Map quality evaluation method and device, vehicle and readable storage medium
By acquiring vehicle location information and calculating location errors, combined with the number of times map elements are collected and confirmed, and their status, the problem of insufficient quality assessment of crowdsourced maps is solved, and efficient and accurate quality assessment is achieved.
Patent Information
- Application Number
- CN202310611819.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-26
- Publication Date
- 2024-11-26
AI Technical Summary
Existing technologies cannot effectively evaluate the quality of crowdsourced maps, resulting in uneven map quality and inadequate evaluation methods.
By acquiring the vehicle's initial location information, the vehicle's position in the high-precision map and the crowdsourced map is confirmed, the position error is calculated, and the quality of the crowdsourced map is determined by combining factors such as the number of times map elements are collected and confirmed, their integrity status, topological connectivity status, and fit.
This enables effective evaluation of the quality of crowdsourced maps, improves the reliability and accuracy of the evaluation, and ensures the objectivity and comprehensiveness of map quality assessment.
Smart Images

Figure CN119022959A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of crowdsourcing maps, and in particular to a map quality assessment method, device, vehicle and readable storage medium. Background Art
[0002] With the development of the field of smart cars, smart driving has become a hot topic. As an important part of strategic emerging industries, smart driving is an important comprehensive technology application subject that has emerged in the process from the Internet era to the artificial intelligence era. It is also one of the strategic commanding heights of the new round of economic and technological development in the world. In autonomous driving technology, maps are an indispensable element for vehicle positioning and path planning. Having high-precision maps can often provide more reliable support for smart driving.
[0003] In the existing technologies of intelligent driving, the realization of long-distance perception and lane-level planning guidance currently mainly relies on high-precision maps, which are difficult to obtain and require professional collection methods and are costly. Crowdsourcing maps have diverse collection methods and low mapping costs, so crowdsourcing map generation technology has now become the first choice for large-scale, low-cost mapping.
[0004] However, due to the different sources of crowdsourced map data, uneven quality distribution and limitations of existing mapping methods, there is currently no effective way to evaluate the map quality of crowdsourced maps. Summary of the invention
[0005] One of the purposes of the present invention is to provide a map quality assessment method to solve the technical problem that the map quality of crowdsourced maps cannot be effectively assessed in the prior art; the second purpose is to provide a map quality assessment device; the third purpose is to provide a vehicle; and the fourth purpose is to provide a readable storage medium.
[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows:
[0007] A map quality assessment method, the method comprising:
[0008] Obtaining first positioning information of the vehicle;
[0009] Based on the first positioning information, confirm a first position of the vehicle in the high-precision map and a second position of the vehicle in the crowdsourced map; wherein the first position represents a relative position of the vehicle compared to a specific map element in the high-precision map; and the second position represents a relative position of the vehicle compared to a specific map element in the crowdsourced map;
[0010] determining a position error between the first position and the second position;
[0011] Based on the position error, a map quality of the crowdsourced map is determined.
[0012] Optionally, determining the map quality of the crowdsourced map based on the position error includes:
[0013] Obtaining map elements in the crowdsourced map;
[0014] Classifying the map elements to obtain a classified map element set;
[0015] Obtaining the number of collection confirmations for each type of map element set; the number of collection confirmations indicates the number of times the geographic elements used to generate the map elements are collected and applied in the crowdsourced map;
[0016] Based on the number of collection confirmations and the position error, a map quality of the crowdsourced map is determined.
[0017] Further, the map quality includes low quality, and determining the map quality of the crowdsourced map based on the number of acquisition confirmations and the position error includes:
[0018] When the number of collection confirmations is greater than or equal to the first threshold, determining the map quality of the crowdsourced map based on the position error;
[0019] When the number of collection confirmations is less than the first threshold, it is determined that the map quality of the crowdsourcing map is low quality.
[0020] Optionally, determining the map quality of the crowdsourced map based on the position error includes:
[0021] Based on the integrity rule and topological connectivity rule of the road design, the integrity state and topological connectivity state of the map element are confirmed; the integrity rule is the integrity of the road component in the map element; the topological connectivity rule is the topological connectivity of the lanes in the map element; the integrity state includes a complete state and a missing state; the topological connectivity state includes a normal state and an abnormal state;
[0022] A map quality of the crowdsourced map is determined based on the completeness status, the topological connectivity status, and the location error.
[0023] Further, the map quality includes low quality, and determining the map quality of the crowdsourced map based on the integrity state, the topological connectivity state and the position error includes:
[0024] When the integrity state is a complete state and the topological connectivity state is a normal state, determining a map quality of the crowdsourced map based on the location error;
[0025] When the integrity state is a true state or the topology connectivity state is an abnormal state, it is determined that the map quality of the crowdsourced map is low quality.
[0026] Optionally, determining the map quality of the crowdsourced map based on the position error includes:
[0027] Determine a local map containing a fork in the crowdsourced map, and obtain a set of driving trajectory data of the vehicle driving multiple times in the fork in the local map;
[0028] Counting the number of occurrences of each driving trajectory in the driving trajectory data set, and determining the driving trajectory with the largest number of occurrences as the target trajectory;
[0029] Determining a fit between the fork in the road and the target trajectory;
[0030] Determining a first confidence level of the fork in the local map based on the fit level;
[0031] Based on the first confidence level and the position error, a map quality of the crowdsourced map is determined.
[0032] Further, the map quality includes low quality, and determining the map quality of the crowdsourced map based on the first confidence level and the position error includes:
[0033] When the first confidence level is greater than or equal to a second threshold, determining a map quality of the crowdsourced map based on the position error;
[0034] When the first confidence level is less than a second threshold, the map quality is determined to be low quality.
[0035] Optionally, confirming a first position of the vehicle in a high-precision map based on the first positioning information includes:
[0036] Acquire the positioning signal state of the first positioning information, the first speed of the vehicle, the second speed of the vehicle and a point cloud image of the surrounding environment of the vehicle; the first speed is obtained by differential calculation based on the first positioning information of the global navigation system; the second speed is obtained based on the speed measuring device of the vehicle itself;
[0037] Associating each point in the point cloud image with the lane line in the high-precision map to obtain a point-line matching relationship;
[0038] Determine a first covariance of the error of the first positioning information and a speed difference between the first speed and the second speed, and obtain a second covariance of the point-line distance between the matched point and the lane line based on the point-line matching relationship; the error of the first positioning information represents a positioning error obtained by a global positioning system through visual observation positioning and differential calculation; the first covariance, the speed difference and the second covariance can be obtained continuously;
[0039] Based on the positioning signal state, the first covariance, the speed difference, and the second covariance, confirming a second confidence level of the first positioning information;
[0040] Based on the second confidence level, the first positioning information is corrected to obtain a first position.
[0041] A map quality assessment device, characterized in that the device comprises:
[0042] A first acquisition module, used to acquire first positioning information of the vehicle;
[0043] A first confirmation module is used to confirm a first position of the vehicle in the high-precision map and a second position of the vehicle in the crowdsourced map based on the first positioning information; wherein the first position represents a relative position of the vehicle compared to a specific map element in the high-precision map; and the second position represents a relative position of the vehicle compared to a specific map element in the crowdsourced map;
[0044] A second confirmation module, used to confirm a position error between the first position and the second position;
[0045] An evaluation module is used to determine the map quality of the crowdsourced map based on the position error.
[0046] Optionally, the evaluation module comprises:
[0047] A first acquisition submodule, used to acquire map elements in the crowdsourced map;
[0048] A classification submodule, used to classify the map elements to obtain a classified map element set;
[0049] The second acquisition submodule is used to obtain the number of collection confirmations of each type of map element set; the number of collection confirmations indicates the number of times the geographical elements used to generate the map elements are collected and applied in the crowdsourced map;
[0050] The first evaluation submodule is used to determine the map quality of the crowdsourced map based on the number of collection confirmations and the location error.
[0051] Further, the map quality includes low quality, and the first evaluation submodule includes:
[0052] A first evaluation unit, configured to determine the map quality of the crowdsourced map based on the position error when the number of collection confirmations is greater than or equal to the first threshold;
[0053] The first determining unit is configured to determine that the map quality of the crowdsourced map is low quality when the number of collection confirmation times is less than the first threshold.
[0054] Optionally, the evaluation module comprises:
[0055] A state confirmation submodule, used for confirming the integrity state and topological connectivity state of the map element based on the integrity rule and topological connectivity rule of the road design; the integrity rule is the integrity of the road component in the map element; the topological connectivity rule is the topological connectivity of the lanes in the map element; the integrity state includes a complete state and a missing state; the topological connectivity state includes a normal state and an abnormal state;
[0056] The second evaluation submodule is used to determine the map quality of the crowdsourced map based on the integrity state, the topological connectivity state and the location error.
[0057] Further, the map quality includes low quality, and the second evaluation submodule includes:
[0058] a second evaluation unit, configured to determine a map quality of the crowdsourced map based on the position error when the integrity state is a complete state and the topological connectivity state is a normal state;
[0059] The second determining unit is configured to determine that the map quality of the crowdsourced map is low quality when the integrity state is a true state or the topology connectivity state is an abnormal state.
[0060] Optionally, the evaluation module comprises:
[0061] A third acquisition submodule is used to determine a local map containing a fork in the crowdsourced map, and obtain a driving trajectory data set of the vehicle driving multiple times in the fork in the local map;
[0062] A first determination submodule, used for counting the number of occurrences of each driving trajectory in the driving trajectory data set, and determining the driving trajectory with the largest number of occurrences as the target trajectory;
[0063] A second determination submodule is used to determine the degree of fit between the fork road and the target trajectory;
[0064] A third determination submodule, configured to determine a first confidence level of the fork in the local map based on the fit level;
[0065] The third evaluation submodule is used to determine the map quality of the crowdsourced map based on the first confidence and the position error.
[0066] Further, the map quality includes low quality, and the third evaluation submodule includes:
[0067] a third evaluation unit, configured to determine the map quality of the crowdsourced map based on the position error when the first confidence level is greater than or equal to a second threshold;
[0068] The third determination unit is configured to determine that the map quality is low quality when the first confidence level is less than a second threshold.
[0069] Optionally, the first confirmation module includes:
[0070] a fourth acquisition submodule, configured to acquire a positioning signal state of the first positioning information, a first speed of the vehicle, a second speed of the vehicle, and a point cloud image of the surrounding environment of the vehicle; the first speed is obtained by differential calculation based on the first positioning information of the global navigation system; the second speed is obtained based on the speed measuring device of the vehicle itself;
[0071] An association submodule, used to associate point and line data of each point in the point cloud image with the lane line in the high-precision map to obtain a point and line matching relationship;
[0072] a fifth acquisition submodule, configured to determine a first covariance of an error of the first positioning information and a speed difference between the first speed and the second speed, and to acquire a second covariance of a point-line distance between a matched point and a lane line based on the point-line matching relationship; the error of the first positioning information represents a positioning error obtained by a global positioning system through visual observation positioning and differential calculation; the first covariance, the speed difference and the second covariance may be obtained continuously;
[0073] a fifth determination submodule, configured to confirm a second confidence level of the first positioning information based on the positioning signal state, the first covariance, the speed difference, and the second covariance;
[0074] A correction submodule is used to correct the first positioning information based on the second confidence level to obtain a first position.
[0075] A vehicle, characterized by comprising the map quality assessment device as described above, used for executing the map quality assessment method as described above.
[0076] A readable storage medium, characterized in that when the instructions in the storage medium are executed by a processor of an electronic device, the electronic device executes the map quality assessment method as described above.
[0077] Beneficial effects of the present invention:
[0078] Compared with the prior art, the present invention obtains first positioning information of a vehicle; based on the first positioning information, confirms a first position of the vehicle in a high-precision map and a second position of the vehicle in a crowdsourced map; wherein the first position represents the relative position of the vehicle compared to a specific map element in the high-precision map; the second position represents the relative position of the vehicle compared to a specific map element in the crowdsourced map; confirms the position error between the first position and the second position; and based on the position error, determines the map quality of the crowdsourced map, thereby effectively evaluating the quality of the crowdsourced map and solving the technical problem that the map quality of the crowdsourced map cannot be effectively evaluated. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] Figure 1 A flowchart of a map quality assessment method provided by an embodiment of the present invention;
[0080] Figure 2 A schematic diagram of a flow chart of another map quality assessment method provided by an embodiment of the present invention;
[0081] Figure 3 A logical block diagram of a map quality assessment module provided by an embodiment of the present invention;
[0082] Figure 4 A schematic diagram of the architecture of a map quality assessment method provided by an embodiment of the present invention;
[0083] Figure 5 A schematic diagram of a positioning correction mode provided by an embodiment of the present invention;
[0084] Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0085] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.
[0086] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.
[0087] In an embodiment of the present invention, Figure 1 A flowchart of a map quality assessment method is shown, and the method may include:
[0088] Step 101, obtaining first positioning information of a vehicle.
[0089] In an embodiment of the present invention, the vehicle can be positioned by a positioning device carried on the vehicle, so that the positioning information of the vehicle can be obtained. The positioning device can be a wireless positioning device equipped with various navigation systems. The positioning device can first calibrate the position of the navigation satellite, and then receive the longitude and latitude and altitude of the positioning device measured by the navigation satellite, so as to confirm the positioning information of the positioning device. Since the positioning device can be carried on the vehicle, the positioning information of the vehicle can also be confirmed, which is called the first positioning information.
[0090] Step 102: Based on the first positioning information, confirm the first position of the vehicle in the high-precision map and the second position of the vehicle in the crowdsourced map; wherein the first position represents the relative position of the vehicle compared to a specific map element in the high-precision map; and the second position represents the relative position of the vehicle compared to a specific map element in the crowdsourced map.
[0091] In an embodiment of the present invention, the position of the vehicle in the high-precision map and the position of the vehicle in the crowdsourced map can be confirmed based on the first positioning information obtained. The position of the vehicle in the high-precision map can be referred to as the first position, and the position of the vehicle in the crowdsourced map can be referred to as the second position. Specifically, a common map element in the high-precision map and the crowdsourced map can be selected as a reference map element, so that the relative position of the vehicle in the high-precision map based on the map element and the relative position of the vehicle in the crowdsourced map based on the map element, that is, the first position and the second position, can be obtained.
[0092] Step 103: confirming a position error between the first position and the second position.
[0093] In an embodiment of the present invention, after obtaining the first position and the second position, the position error between the first position and the second position can be confirmed. The first position can be used as the true value, and then the position error between the position of the vehicle in the crowdsourced map and the position of the high-precision map can be determined based on the second position. In this process, specific map elements of the first position and the second position can be obtained as a reference.
[0094] Step 104: Determine the map quality of the crowdsourced map based on the position error.
[0095] In an embodiment of the present invention, after obtaining the position error between the first position and the second position, the map quality of the crowdsourced map can be determined according to the value of the position error. Specifically, a value can be calibrated as a third threshold, and when each value included in the position error is less than the third threshold, it can be determined that the map quality of the crowdsourced map is high; when at least one of the values included in the position error is greater than or equal to the third threshold, it can be determined that the quality of the crowdsourced map is poor.
[0096] In an embodiment of the present invention, by acquiring first positioning information of a vehicle; based on the first positioning information, confirming a first position of the vehicle in a high-precision map and a second position of the vehicle in a crowdsourced map; wherein the first position represents a relative position of the vehicle compared to a specific map element in the high-precision map; the second position represents a relative position of the vehicle compared to a specific map element in the crowdsourced map; confirming a position error between the first position and the second position; and based on the position error, determining a map quality of the crowdsourced map, the quality of the crowdsourced map can be effectively evaluated, thereby solving the technical problem that the map quality of the crowdsourced map cannot be effectively evaluated.
[0097] In an embodiment of the present invention, Figure 2 A schematic diagram of a process of another map quality assessment method provided by an embodiment of the present invention is shown. The method may include:
[0098] Step 201, obtaining first positioning information of a vehicle.
[0099] In the embodiment of the present invention, the meaning and acquisition method of the first positioning information can refer to the content of step 101 in the above embodiment, and will not be repeated here.
[0100] Step 202: Based on the first positioning information, confirm the first position of the vehicle in the high-precision map and the second position of the vehicle in the crowdsourced map.
[0101] In the embodiment of the present invention, the meanings of the first position and the second position and the confirmation method can refer to the content of step 102 in the above embodiment, and will not be repeated here.
[0102] Specifically, in an embodiment of the present invention, the following sub-steps may be included:
[0103] Sub-step 2021, obtaining the positioning signal state of the first positioning information, the first speed of the vehicle, the second speed of the vehicle and the point cloud image of the surrounding environment of the vehicle; the first speed is obtained by differential calculation based on the first positioning information of the global navigation system; the second speed is obtained based on the vehicle's own speed measuring device.
[0104] In an embodiment of the present invention, when acquiring the first positioning information, the positioning signal state corresponding to the first positioning information can be acquired at the same time. The positioning signal state includes but is not limited to invalid positioning, single-point positioning, pseudo-range differential positioning, differential global positioning, real-time kinematic (RTK) floating-point solution, real-time dynamic fixed solution, inertial navigation positioning and other states, which are represented by seven numbers from 0 to 6 respectively. Since the positioning information of the vehicle can be obtained continuously, the speed value of the vehicle can be obtained by differential calculation based on the positioning information, which is called the first speed. The vehicle can also be equipped with other speed measuring devices, so that the speed value of the vehicle can be obtained based on these speed measuring devices, which is called the second speed. In addition, the vehicle can also be equipped with an environmental perception device, through which a point cloud image of the vehicle's surrounding environment can be obtained.
[0105] Sub-step 2022, performing point-line data association between each point in the point cloud image and the lane line in the high-precision map to obtain a point-line matching relationship.
[0106] In an embodiment of the present invention, after acquiring a point cloud image of the vehicle's surroundings, the point cloud coordinates of the point cloud image can be obtained. The point cloud coordinates are based on the vehicle body as the coordinate center, so the point cloud coordinates can be converted from the vehicle body coordinate system to the world coordinate system. At the same time, the position of the environmental perception device carried by the vehicle can be obtained, which is called a visual point. In the point cloud coordinates, the point cloud coordinates of the lane lines around the vehicle are included. Therefore, first, a K-dimensional search tree can be constructed based on the lane line point cloud image of the high-precision map, so that the lane line point cloud can be traversed. In the K-dimensional search tree, the nearest k points to each visual point are found through the nearest neighbor search and indexed and sorted, so that the visual point-lane line matching and corresponding weights can be constructed, wherein the farther the obtained lane line point is from the visual point, the lower its corresponding weight.
[0107] Sub-step 2023, determines a first covariance of the error of the first positioning information and a speed difference between the first speed and the second speed, and obtains a second covariance of the point-line distance between the matching point and the lane line based on the point-line matching relationship; the error of the first positioning information represents the positioning error obtained by the global positioning system through visual observation positioning and differential calculation.
[0108] In an embodiment of the present invention, after obtaining the first positioning information, the navigation and positioning system carried by the positioning device can obtain the error of the first positioning information through the constraint algorithm of the corresponding positioning system. Among them, the constraint algorithm can be that when performing positioning, visual observation positioning can be performed at intervals of a fixed number of frames to obtain the positioning error. Based on this error, the covariance of the error can be calculated, which is called the first covariance. The difference between the first speed and the second speed can also be calculated and the absolute value can be taken, so that the speed difference between the first speed and the second speed can be obtained. In addition, according to the point-line matching relationship, the distance between the matched visual point and the lane line formed by the point cloud of the lane line can be calculated, so that the covariance of the distance can be calculated, which is called the second covariance.
[0109] Sub-step 2024: confirming a second confidence level of the first positioning information based on the positioning signal state, the first covariance, the speed difference, and the second covariance.
[0110] In an embodiment of the present invention, after obtaining the first covariance, the average value of the first covariance can be calculated and calibrated. Specifically, a certain positive integer can be set as the number of cycles, and the number of acquisitions of the first covariance can be accumulated after obtaining the first covariance. When the number of acquisitions is greater than or equal to the number of cycles, the average value of each first covariance obtained is calculated, which is called the average value of the first covariance. Similarly, the speed difference and the second covariance can also obtain the average value of the speed difference and the average value of the second covariance by this method. In addition, when the first average value calibration is completed, it can be confirmed that the initialization process is completed, and then the confidence of the first positioning information can be calculated in combination with the signal strength model of the differential global navigation positioning system, which is called the second confidence. Among them, the signal strength model of the differential global navigation positioning system can be expressed as the following formula 1:
[0111]
[0112] In the above model, Indicates signal strength, status indicates positioning signal status, cov indicates first covariance, V gps Indicates the first speed, V imu Indicates the second speed.
[0113] The calculation formula of the second confidence level is shown in Formula 2 below:
[0114]
[0115] Among them, α represents the second confidence level, represents the signal strength, cov represents the first covariance, cov avgrepresents the mean value of the first covariance, ΔV represents the speed difference, ΔV avg represents the average value of the speed difference, cov error represents the second covariance, cov error avg represents the mean of the second covariance.
[0116] Through the above method, the value of the second confidence level can be obtained.
[0117] Sub-step 2025: Based on the second confidence level, correct the first positioning information to obtain a first position.
[0118] In an embodiment of the present invention, after obtaining the second confidence level, the first positioning information can be corrected to obtain a corrected position, referred to as the first position. Among them, the first positioning information can be corrected by the differential positioning principle of the satellite inertial navigation (GNSS-IMU) integrated navigation system. The differential positioning principle refers to setting one or more receivers within a certain geographical range, and using a receiver with known precise coordinates as a differential reference station. The reference station continuously receives positioning signals and compares them with the known position and distance data of the reference station to calculate the differential correction amount. Then, the reference station will send this differential correction amount to the mobile station within its range for data correction, thereby reducing or even eliminating the errors caused by satellite clocks, satellite ephemeris, ionospheric delays and tropospheric delays, and improving positioning accuracy. In addition, the satellite inertial navigation integrated navigation system can also realize the time service function, which can provide a unified time for multi-sensor data sources.
[0119] In an embodiment of the present invention, by acquiring the positioning signal state of the first positioning information, the first speed of the vehicle, the second speed of the vehicle and the point cloud image of the surrounding environment of the vehicle, the first speed is obtained by differential calculation based on the first positioning information of the global navigation system; the second speed is obtained based on the speed measuring device of the vehicle itself, and point-line data association is performed between each point in the point cloud image and the lane line in the high-precision map to obtain a point-line matching relationship, and a first covariance of the error of the first positioning information and a speed difference between the first speed and the second speed are determined. Based on the point-line matching relationship, a second covariance of the point-line distance between the matched point and the lane line is obtained. The error of the first positioning information represents the positioning error obtained by the global positioning system through visual observation positioning and differential calculation. Based on the positioning signal state, the first covariance, the speed difference, and the second covariance, a second confidence of the first positioning information is confirmed. Based on the second confidence, the first positioning information is corrected to obtain a first position. The first positioning information can be corrected to obtain a more accurate positioning result, thereby providing more reliable support for the quality assessment of crowdsourced maps.
[0120] Step 203: confirming a position error between the first position and the second position.
[0121] In the embodiment of the present invention, the meaning and confirmation method of the position error can refer to the content of step 103 in the above embodiment, and will not be repeated here.
[0122] Step 204: Obtain map elements of the crowdsourced map.
[0123] In an embodiment of the present invention, after obtaining a crowdsourced map, the crowdsourced map may be parsed to obtain map elements of the crowdsourced map. The parsing method may include a threshold segmentation method, a template matching method, and a deep learning method. Through parsing, various map elements of the crowdsourced map may be obtained.
[0124] Step 205: classify the map elements to obtain a classified map element set.
[0125] In an embodiment of the present invention, after obtaining various map elements of the crowdsourced map, the map elements can be classified based on the level evaluation model of the map elements. The model can classify map elements into five categories: one is basic, which may include lane lines, stop lines, pedestrian crossings / non-motorized lanes, bus stops, bridges and tunnels; the second is safety, which may include speed limits, turning and slope signs, and alternative safety zone signs; the third is comfort, which may include speed bumps and smooth curves; the fourth is supplementary information, which may include traffic signs, traffic lights, etc.; the fifth is environmental richness information, which may include weather, surrounding environment, and road conditions. After classifying the map elements through the above model, a classified set of map elements can be obtained.
[0126] Step 206: Obtain the number of collection confirmations for each type of map element set.
[0127] In an embodiment of the present invention, after classifying the map elements, the number of collection confirmations of various categories of map elements in the crowdsourcing map drawing process can be obtained. The number of collection confirmations can represent the number of times the geographical elements used to generate the map elements are collected and applied in the crowdsourcing map. For example, there can be a straight lane in the crowdsourcing map, and all vehicles with map element collection functions and crowdsourcing map drawing functions can collect information about the straight lane when passing through the straight lane, and the collected straight lane information can be used in the crowdsourcing map drawing process. Each time a vehicle completes this process, the number of collection confirmations can be increased by one.
[0128] Step 207: Determine the map quality of the crowdsourced map based on the number of collection confirmations and the location error.
[0129] In an embodiment of the present invention, the map quality of the crowdsourced map may be determined according to the size of the location error and the number of collection confirmation times.
[0130] Specifically, step 207 may include the following sub-steps:
[0131] Sub-step 2071, when the number of collection confirmations is greater than or equal to the first threshold, determining the map quality of the crowdsourced map based on the position error.
[0132] Sub-step 2072, when the number of collection confirmations is less than the first threshold, determining that the map quality of the crowdsourced map is low quality.
[0133] In an embodiment of the present invention, a numerical value can be calibrated as the first threshold, and then the first threshold and the number of collection confirmations can be compared. If the number of collection confirmations is greater than or equal to the first threshold, it can be considered that the map element corresponding to this number of collection confirmations has high quality. Then, each map element in the crowdsourced map is traversed. If the number of collection confirmations of all map elements is greater than or equal to the first threshold, the map quality of the crowdsourced map can be evaluated in combination with the position error. When the number of collection confirmations of at least one map element in the crowdsourced map is less than the first threshold, the crowdsourced map can be considered to be of low quality.
[0134] In an embodiment of the present invention, by acquiring first positioning information of a vehicle, confirming a first position of the vehicle in a high-precision map and a second position of the vehicle in a crowdsourced map based on the first positioning information, confirming a position error between the first position and the second position, acquiring map elements of the crowdsourced map, classifying the map elements to obtain a classified map element set, acquiring the number of collection confirmations for each type of the map element set, determining the map quality of the crowdsourced map based on the number of collection confirmations and the position error, a comprehensive evaluation of the crowdsourced map can be performed based on the map elements of the crowdsourced map and the position error of the first positioning information, thereby improving the reliability and accuracy of the evaluation.
[0135] In an embodiment of the present invention, when the number of collection confirmations is greater than or equal to the first threshold, the map quality of the crowdsourced map is determined based on the position error; when the number of collection confirmations is less than the first threshold, the map quality of the crowdsourced map is determined to be low quality, thereby increasing the direction of crowdsourced map evaluation and making the evaluation results of crowdsourced maps more objective and accurate.
[0136] Step 208: confirming the integrity status and topological connectivity status of the map element based on the integrity rule and topological connectivity rule of the road design.
[0137] In an embodiment of the present invention, lanes need to follow integrity and topological connectivity rules during the design process. The integrity rule may indicate the integrity of road components in a map element, such as a curve sign should be provided at a curve, and a fork sign should be provided at a fork. The topological connectivity rule may be the topological connectivity of lanes in a map element, such as whether a lane is redundant, whether a lane is broken or misaligned, and whether a lane has a slight intersection. The integrity state may include a complete state and a missing state, and the topological connectivity state may include a normal state and an abnormal state.
[0138] Step 209: determining the map quality of the crowdsourced map based on the integrity status, the topological connectivity status and the location error.
[0139] In an embodiment of the present invention, the map quality of the crowdsourced map may be confirmed based on the integrity of the map elements and the topological connectivity of the lanes and the position error of the first positioning information.
[0140] Specifically, step 209 may include the following sub-steps:
[0141] Sub-step 2091, when the integrity state is a complete state and the topology connectivity state is a normal state, determining the map quality of the crowdsourced map based on the position error.
[0142] Sub-step 2092, when the integrity state is a true state or the topology connectivity state is an abnormal state, determining that the map quality of the crowdsourced map is low quality.
[0143] In an embodiment of the present invention, the integrity of the road elements of the crowdsourced map and the topological connectivity of the roads can be detected, so that the integrity status of the road elements and the topological connectivity status of the roads can be obtained. The integrity status can include a complete state and a missing state, and the topological connectivity status can include a normal state and an abnormal state. If the integrity status of the map elements of the crowdsourced map is a complete state, and the topological connectivity status of the roads is a normal state, the map quality of the crowdsourced map can be evaluated in combination with the position error; if the integrity status of the map elements is a missing state, or the topological connectivity status of the roads is an abnormal state, it can be confirmed that the crowdsourced map is of low quality. Among them, when there is no redundancy in the lanes, no broken or misaligned connections, and no slight intersections, it can be determined that the topological connectivity status of the road is a normal state; when there is redundancy in the lanes, or there are broken or misaligned connections, or there are slight intersections, it can be determined that the topological connectivity status of the road is an abnormal state.
[0144] In an embodiment of the present invention, the integrity status and topological connectivity status of the map elements are confirmed based on the integrity rules and topological connectivity rules of road design, and the map quality of the crowdsourced map is determined based on the integrity status, the topological connectivity status and the position error. The quality of the crowdsourced map can be evaluated from three perspectives: the integrity and topological connectivity of the crowdsourced map and the position error of the first positioning information, thereby further improving the reliability and accuracy of the map quality evaluation.
[0145] In an embodiment of the present invention, by determining the map quality of the crowdsourced map based on the position error when the integrity state is a complete state and the topological connectivity state is a normal state, and determining that the map quality of the crowdsourced map is of low quality when the integrity state is a true state or the topological connectivity state is an abnormal state, the direction of crowdsourced map evaluation is increased, thereby making the evaluation result of the crowdsourced map more objective and accurate.
[0146] Step 210, determining a local map including a fork in the road from the crowdsourced map, and obtaining a set of driving trajectory data of the vehicle driving multiple times in the fork in the local map.
[0147] In an embodiment of the present invention, a local map containing a fork in the road can be determined in the crowdsourced map through lane-level map matching. When a vehicle passes through the fork in the road, a driving trajectory can be generated and saved, and the driving trajectory data can be uploaded to a cloud server. The vehicle can call the driving trajectory data corresponding to the current fork in the road saved by all cloud servers, thereby forming a data set.
[0148] Step 211 , counting the number of occurrences of each driving trajectory in the driving trajectory data set, and determining the driving trajectory with the largest number of occurrences as the target trajectory.
[0149] In an embodiment of the present invention, after obtaining the driving trajectory data set corresponding to the current fork in the road, the number of occurrences of each driving trajectory in the driving trajectory data set can be counted by the precise matching method. Specifically, one of the paths can be randomly selected as the first template, and the number of occurrences can be accumulated once, and then all the remaining driving trajectories in the driving trajectory data set can be traversed and compared with the first template. If they completely overlap with the first template, the number of occurrences of the first template is accumulated. If they do not overlap, the driving trajectory is used as the second template and the number of occurrences is accumulated once, and so on. After traversing all the driving trajectories in the driving trajectory data set, the number of occurrences of each driving trajectory can be obtained, so that the driving trajectory with the most occurrences can be determined, and the driving trajectory can be used as the target trajectory.
[0150] Step 212, determining the degree of fit between the fork in the road and the target trajectory.
[0151] In an embodiment of the present invention, after obtaining the target trajectory, the degree of fit between the fork road and the target trajectory can be determined based on a mathematical method. For example, scatter points of the fork road and the target trajectory can be collected by MATLAB, and the degree of fit can be obtained through these scatter points; or a mathematical model can be established to calculate the degree of fit between the fork road and the target trajectory by means of computer program code.
[0152] Step 213: Determine a first confidence level of the fork in the local map based on the fit level.
[0153] In an embodiment of the present invention, the degree of fit between the fork road and the target trajectory can be directly determined as the first confidence of the local map. The degree of fit can represent the similarity between the fork road and the target trajectory, and therefore, the degree of fit can be used to represent the confidence of the fork road in the local map.
[0154] Step 214: determine the map quality of the crowdsourced map based on the first confidence level and the position error.
[0155] In an embodiment of the present invention, after obtaining the first confidence level, the map quality of the crowdsourced map may be confirmed in combination with the location error.
[0156] Specifically, step 214 may include the following sub-steps:
[0157] Sub-step 2141, when the first confidence level is greater than or equal to a second threshold, determining the map quality of the crowdsourced map based on the position error.
[0158] Sub-step 2142: when the first confidence level is less than a second threshold, determining that the map quality is low quality.
[0159] In an embodiment of the present invention, a numerical value may be calibrated as the second threshold, and then the second threshold and the first confidence level may be compared. When the first confidence level is greater than or equal to the second threshold, the quality of the crowdsourced map may be determined in combination with the position error; when the first confidence level is less than the second threshold, the map quality may be determined to be low quality.
[0160] In an embodiment of the present invention, a local map including a fork in the road is determined from the crowdsourced map, and a driving trajectory data set of the vehicle driving multiple times in the fork in the local map is obtained, the number of occurrences of each driving trajectory in the driving trajectory data set is counted, and the driving trajectory with the largest number of occurrences is determined as the target trajectory, the degree of fit between the fork in the road and the target trajectory is determined, and a first confidence of the fork in the local map is determined based on the degree of fit, and the map quality of the crowdsourced map is determined based on the first confidence and the position error. The crowdsourced map can be comprehensively evaluated based on the confidence of the lane and the position error of the first positioning information, thereby improving the reliability and accuracy of the map evaluation.
[0161] In an embodiment of the present invention, by determining the map quality of the crowdsourced map based on the position error when the first confidence is greater than or equal to the second threshold, and determining that the map quality is low quality when the first confidence is less than the second threshold, the direction of crowdsourced map evaluation is increased, thereby making the evaluation results of the crowdsourced map more objective and accurate.
[0162] It should be noted that, in the embodiments of the present invention, various evaluation methods can be integrated to obtain a comprehensive evaluation method. Specifically, when the position error is less than the third threshold, and the number of collection confirmations is greater than or equal to the first threshold, and the integrity state is complete, and the topological connectivity is normal, and the first confidence is greater than or equal to the second threshold, it can be confirmed that the crowdsourced map has high quality; when the position error is greater than or equal to the third threshold, or the number of collection confirmations is less than the first threshold, or the integrity state is missing, or the topological connectivity is abnormal, or the first confidence is less than the second threshold, it can be confirmed that the crowdsourced map is of low quality.
[0163] In an embodiment of the present invention, Figure 3 The following is a logic block diagram of a map quality assessment module provided by an embodiment of the present invention. The map quality assessment module 300 may include:
[0164] A first acquisition module 301 is used to acquire first positioning information of a vehicle;
[0165] A first confirmation module 302 is used to confirm a first position of the vehicle in the high-precision map and a second position of the vehicle in the crowdsourced map based on the first positioning information; wherein the first position represents a relative position of the vehicle compared to a specific map element in the high-precision map; and the second position represents a relative position of the vehicle compared to a specific map element in the crowdsourced map;
[0166] A second confirmation module 303, used to confirm the position error between the first position and the second position;
[0167] The evaluation module 304 is used to determine the map quality of the crowdsourced map based on the position error.
[0168] Optionally, the evaluation module 304 includes:
[0169] A first acquisition submodule, used to acquire map elements in the crowdsourced map;
[0170] A classification submodule, used to classify the map elements to obtain a classified map element set;
[0171] The second acquisition submodule is used to obtain the number of collection confirmations of each type of map element set; the number of collection confirmations indicates the number of times the geographical elements used to generate the map elements are collected and applied in the crowdsourced map;
[0172] The first evaluation submodule is used to determine the map quality of the crowdsourced map based on the number of collection confirmations and the location error.
[0173] Further, the map quality includes low quality, and the first evaluation submodule includes:
[0174] A first evaluation unit, configured to determine the map quality of the crowdsourced map based on the position error when the number of collection confirmations is greater than or equal to the first threshold;
[0175] The first determining unit is configured to determine that the map quality of the crowdsourced map is low quality when the number of collection confirmation times is less than the first threshold.
[0176] Optionally, the evaluation module 304 includes:
[0177] A state confirmation submodule, used for confirming the integrity state and topological connectivity state of the map element based on the integrity rule and topological connectivity rule of the road design; the integrity rule is the integrity of the road component in the map element; the topological connectivity rule is the topological connectivity of the lanes in the map element; the integrity state includes a complete state and a missing state; the topological connectivity state includes a normal state and an abnormal state;
[0178] The second evaluation submodule is used to determine the map quality of the crowdsourced map based on the integrity state, the topological connectivity state and the location error.
[0179] Further, the map quality includes low quality, and the second evaluation submodule includes:
[0180] a second evaluation unit, configured to determine a map quality of the crowdsourced map based on the position error when the integrity state is a complete state and the topological connectivity state is a normal state;
[0181] The second determining unit is configured to determine that the map quality of the crowdsourced map is low quality when the integrity state is a true state or the topology connectivity state is an abnormal state.
[0182] Optionally, the evaluation module 304 includes:
[0183] A third acquisition submodule is used to determine a local map containing a fork in the crowdsourced map, and obtain a driving trajectory data set of the vehicle driving multiple times in the fork in the local map;
[0184] A first determination submodule, used for counting the number of occurrences of each driving trajectory in the driving trajectory data set, and determining the driving trajectory with the largest number of occurrences as the target trajectory;
[0185] A second determination submodule is used to determine the degree of fit between the fork road and the target trajectory;
[0186] A third determination submodule, configured to determine a first confidence level of the fork in the local map based on the fit level;
[0187] The third evaluation submodule is used to determine the map quality of the crowdsourced map based on the first confidence and the position error.
[0188] Further, the map quality includes low quality, and the third evaluation submodule includes:
[0189] a third evaluation unit, configured to determine the map quality of the crowdsourced map based on the position error when the first confidence level is greater than or equal to a second threshold;
[0190] The third determination unit is configured to determine that the map quality is low quality when the first confidence level is less than a second threshold.
[0191] Optionally, the first confirmation module 302 includes:
[0192] a fourth acquisition submodule, configured to acquire a positioning signal state of the first positioning information, a first speed of the vehicle, a second speed of the vehicle, and a point cloud image of the surrounding environment of the vehicle; the first speed is obtained by differential calculation based on the first positioning information of the global navigation system; the second speed is obtained based on the speed measuring device of the vehicle itself;
[0193] An association submodule, used to associate point and line data of each point in the point cloud image with the lane line in the high-precision map to obtain a point and line matching relationship;
[0194] a fifth acquisition submodule, configured to determine a first covariance of an error of the first positioning information and a speed difference between the first speed and the second speed, and to acquire a second covariance of a point-line distance between a matched point and a lane line based on the point-line matching relationship; the error of the first positioning information represents a positioning error obtained by a global positioning system through visual observation positioning and differential calculation; the first covariance, the speed difference and the second covariance may be obtained continuously;
[0195] a fifth determination submodule, configured to confirm a second confidence level of the first positioning information based on the positioning signal state, the first covariance, the speed difference, and the second covariance;
[0196] A correction submodule is used to correct the first positioning information based on the second confidence level to obtain a first position.
[0197] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0198] The map quality assessment device has the same advantages as the map quality assessment method described in the above embodiment over the prior art, which will not be described in detail here.
[0199] In an embodiment of the present invention, Figure 4 A schematic diagram of the architecture of a map quality assessment method provided by an embodiment of the present invention is shown. Among them, the crowdsourcing map quality assessment architecture includes position accuracy evaluation and map element evaluation. In the position accuracy evaluation, the vehicle's own signal categories include vehicle speed, wheel speed, steering, attitude data (IMU), positioning signal (GNSS), and camera. In addition, other factors such as lighting and external environment can also be obtained. These signals can be transmitted to the signal processing unit, and through signal processing and attitude solution, the positioning algorithm is performed to obtain the confidence of the signal. In the map element evaluation, the map elements of the crowdsourcing map are first obtained, including lane information, road components, and auxiliary information. With this information, conventional quality detection is performed on the map elements, such as integrity detection, topological connectivity detection, etc. In addition, deep quality detection of map elements can also be performed through driving trajectories, traffic rules, curve divergence intersections, etc.
[0200] In an embodiment of the present invention, Figure 5 FIG. 1 is a schematic diagram of a positioning correction mode provided by an embodiment of the present invention. Where M represents the positioning missing section and the sections before and after the positioning missing section, and N maxIndicates the positioning point obtained when positioning offset occurs, N min Indicates the minimum number of positioning points required for positioning correction after the positioning is restored to normal, and flag is the identification positioning point; Figure 5 The lower part of represents the road map after positioning and repair based on the identification points and the positioning points after the positioning is restored to normal.
[0201] An embodiment of the present invention provides a vehicle, comprising the map quality assessment device as described in the above embodiment, and configured to execute the map quality assessment method as described in the above embodiment.
[0202] An embodiment of the present invention provides a readable storage medium. When instructions in the storage medium are executed by a processor of an electronic device, the electronic device can execute the aforementioned map quality assessment method.
[0203] An embodiment of the present invention provides an electronic device, see Figure 6 The electronic device 60 includes: a processor 601, a memory 602, and a computer program 6021 stored in the memory 602 and executable on the processor 601. When the processor 601 executes the program, the map quality assessment method of the aforementioned embodiment is implemented.
[0204] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other equipment. According to the above description, it is obvious that the structure required for constructing such a system is. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages can be utilized to realize the content of the present invention described herein, and the description of the above specific language is for disclosing the best mode of the present invention.
[0205] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.
[0206] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recited in each claim. More specifically, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.
[0207] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.
[0208] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the sorting device according to the present invention. The present invention may also be implemented as a device or apparatus program for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.
[0209] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.
[0210] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0211] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
[0212] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
[0213] It should be noted that the various data-related processes in the embodiments of the present invention are all carried out in compliance with the corresponding data protection laws and policies of the country where the device is located and with the authorization given by the corresponding device owner.
[0214] The above embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Any equivalent substitution or change made by a person skilled in the art based on the present invention is within the protection scope of the present invention.
Claims
1. A map quality assessment method, characterized in that: The method comprises: Obtaining first positioning information of the vehicle; Based on the first positioning information, confirm a first position of the vehicle in the high-precision map and a second position of the vehicle in the crowdsourced map; wherein the first position represents a relative position of the vehicle compared to a specific map element in the high-precision map; and the second position represents a relative position of the vehicle compared to a specific map element in the crowdsourced map; determining a position error between the first position and the second position; A map quality of the crowdsourced map is determined based on the position error.
2. The method according to claim 1, characterized in that The determining the map quality of the crowdsourced map based on the position error comprises: Obtaining map elements in the crowdsourced map; Classifying the map elements to obtain a classified map element set; Obtaining the number of collection confirmations for each type of map element set; the number of collection confirmations represents the number of times the geographic elements used to generate the map elements are collected and applied in the crowdsourced map; Based on the number of collection confirmations and the position error, a map quality of the crowdsourced map is determined.
3. The method according to claim 2, wherein the map quality comprises low quality, characterized in that: The determining the map quality of the crowdsourced map based on the number of collection confirmations and the location error includes: When the number of collection confirmations is greater than or equal to the first threshold, determining the map quality of the crowdsourced map based on the position error; When the number of collection confirmations is less than the first threshold, it is determined that the map quality of the crowdsourcing map is low quality.
4. The method according to claim 1, characterized in that The determining the map quality of the crowdsourced map based on the position error comprises: Based on the integrity rule and topological connectivity rule of the road design, the integrity state and topological connectivity state of the map element are confirmed; the integrity rule is the integrity of the road component in the map element; the topological connectivity rule is the topological connectivity of the lanes in the map element; the integrity state includes a complete state and a missing state; the topological connectivity state includes a normal state and an abnormal state; A map quality of the crowdsourced map is determined based on the completeness status, the topological connectivity status, and the location error.
5. The method according to claim 4, wherein the map quality comprises low quality, characterized in that: The determining the map quality of the crowdsourced map based on the integrity state, the topological connectivity state and the position error comprises: When the integrity state is a complete state and the topological connectivity state is a normal state, determining a map quality of the crowdsourced map based on the location error; When the integrity state is a true state or the topology connectivity state is an abnormal state, it is determined that the map quality of the crowdsourced map is low quality.
6. The method according to claim 1, characterized in that The determining the map quality of the crowdsourced map based on the position error comprises: Determine a local map containing a fork in the crowdsourced map, and obtain a set of driving trajectory data of the vehicle driving multiple times in the fork in the local map; Counting the number of occurrences of each driving trajectory in the driving trajectory data set, and determining the driving trajectory with the largest number of occurrences as the target trajectory; Determining a fit between the fork in the road and the target trajectory; Determining a first confidence level of the fork in the local map based on the fit level; Based on the first confidence level and the position error, a map quality of the crowdsourced map is determined.
7. The method according to claim 6, wherein the map quality comprises low quality, characterized in that: The determining, based on the first confidence level and the position error, the map quality of the crowdsourced map includes: When the first confidence level is greater than or equal to a second threshold, determining a map quality of the crowdsourced map based on the position error; When the first confidence level is less than a second threshold, the map quality is determined to be low quality.
8. The method according to claim 1, characterized in that The step of confirming a first position of the vehicle in a high-precision map based on the first positioning information includes: Acquire the positioning signal state of the first positioning information, the first speed of the vehicle, the second speed of the vehicle and a point cloud image of the surrounding environment of the vehicle; the first speed is obtained by differential calculation based on the first positioning information of the global navigation system; the second speed is obtained based on the speed measuring device of the vehicle itself; Associating each point in the point cloud image with the lane line in the high-precision map to obtain a point-line matching relationship; Determine a first covariance of the error of the first positioning information and a speed difference between the first speed and the second speed, and obtain a second covariance of the point-line distance between the matched point and the lane line based on the point-line matching relationship; the error of the first positioning information represents a positioning error obtained by a global positioning system through visual observation positioning and differential calculation; Based on the positioning signal state, the first covariance, the speed difference and the second covariance, confirming a second confidence level of the first positioning information; Based on the second confidence level, the first positioning information is corrected to obtain a first position.
9. A map quality assessment device, characterized in that: The device comprises: A first acquisition module, used to acquire first positioning information of the vehicle; A first confirmation module is used to confirm a first position of the vehicle in the high-precision map and a second position of the vehicle in the crowdsourced map based on the first positioning information; wherein the first position represents a relative position of the vehicle compared to a specific map element in the high-precision map; and the second position represents a relative position of the vehicle compared to a specific map element in the crowdsourced map; A second confirmation module, used to confirm a position error between the first position and the second position; An evaluation module is used to determine the map quality of the crowdsourced map based on the position error.
10. A vehicle, characterized in that: It comprises the map quality assessment device as claimed in claim 9, which is used to execute the map quality assessment method as claimed in any one of claims 1-8.
11. A readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device executes the map quality assessment method as described in any one of claims 1 to 8.
Citation Information
Cited By
Fusion positioning precision test method and device, electronic equipment and storage medium
CN117146863A
