spot angle in the true north direction. The observation blind spot angle refers to the angle of the observation blind spot between two adjacent target satellites in the target satellite sequence. The target satellite sequence is obtained by sorting the target satellites in ascending order according to their azimuth angles.

[0096] The second determining module 603 includes:

[0097] The second determination submodule is configured to determine that the target observation position is in a second occlusion scenario when the number of carriers without cycle slips is greater than or equal to a preset number threshold and at least one of the maximum observation blind spot angle and the north direction observation blind spot angle meets the preset conditions, wherein the occlusion degree of the first occlusion scenario is greater than the occlusion degree of the second occlusion scenario.

[0098] As an optional embodiment, the scene recognition device is further specifically configured as follows:

[0099] Determine whether the maximum observation blind area angle meets the preset conditions by the following methods:

[0100] When the maximum observation blind spot angle is greater than or equal to the first angle threshold, it is determined that the maximum observation blind spot angle meets the preset condition.

[0101] As an optional embodiment, the scene recognition device is further specifically configured as follows:

[0102] Determine whether the blind spot angle in the north direction meets the preset conditions by the following methods:

[0103] Get the target latitude value of the target observation position;

[0104] Determining the target second angle threshold according to the target latitude value, the correspondence between the latitude value and the second angle threshold;

[0105] When the blind spot angle of the observation in the due north direction is greater than or equal to the target second angle threshold, it is determined that the blind spot angle of the observation in the due north direction meets the preset condition.

[0106] As an optional embodiment, the latitude value is positively correlated with the second angle threshold.

[0107] As an optional embodiment, the second determining module 603 includes:

[0108] The third determination submodule is configured to determine that the target observation position is in a third occlusion scenario when the number of carriers without cycle slips is greater than or equal to a preset number threshold and the maximum observation blind spot angle and the north direction observation blind spot angle do not meet the preset conditions, wherein the occlusion degree of the second occlusion scenario is greater than the occlusion degree of the third occlusion scenario.

[0109] As an optional embodiment, the acquisition module 601 is specifically configured to:

[0110] Obtain observation data from multiple satellites;

[0111] For each satellite, if the observation data of the satellite includes a carrier phase observation value, the satellite is determined as a target satellite, and the observation data of the satellite is determined as target observation data.

[0112] Regarding the scene recognition device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the scene recognition method, and will not be elaborated here.

[0113] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, which implement the steps of the scene recognition method provided by the present disclosure when the program instructions are executed by a processor.

[0114] The present disclosure also provides a vehicle, comprising:

[0115] a storage device for storing a computer program;

[0116] A processing device is used to execute the computer program to implement the scene recognition method provided by the present disclosure.

[0117] Figure 7 FIG2 is a block diagram of a vehicle 700 according to an exemplary embodiment. For example, vehicle 700 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or another type of vehicle. Vehicle 700 may be an autonomous vehicle or a semi-autonomous vehicle.

[0118] Reference Figure 7 Vehicle 700 may include various subsystems, such as an infotainment system 710, a perception system 720, a decision-making control system 730, a drive system 740, and a computing platform 750. Vehicle 700 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and each component of vehicle 700 may be interconnected via wired or wireless means.

[0119] In some embodiments, the infotainment system 710 may include a communication system, an entertainment system, a navigation system, and the like.

[0120] The perception system 720 may include several sensors for sensing information about the environment surrounding the vehicle 700. For example, the perception system 720 may include a global positioning system (which may be a GPS system, a BeiDou system, or other positioning systems), an inertial measurement unit (IMU), a laser radar, a millimeter-wave radar, an ultrasonic radar, and a camera.

[0121] The decision control system 730 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0122] The drive system 740 may include components that provide power to the vehicle 700. In one embodiment, the drive system 740 may include an engine, a power source, a transmission system, and wheels. The engine may be an internal combustion engine, an electric motor, an air compression engine, or a combination thereof. The engine is capable of converting energy provided by the power source into mechanical energy.

[0123] Some or all functions of the vehicle 700 are controlled by a computing platform 750. The computing platform 750 may include at least one processor 751 and a memory 752. The processor 751 may execute instructions 753 stored in the memory 752.

[0124] The processor 751 can be any conventional processor, such as a commercially available CPU. The processor can also include a graphics processor (GPU), a field programmable gate array (FPGA), a system on chip (SOC), an application specific integrated circuit (ASIC), or a combination thereof.

[0125] The memory 752 can be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0126] In addition to instructions 753 , memory 752 may also store data, such as road maps, route information, and vehicle location, direction, speed, etc. The data stored in memory 752 may be used by computing platform 750 .

[0127] In the embodiment of the present disclosure, the processor 751 may execute the instruction 753 to complete all or part of the steps of the above-mentioned scene recognition method.

[0128] In another exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program executable by a programmable device, and has a code portion for performing the above-mentioned scene recognition method when executed by the programmable device.

[0129] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0130] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A scene recognition method, characterized in that: include: Acquire target observation data of target satellite; Determining target feature information related to the scene based on the target observation data; The scene type of the target observation position corresponding to the target observation data is determined according to the target feature information.

2. The method according to claim 1, characterized in that The target characteristic information includes the number of carrier waves in the target observation data that do not have cycle slips; The determining, based on the target feature information, the scene type of the target observation position corresponding to the target observation data includes: When the number of carrier waves without cycle slips is less than a preset threshold, it is determined that the target observation position is in a first occlusion scenario.

3. The method according to claim 2, characterized in that There are multiple target satellites, and the target characteristic information further includes a maximum observation blind area angle among the multiple observation blind area angles, and an observation blind area angle in the true north direction, wherein the observation blind area angle refers to an observation blind area angle between two adjacent target satellites in a target satellite sequence, and the target satellite sequence is obtained by sorting the target satellites in ascending order according to their azimuth angles; The determining, based on the target feature information, the scene type of the target observation position corresponding to the target observation data includes: When the number of carriers that have not experienced cycle slips is greater than or equal to the preset number threshold, and at least one of the maximum observation blind spot angle and the north direction observation blind spot angle meets the preset conditions, it is determined that the target observation position is in the second occlusion scene, wherein the occlusion degree of the first occlusion scene is greater than the occlusion degree of the second occlusion scene.

4. The method according to claim 3, characterized in that Whether the maximum observation blind spot angle meets the preset conditions is determined by the following method: When the maximum observation blind spot angle is greater than or equal to the first angle threshold, it is determined that the maximum observation blind spot angle meets the preset condition.

5. The method according to claim 3, characterized in that Determine whether the north direction blind spot angle meets the preset conditions by the following method: Obtaining a target latitude value of the target observation position; Determining a target second angle threshold according to the target latitude value, a correspondence between the latitude value and the second angle threshold; When the north direction observation blind spot angle is greater than or equal to the target second angle threshold, it is determined that the north direction observation blind spot angle meets the preset condition.

6. The method according to claim 5, characterized in that The latitude value is positively correlated with the second angle threshold.

7. The method according to claim 3, characterized in that The determining, based on the target feature information, the scene type of the target observation position corresponding to the target observation data includes: When the number of carriers that have not experienced cycle slips is greater than or equal to the preset number threshold, and the maximum observation blind spot angle and the north direction observation blind spot angle do not meet the preset conditions, it is determined that the target observation position is in the third occlusion scene, wherein the occlusion degree of the second occlusion scene is greater than the occlusion degree of the third occlusion scene.

8. The method according to any one of claims 1 to 7, characterized in that The step of obtaining target observation data of a target satellite includes: Obtain observation data from multiple satellites; For each satellite, if the observation data of the satellite includes a carrier phase observation value, the satellite is determined as the target satellite, and the observation data of the satellite is determined as the target observation data.

9. A scene recognition device, characterized in that: include: an acquisition module, configured to acquire target observation data of a target satellite; A first determining module is configured to determine target feature information related to the scene based on the target observation data; The second determining module is configured to determine the scene type of the target observation position corresponding to the target observation data according to the target feature information.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the method according to any one of claims 1 to 8 is implemented.

11. A vehicle, characterized in that: include: a storage device for storing a computer program; A processing device, configured to execute the computer program to implement the method according to any one of claims 1 to 8.

Images