Vehicle model coordinate system calibration method and device

The vehicle model coordinate system calibration method addresses coordinate system inconsistencies by generating and comparing probabilities to align vehicle trajectory data with road networks, improving accuracy and consistency for data integration.

CN119085652APending Publication Date: 2024-12-06BEIJING INST OF TECH XINYUAN INFORMATION TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411199923.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The diversity of trajectory coordinate systems used by different car companies leads to inconsistencies in vehicle trajectory data, causing information loss, error accumulation, and inaccurate results in data processing and analysis.

Method used

A vehicle model coordinate system calibration method that involves generating observation and state transition probabilities for trajectory points, determining path likelihood probabilities, and identifying positioning trajectory points to align the vehicle model's coordinate system with a candidate road system, using Gaussian distribution and threshold-based ratio comparisons.

Benefits of technology

This method enhances the accuracy and consistency of vehicle trajectory data, enabling effective comparison and integration of data from different sources, reducing errors and deviations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119085652A_ABST
    Figure CN119085652A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of Internet of Vehicles, and discloses a vehicle type coordinate system calibration method, which comprises the following steps: acquiring a target trajectory which is each of a plurality of representative trajectories of a vehicle type; for a target track point in the target track, generating an observation probability that the target track point is mapped to a target mapping point of the target candidate road to obtain an observation probability set corresponding to the target track, and generating a state transition probability between the target mapping point and a next target mapping point to obtain a state transition probability set of the target candidate road; generating a path likelihood probability of the target candidate road according to the observation probability set and the state transition probability set, and determining a target path corresponding to the path likelihood probability from the candidate road set; the positioning track points are determined in all the track points of the target track, and the coordinate system of the vehicle model is determined based on the number of the positioning track points in the plurality of representative tracks and the number of all the track points in the plurality of representative tracks, so that the accuracy of the calibration result is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the technical field of vehicle networking, and in particular to a method and device for calibrating a vehicle model coordinate system. Background Art

[0002] The vehicle trajectory data involved in the Internet of Vehicles comes from different car companies, and different car companies may use different coordinate systems to record and process trajectory data. The GPS84 coordinate system is a universal geographic coordinate system that is widely used in the global positioning system. The GCJ02 coordinate system is a Chinese-specific encrypted coordinate system used to protect the security and privacy of geographic information. Therefore, different data sources may use different coordinate systems, which leads to the diversity and inconsistency of trajectory data.

[0003] The diversity of trajectory coordinate systems brings challenges to data processing and analysis. If these data are not calibrated in a unified coordinate system, trajectory data from different data sources cannot be directly compared and integrated. This may lead to problems such as information loss, error accumulation, and inaccurate results. Therefore, how to accurately calibrate the trajectory coordinate system to ensure the accuracy and consistency of vehicle trajectory data is an urgent problem that needs to be solved. Summary of the invention

[0004] In view of this, the present disclosure provides a vehicle model coordinate system calibration method to solve the problem of how to accurately calibrate the trajectory coordinate system.

[0005] In a first aspect, the present disclosure provides a vehicle model coordinate system calibration method, the method comprising: obtaining a target trajectory, the target trajectory being each trajectory in a plurality of representative trajectories of a vehicle model; for a target trajectory point in the target trajectory, generating an observation probability of mapping the target trajectory point to a target mapping point of a target candidate road, obtaining an observation probability set corresponding to the target trajectory, and generating a state transition probability between the target mapping point and the target next mapping point, obtaining a state transition probability set of the target candidate road; wherein the target trajectory point is each trajectory point in the target trajectory, the target candidate road is each candidate road in the candidate road set, the target next mapping point is a mapping point of the target next trajectory point on the target candidate road, and the target next trajectory point is a next trajectory point of the target trajectory point; according to the observation probability set and the state transition probability set, generating a path likelihood probability of the target candidate road, and determining a target path corresponding to the path likelihood probability from the candidate road set; determining a positioning trajectory point among all trajectory points of the target trajectory, and determining the coordinate system of the vehicle model based on the number of positioning trajectory points in the plurality of representative trajectories and the number of all trajectory points in the plurality of representative trajectories; wherein the distance between the positioning trajectory point and its mapping point on the target path is less than a distance threshold.

[0006] The vehicle model coordinate system calibration method provided in this embodiment can more accurately identify the target path in the road network set corresponding to the target trajectory by generating and comparing the observation probability and state transition probability of the target trajectory in multiple representative trajectories of the vehicle model on different candidate roads, and then determine the coordinate system of the vehicle model based on the proportion of the positioning trajectory points in the multiple representative trajectories of the vehicle model in all trajectory points. This process improves the accuracy of the coordinate system corresponding to the determined vehicle model trajectory, thereby ensuring the accuracy and consistency of the vehicle trajectory data involved in the Internet of Vehicles, which is facilitating the subsequent comparison and integration of trajectory data.

[0007] In some optional implementations of this embodiment, multiple representative trajectories of a vehicle model are determined based on the following steps: a candidate trajectory sample area is determined based on at least one of the following: the distribution of vehicle model data of vehicles traveling in the area meets the distribution requirements; the network infrastructure in the area meets the facility requirements; the road network data in the area meets the road network requirements; from the candidate trajectory sample area, a preset number of target vehicles corresponding to the vehicle model are determined; for each target vehicle, a predetermined number of target driving trajectories are selected to obtain multiple representative trajectories of the vehicle model; wherein the target driving trajectory meets at least one of the following: includes a preset number of trajectory points; the driving speed of each trajectory point is greater than a preset speed.

[0008] In some optional implementations of the present embodiment, an observation probability of mapping a target trajectory point to a target mapping point of a target candidate road is generated, including: selecting a set of candidate roads whose distance from the target trajectory point is less than a given radius in a target road network data set; determining a target distance between the target trajectory point and the target mapping point mapped to the target candidate road; and using a Gaussian distribution function to generate a probability density function of mapping the target trajectory point to the target mapping point according to the target distance as the observation probability of the target trajectory point being mapped to the target mapping point.

[0009] In some optional implementations of the present embodiment, the target trajectory point is mapped to the target mapping point of the target candidate road based on at least one of the following determinations: when the foot of a perpendicular line from the target trajectory point to the target candidate road is located on the target candidate road, the foot of the perpendicular is used as the target mapping point of the target trajectory point on the target candidate road; when the foot of the perpendicular line from the target trajectory point to the target candidate road is located on the extension line of the target candidate road, the endpoint of the target candidate road adjacent to the foot of the perpendicular is used as the target mapping point of the target trajectory point on the target candidate road.

[0010] In some optional implementations of the present embodiment, generating a state transition probability between a target mapping point and a target next mapping point includes: determining a first distance between a target trajectory point and a target next trajectory point, and determining a second distance between the target mapping point and the target next mapping point; respectively determining a minimum value and a maximum value in the first distance and the second distance; and using a ratio of the minimum value to the maximum value as a state transition probability between the target mapping point and the target next mapping point.

[0011] In some optional implementations of the present embodiment, a path likelihood probability of a target candidate road is generated based on an observation probability set and a state transition probability set, and a target path corresponding to the path likelihood probability is determined from the candidate road set, including: generating the product of the observation probability of mapping a target trajectory point to a target mapping point of the target candidate road and the state transition probability between the target mapping point and the next mapping point of the target; for all trajectory points in the target trajectory, calculating the sum of the products as the path likelihood probability of the target candidate road; and selecting the candidate road with the largest path likelihood probability from the candidate road set as the target path of the representative trajectory in the candidate road set.

[0012] In some optional implementations of the present embodiment, the coordinate system of the vehicle model is determined based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in the multiple representative trajectories, including: determining the ratio of positioning trajectory points based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in the multiple representative trajectories; determining the coordinate system of the vehicle model based on the ratio of positioning trajectory points.

[0013] In some optional implementations of the present embodiment, the coordinate system of the vehicle model is determined based on the ratio of the positioning trajectory points, including at least one of the following: when the ratio of the positioning trajectory points is greater than or equal to the ratio threshold, the coordinate system of the vehicle model is determined to be the same as the coordinate system of the candidate road set; when the ratio of the positioning trajectory points is less than the ratio threshold, the target trajectory is obtained, and the target trajectory is each trajectory of multiple representative trajectories of the vehicle model, including: obtaining the target trajectory, and the multiple representative trajectories to which the target trajectory belongs are generated based on the following steps: selecting a target coordinate system conversion function from a coordinate system conversion function set, and using the target coordinate system conversion function to update the multiple representative trajectories to multiple representative trajectories that describe their positions in the coordinate system of the candidate road set.

[0014] In a second aspect, the present disclosure provides a vehicle coordinate system calibration device, the device comprising: a trajectory acquisition module, used to acquire a target trajectory, the target trajectory being each of a plurality of representative trajectories of the vehicle model; a probability generation module, used to generate, for a target trajectory point in the target trajectory, an observation probability of mapping the target trajectory point to a target mapping point of a target candidate road, to obtain an observation probability set corresponding to the target trajectory, and to generate a state transition probability between the target mapping point and the next mapping point of the target, to obtain a state transition probability set of the target candidate roads; wherein the target trajectory point is each trajectory point in the target trajectory, and the target candidate road is each candidate road in the candidate road set, The target next mapping point is the mapping point of the target next trajectory point on the target candidate road, and the target next trajectory point is the next trajectory point of the target trajectory point; the path determination module is used to generate the path likelihood probability of the target candidate road according to the observation probability set and the state transition probability set, and determine the target path corresponding to the path likelihood probability from the candidate road set; the coordinate system determination module is used to determine the positioning trajectory point among all trajectory points of the target trajectory, and determine the coordinate system of the vehicle model based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in multiple representative trajectories; wherein the distance between the positioning trajectory point and its mapping point on the target path is less than the distance threshold.

[0015] In some optional implementations of the present embodiment, multiple representative trajectories of vehicle models in the trajectory acquisition module are determined based on the following sub-units: an area determination sub-unit, which is used to determine a candidate trajectory sample area based on at least one of the following: the distribution of vehicle model data of vehicles traveling in the area meets the distribution requirements; the network infrastructure in the area meets the facility requirements; the road network data in the area meets the road network requirements; a vehicle determination sub-unit, which is used to determine a preset number of target vehicles corresponding to the vehicle model from the candidate trajectory sample area; a trajectory selection sub-unit, which is used to select a predetermined number of target driving trajectories for each target vehicle to obtain multiple representative trajectories of the vehicle model; wherein the target driving trajectory meets at least one of the following: includes a preset number of trajectory points; the driving speed of each trajectory point is greater than a preset speed.

[0016] In some optional implementations of the present embodiment, the probability generation module includes: a set selection subunit, which is used to select a set of candidate roads whose distance from the target trajectory point is less than a given radius in the target road network data set; a distance determination subunit, which is used to determine the target distance between the target trajectory point and the target mapping point mapped to the target candidate road; and a probability generation subunit, which is used to generate a probability density function of mapping the target trajectory point to the target mapping point according to the target distance using a Gaussian distribution function as the observation probability of the target trajectory point being mapped to the target mapping point.

[0017] In some optional implementations of the present embodiment, the target trajectory point in the distance determination subunit is mapped to the target mapping point of the target candidate road based on at least one of the following determinations: when the foot of a perpendicular line from the target trajectory point to the target candidate road is located on the target candidate road, the foot of the perpendicular is used as the target mapping point of the target trajectory point on the target candidate road; when the foot of the perpendicular line from the target trajectory point to the target candidate road is located on the extension line of the target candidate road, the endpoint of the target candidate road adjacent to the foot of the perpendicular is used as the target mapping point of the target trajectory point on the target candidate road.

[0018] In some optional implementations of this embodiment, the probability generation module includes: a first distance determination subunit, used to determine the first distance between the target trajectory point and the target next trajectory point, and a second distance determination subunit, used to determine the second distance between the target mapping point and the target next mapping point; an extreme value determination subunit, used to determine the minimum value and the maximum value of the first distance and the second distance respectively; a probability determination subunit, used to use the ratio of the minimum value to the maximum value as the state transition probability between the target mapping point and the target next mapping point.

[0019] In some optional implementations of the present embodiment, the path determination module includes: a product generation subunit, which is used to generate the product of the observation probability of mapping the target trajectory point to the target mapping point of the target candidate road and the state transition probability between the target mapping point and the target next mapping point; a sum calculation subunit, which is used to calculate the sum of the products for all trajectory points in the target trajectory as the path likelihood probability of the target candidate road; and a path selection subunit, which is used to select the candidate road with the largest path likelihood probability from the candidate road set as the target path of the representative trajectory in the candidate road set.

[0020] In some optional implementations of this embodiment, the coordinate system determination module includes: a ratio determination subunit, used to determine the ratio of positioning trajectory points based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in multiple representative trajectories; a coordinate system determination subunit, used to determine the coordinate system of the vehicle model based on the ratio of positioning trajectory points.

[0021] In some optional implementations of the present embodiment, the coordinate system determination subunit is further used for at least one of the following: when the ratio of the positioning trajectory points is greater than or equal to the ratio threshold, determining the coordinate system of the vehicle model to be the same coordinate system as the coordinate system of the candidate road set; when the ratio of the positioning trajectory points is less than the ratio threshold, obtaining the target trajectory, the target trajectory being each trajectory of the multiple representative trajectories of the vehicle model, including: obtaining the target trajectory, the multiple representative trajectories to which the target trajectory belongs are generated based on the following steps: selecting a target coordinate system conversion function from a coordinate system conversion function set, and using the target coordinate system conversion function to update the multiple representative trajectories to multiple representative trajectories that describe their positions in the coordinate system of the candidate road set.

[0022] In a third aspect, the present disclosure provides a computer device, including: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the vehicle model coordinate system calibration method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0023] In a fourth aspect, the present disclosure provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the vehicle model coordinate system calibration method of the first aspect or any corresponding embodiment thereof.

[0024] In a fifth aspect, the present disclosure provides a computer program product, including computer instructions, which are used to enable a computer to execute the vehicle model coordinate system calibration method of the above-mentioned first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the specific embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1 is a flow chart of a vehicle model coordinate system calibration method according to an embodiment of the present disclosure;

[0027] Figure 2 is a flow chart of a method for determining multiple representative trajectories of a vehicle model according to an embodiment of the present disclosure;

[0028] Figure 3 is a flow chart of a method for generating an observation probability of mapping a target trajectory point to a target candidate road according to an embodiment of the present disclosure;

[0029] Figure 4 is a flow chart of a method for generating state transition probabilities of a target mapping point and a target next mapping point according to an embodiment of the present disclosure;

[0030] Figure 5 is a flowchart of a method for determining a target path corresponding to a path likelihood probability according to an embodiment of the present disclosure;

[0031] Figure 6 is a flowchart of a specific application scenario of the vehicle model coordinate system calibration method according to an embodiment of the present disclosure;

[0032] Figure 7is a structural block diagram of a vehicle model coordinate system calibration device according to an embodiment of the present disclosure;

[0033] Figure 8 It is a schematic diagram of the hardware structure of the computer device of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0035] Professional terms related to this disclosure:

[0036] WGS84 coordinate system: It is a universal coordinate system and is widely used in GPS navigation and geographic information systems. WGS84 uses the center of mass of the earth as the origin and uses an ellipsoid model to represent the shape of the earth. Its coordinates are expressed in longitude, latitude and altitude. The WGS84 coordinate system provides precise geographic positioning on a global scale. Therefore, in the Internet of Vehicles, the WGS84 coordinate system is often used for vehicle trajectory recording and positioning on a global scale.

[0037] GCJ02 coordinate system: It is an encrypted coordinate system developed by the National Bureau of Surveying, Mapping and Geoinformation used in China, also known as the Martian coordinate system. GCJ02 encrypts and converts the WGS84 coordinate system. In Internet of Vehicles applications, if the vehicle is traveling in China, the use of the GCJ02 coordinate system can ensure the security and privacy of the trajectory data.

[0038] CGCS2000 coordinate system: It is the Chinese Geodetic Coordinate System 2000, which takes the center of mass of the earth as the origin and uses an ellipsoid model to represent the shape of the earth. It is similar to the WGS84 coordinate system, but has higher accuracy and wider applicability. In Internet of Vehicles applications, the CGCS2000 coordinate system can be used for high-precision vehicle positioning and trajectory recording.

[0039] BD09 coordinate system: It is the coordinate offset standard used by Baidu, also known as the Baidu coordinate system. BD09 is encrypted again on the basis of the GCJ02 coordinate system, which further enhances the security of data. In the Internet of Vehicles, if the vehicle trajectory data needs to be connected with Baidu products such as Baidu Maps, it is necessary to use the BD09 coordinate system.

[0040] Electronic map data, the data content can cover various types of geographic information such as road and railway networks, buildings, water bodies, land use, points of interest, administrative boundaries, etc.

[0041] In the Internet of Vehicles technology, the vehicle trajectory data involved comes from different car companies, and different car companies may use different coordinate systems to record and process trajectory data. For example, the GPS84 coordinate system is a universal geographic coordinate system that is widely used in the global positioning system. The GCJ02 coordinate system is a Chinese-specific encrypted coordinate system used to protect the security and privacy of geographic information. Therefore, different data sources may use different coordinate systems, which leads to the diversity and inconsistency of trajectory data.

[0042] The diversity of trajectory coordinate systems brings challenges to data processing and analysis. If these data are not calibrated to a unified coordinate system, trajectory data from different data sources cannot be directly compared and integrated. This may lead to problems such as information loss, error accumulation, and inaccurate results. Therefore, calibrating the trajectory coordinate system is a key step to ensure data accuracy and consistency.

[0043] When determining the coordinate system used for the longitude and latitude in vehicle trajectory data, the core technical issues include the conversion algorithm between coordinate systems and the trajectory matching algorithm. The specific challenges are as follows:

[0044] (1) Diversity: There are many geographic coordinate systems in the world, such as WGS-84 (the coordinate system used by the Global Positioning System), GCJ-02 (the official coordinate system of China, also known as the Martian coordinate system), BD-09 (Baidu coordinate system), etc. Each coordinate system has its own specific conversion algorithm and offset.

[0045] (2) Confidentiality: Some coordinate systems (such as GCJ-02) are used for national geographic information security reasons, and their conversion algorithms may not be public, which increases the difficulty of identifying the coordinate system.

[0046] (3) Complexity: Conversion between different coordinate systems may involve complex mathematical operations and geographical knowledge, especially when encrypted coordinate systems are involved.

[0047] (4) Accuracy: The accuracy and error of coordinate data will also affect the determination of the coordinate system. If the data accuracy is low or there are errors, the determination result may be inaccurate.

[0048] Based on this, it is necessary to design a calibration scheme for the vehicle trajectory data coordinate system to determine the coordinate system used by the longitude and latitude in the vehicle trajectory data to solve the above problems. In the present disclosure, the vehicle model dimension is selected for the calibration of the coordinate system because it is found through the study of vehicle trajectory data that the positioning coordinate system used by the vehicle-mounted terminal of the same vehicle model is basically the same and generally does not change. Therefore, calibrating the coordinate system of the vehicle model instead of the vehicle dimension can effectively improve the calibration efficiency.

[0049] In the present disclosure, the vehicle model coordinate system calibration method based on map matching is to determine the coordinate system by comparing the deviation between the vehicle's driving trajectory and the electronic map data. This method can quickly and accurately locate the coordinate system of the vehicle's trajectory data, thereby improving the availability and accuracy of the trajectory data.

[0050] Through the vehicle coordinate system calibration method provided by the present disclosure, trajectory data from different data sources can be integrated. This enables researchers to have a more comprehensive understanding of the vehicle's driving path, speed changes and other information, and then conduct more in-depth analysis and mining; at the same time, the unified coordinate system can eliminate errors and deviations caused by inconsistent coordinate systems, which helps to improve the accuracy and reliability of trajectory data analysis.

[0051] According to an embodiment of the present disclosure, an embodiment of a vehicle model coordinate system calibration method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0052] In this embodiment, a vehicle coordinate system calibration method is provided, which can be used in the above-mentioned computer system, such as a desktop computer, an industrial computer, a mobile terminal (such as a tablet computer), etc. Figure 1 is a flow chart of a vehicle model coordinate system calibration method according to an embodiment of the present disclosure. Figure 1 As shown, the process includes the following steps:

[0053] Step S101, obtaining the target trajectory.

[0054] In this embodiment, the target trajectory is each of the multiple representative trajectories of the vehicle type.

[0055] When acquiring the target trajectory, the execution subject generally acquires a set of multiple representative trajectories of multiple vehicle models. Afterwards, for each of the multiple vehicle models, each of the multiple representative trajectories can be used as a target trajectory, and subsequent data processing is performed on the target trajectory.

[0056] Here, in order to improve the data quality of representative trajectories, the vehicle's driving area, vehicle speed and other aspects can usually be used as reference factors to select appropriate representative trajectories.

[0057] Step S102, for the target trajectory point in the target trajectory, generate the observation probability of mapping the target trajectory point to the target mapping point of the target candidate road, obtain the observation probability set corresponding to the target trajectory, and generate the state transition probability between the target mapping point and the target next mapping point, and obtain the state transition probability set of the target candidate road.

[0058] In this embodiment, the target trajectory point is each trajectory point in the target trajectory, the target candidate road is each candidate road in the candidate road set, the target next mapping point is the mapping point of the target next trajectory point on the target candidate road, and the target next trajectory point is the next trajectory point of the target trajectory point.

[0059] The road network set can adopt existing public electronic map data, which is not limited in this application. For example, the road network data corresponding to all provinces and cities in the selected area can be downloaded from the electronic map data provider to obtain the road network set.

[0060] For a target trajectory point in a target trajectory, a road set corresponding to the target trajectory point can be selected from a road network set according to a preset rule to obtain a candidate road set.

[0061] When generating the observation probability of the target trajectory point being mapped to the target mapping point of the target candidate road, for the target trajectory point set P = {p1, p2, ..., p n}, where each pi contains the longitude and latitude coordinates (lat i ,lng i ) and possible timestamps, there may be a candidate road set R = {r1, r2, ..., r n}, where each road r j It consists of a series of road coordinate points that define the shape and location of the road.

[0062] Each road r j As target candidate roads, calculate P = {p1, p2, ..., p n}, as the target mapping point Q = {q1, q2, ..., q n}. Then, based on the target trajectory point p i Target mapping point q to target candidate road i The distance and other possible factors (such as road type, speed limit, direction consistency, etc.) are used to calculate the observation probability p(q|p) of the target trajectory point mapped to the target mapping point of the target candidate road, thereby obtaining the observation probability set of the target mapping point of the entire target trajectory mapped to the target candidate road, that is, the corresponding target observation probability set is obtained.

[0063] State transition probability p(q i →q j ) indicates that at a given target trajectory point p i At the target mapping point q on the target candidate road i , transfer to the next mapping point q j The state transition probability describes the continuity between the target mapping points on the target candidate road.

[0064] The target mapping point may include the following information: the location (latitude and longitude) of the mapping point, the ID of the target candidate road, and in some cases, the target mapping point may also include other attributes (such as speed, direction, etc.).

[0065] The state transition probability can be calculated by at least one of the following: Distance-based transition probability: If two mapping points are on the same road and close to each other, the transition probability between them is high. Speed-based transition probability: If the vehicle's moving speed is within a reasonable range and matches the distance between the two mapping points, the transition probability increases. Direction-based transition probability: If the vehicle's direction remains consistent or changes smoothly between the two mapping points, the transition probability increases.

[0066] For the target trajectory, the state transition probability between the target mapping point where the target trajectory point is mapped to the target candidate road and the next possible mapping point thereof is calculated, and a state transition probability set can be obtained.

[0067] Step S103, generating the path likelihood probability of the target candidate road according to the observation probability set and the state transition probability set, and determining the target path corresponding to the path likelihood probability from the candidate road set.

[0068] In this embodiment, the path likelihood probability of the target candidate road can be the sum of the product of the observation probability and the state transition probability of each mapping point. Afterwards, the candidate road with the largest path likelihood probability can be used as the target path.

[0069] Step S104, determining a positioning trajectory point among all trajectory points of the target trajectory, and determining a coordinate system of the vehicle model based on the number of positioning trajectory points in the multiple representative trajectories and the number of all trajectory points in the multiple representative trajectories.

[0070] In this embodiment, the distance between the positioning trajectory point in the multiple representative trajectories of the vehicle model and its mapping point in the candidate road set is less than the distance threshold. By determining the relationship between the positioning trajectory point in the multiple representative trajectories of the vehicle model and all trajectory points, such as the difference in data volume (difference or ratio, etc.), the degree of correspondence between the positioning trajectory point and the target path can be determined, thereby determining whether the coordinate system of the vehicle model is the same as the coordinate system of the road network set.

[0071] The vehicle model coordinate system calibration method provided in this embodiment can more accurately identify the target path in the road network set corresponding to the target trajectory by generating and comparing the observation probability and state transition probability of the target trajectory in multiple representative trajectories of the vehicle model on different candidate roads, and then determine the coordinate system of the vehicle model based on the proportion of the positioning trajectory points in the multiple representative trajectories of the vehicle model in all trajectory points. This process improves the accuracy of the coordinate system corresponding to the determined vehicle model trajectory, thereby ensuring the accuracy and consistency of the vehicle trajectory data involved in the Internet of Vehicles, which is facilitating the subsequent comparison and integration of trajectory data.

[0072] In some optional implementations of the above embodiments, the above-mentioned determining the coordinate system of the vehicle model based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in multiple representative trajectories may include: determining the ratio of positioning trajectory points based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in multiple representative trajectories; determining the coordinate system of the vehicle model based on the ratio of positioning trajectory points.

[0073] In this implementation, the ratio of the number of positioning trajectory points in multiple representative trajectories to the number of all trajectory points in multiple representative trajectories, that is, the positioning trajectory point ratio, can determine the degree of correspondence between the positioning trajectory points and the candidate roads in the candidate road set, so that according to the degree of correspondence, it can be determined whether the coordinate systems of the multiple representative trajectories of the vehicle model and the candidate road set are the same, thereby determining the coordinate system of the vehicle model. In this process, it is clarified that the coordinate system of the vehicle model is determined by using the positioning trajectory point ratio, so that the efficiency and accuracy of the determined coordinate system of the vehicle model can be improved.

[0074] In some optional implementations of the above embodiments, determining the coordinate system of the vehicle model based on the ratio of positioning trajectory points may include: when the ratio of positioning trajectory points is greater than or equal to a ratio threshold, determining the coordinate system of the vehicle model to be the same as the coordinate system of the candidate road set.

[0075] In this implementation, when the proportion of positioning trajectory points in the representative trajectory is greater than or equal to the proportion threshold, it means that the proportion of positioning trajectory points to all trajectory points is high, and the representative trajectory of the vehicle model has a high correspondence with the candidate road set in the road network set. At this time, it can be determined that the coordinate system of the vehicle model is the same as the coordinate system of the candidate road set. This process clarifies the specific application scenarios for determining the coordinate system of the vehicle model and improves the efficiency and accuracy of determining the coordinate system of the vehicle model.

[0076] Alternatively or additionally, the above-mentioned determination of the coordinate system of the vehicle model based on the ratio of the positioning trajectory points may include: when the ratio of the positioning trajectory points is less than a ratio threshold, obtaining a target trajectory, the target trajectory being each trajectory of a plurality of representative trajectories of the vehicle model, including: obtaining the target trajectory, the plurality of representative trajectories to which the target trajectory belongs are generated based on the following steps: selecting a target coordinate system conversion function from a set of coordinate system conversion functions, and using the target coordinate system conversion function to update the plurality of representative trajectories into a plurality of representative trajectories for position description in the coordinate system of the candidate road set.

[0077] In this implementation, when the proportion of positioning trajectory points in the representative trajectory to all trajectory points is less than the proportion threshold, it means that the correspondence between the representative trajectory and the target path in the candidate road set is low. At this time, there is a high probability that the coordinate system of the vehicle model trajectory data is different from the coordinate system of the road network set.

[0078] Therefore, a target coordinate system conversion function can be selected from the coordinate system conversion function set, and the target coordinate system conversion function can be used to update the multiple representative trajectories into multiple representative trajectories that describe the positions in the coordinate system of the candidate road set, and then the updated multiple representative trajectories can be used to obtain the target trajectory and based on Figure 1 The content of the embodiment shown executes subsequent steps to finally determine the coordinate system of the vehicle model based on the positioning trajectory point ratio.

[0079] In this implementation, when determining the coordinate system of a vehicle model, it is clarified that when the result of determining the coordinate system of the vehicle model indicates that the coordinate system of the vehicle model trajectory data is different from the coordinate system of the road network set, a target coordinate system conversion function can be used to convert multiple representative trajectories of the vehicle model, and then the specific application scenario of the coordinate system of the vehicle model is re-determined, so that the vehicle model trajectory data recorded in different coordinate systems can be calibrated with the coordinate system, thereby improving the efficiency, accuracy and applicability of determining the coordinate system of the vehicle model.

[0080] In some optional implementations of the above embodiments, the above Figure 1 The method for determining multiple representative trajectories of the vehicle model in the embodiment can be used for the above-mentioned execution subject or other related Figure 1 The execution subjects of the methods shown are different from those of the execution subjects. Figure 2 4 is a flowchart of a method for determining multiple representative trajectories of a vehicle model according to an embodiment of the present disclosure.

[0081] like Figure 2 As shown, the process includes the following steps:

[0082] Step S201, determining a candidate trajectory sample area based on at least one of the following: the distribution of vehicle model data traveling in the area meets the distribution requirements; the network infrastructure in the area meets the facility requirements; the road network data in the area meets the road network requirements.

[0083] In this embodiment, the execution subject of the method for determining multiple representative trajectories of a vehicle model can select the trajectory sample area according to a preset selection condition. The selection condition can be determined according to the actual application scenario and the experience of those skilled in the art. For example, one or more of the following selection conditions can be considered:

[0084] ① The distribution breadth of vehicle model data in the area is greater than the breadth threshold;

[0085] ② The network infrastructure in the region is complete and can meet the needs of vehicle network data transmission;

[0086] ③The road network data in the region is complete and can meet the preset road network requirements.

[0087] In some specific examples, based on the above selection conditions, the selected area set (Area Set, abbreviated as AS) can be Beijing, Shanghai, Guangdong, Zhejiang, Jiangsu and Tianjin. The size of the AS set can be set as a parameter. Generally speaking, the larger the selected area, the more car models are covered, but the more computing resources are consumed.

[0088] Step S202: determining a preset number of target vehicles of corresponding vehicle types from the candidate trajectory sample area.

[0089] Step S203: for each target vehicle, a predetermined number of target driving trajectories are selected to obtain a plurality of representative trajectories of the vehicle model.

[0090] In this embodiment, the target driving trajectory meets at least one of the following conditions: it includes a preset number of trajectory points; and the driving speed of each trajectory point is greater than a preset speed.

[0091] Among them, the preset number can be determined according to actual needs or the experience of technical personnel in this field, so as to select a target driving trajectory with an appropriate number of trajectory points, thereby achieving subsequent efficient and accurate coordinate system calibration.

[0092] The preset speed may also be determined according to actual needs or the experience of those skilled in the art, thereby ensuring that each trajectory point of the acquired target driving trajectory has accurate coordinate data.

[0093] In some specific examples, the selection of vehicle trajectory data can be achieved by the following steps: select m (for example, m=10) vehicles for each vehicle model, select n (for example: n=20) driving trajectories for each vehicle, require that each trajectory contains h (for example: h=50) trajectory points, and the driving speed of each trajectory point is greater than 20 km / h. According to these conditions, vehicle trajectory data of different vehicle models can be obtained.

[0094] The method for determining multiple representative trajectories of vehicle models in this embodiment can ensure that the collected vehicle driving trajectory data is highly accurate and representative by screening candidate trajectory sample areas that meet specific conditions (such as vehicle model data distribution, network infrastructure, and road network data requirements). This helps subsequent analysis to be closer to the actual situation and improves the accuracy of research or application. Furthermore, by presetting the number of target vehicles and focusing on the target driving trajectories of these vehicles, the complexity and amount of calculation of data processing are effectively reduced. At the same time, selecting trajectories that contain sufficient trajectory points and whose driving speeds meet certain standards further improves the quality of the data and the efficiency of analysis.

[0095] In some optional implementations of the above embodiments, the above Figure 1 The observation probability of mapping the generated target trajectory point to the target mapping point of the target candidate road can be obtained as follows: Figure 3 The flowchart shown in is implemented. Figure 3 The present invention is a flowchart of a method for generating an observation probability of mapping a target trajectory point to a target candidate road according to an embodiment of the present disclosure.

[0096] like Figure 3 As shown, the process includes the following steps:

[0097] Step S301: In the target road network data set, a candidate road set whose distance to the target trajectory point is less than a given radius is selected.

[0098] In this embodiment, the road segment most likely to be related to the target trajectory point (usually the real-time vehicle location point obtained by GPS or other equipment) is found from the huge road network data. Given a target trajectory point and a preset search radius, the algorithm will traverse the entire road network data set, calculate the distance from each point on the road segment to the target trajectory point, and filter out those road segments whose distance is less than the given radius to form a candidate road set, thereby narrowing the scope of subsequent processing and improving the efficiency of the algorithm.

[0099] Step S302, determining a target distance between a target trajectory point and a target mapping point where the target trajectory point is mapped to a target candidate road.

[0100] In this embodiment, after determining the set of candidate roads, the next step is to find the point on each candidate road that best matches the target trajectory point, that is, the target mapping point. This is usually achieved by calculating the distance from the target trajectory point to each point on the candidate road, and selecting the point with the smallest distance as the target mapping point. The target distance is the distance from the target trajectory point to its corresponding target mapping point, which can be Euclidean distance, Manhattan distance, Chebyshev distance, Minkowski distance, standardized Euclidean distance, etc. Through this step, the degree of proximity between the target trajectory point and the candidate road can be quantified, providing a basis for subsequent probability calculations.

[0101] Step S303, using the Gaussian distribution function, according to the target distance, a probability density function of mapping the target trajectory point to the target mapping point is generated as the observation probability of mapping the target trajectory point to the target mapping point.

[0102] In this embodiment, the algorithm uses a Gaussian distribution (also called a normal distribution) function to simulate the probability of mapping a target trajectory point to a target mapping point on a candidate road. The Gaussian distribution function is a continuous probability distribution whose shape is determined by the mean (μ) and standard deviation (σ). Here, the target distance can be used as a parameter of the Gaussian distribution (usually a function of the standard deviation σ), and the parameters of the Gaussian distribution can be adjusted to reflect the change in mapping probability at different target distances. The smaller the target distance, the higher the probability of mapping to the point; otherwise, the lower. The advantage of this is that the distance information in the geographic space can be converted into probability information, which facilitates subsequent decision-making and reasoning processes.

[0103] In some specific examples, the observation probability of mapping a target trajectory point to a target mapping point can be determined based on the following steps:

[0104] (1) A trajectory (also known as the target trajectory) s is selected from multiple representative trajectories (TRS) of the vehicle model. For each trajectory point p in the trajectory, a candidate road set (RS) is selected from the road network set G with a given radius R (e.g., R = 100 m).

[0105] (2) Calculate the perpendicular line from the target trajectory point p to each road r in RS (i.e., the target candidate road), and the foot of the perpendicular q is the mapping point of point p on road r. Calculate the target distance D, which can be defined as the Euclidean distance between points p and q.

[0106]

[0107] Among them: lng p is the longitude of point p, lat p is the latitude of point p.

[0108] (3) Calculate the observation probability: Use Gaussian distribution to measure. The farther the target trajectory point p is from the target mapping point q, the lower the observation probability. The larger the noise of the parameter measurement, the lower the credibility of the trajectory point position. Here, a robust estimate can be calculated based on real data. It is defined as:

[0109]

[0110] σ ρ =1.48*median(D), where median(D) is the median of the distances from the target trajectory point p to all candidate target mapping points q. The meaning is that the greater the noise, the lower the credibility of the trajectory point; D is the distance between point p and point q expressed in the above formula. The larger the distance D, the lower the observation probability according to the formula. e is a natural constant, and its value is approximately 2.718281828459045.

[0111] The method of generating the observation probability of mapping the target trajectory point to the target candidate road in the present implementation can more accurately determine on which road the target trajectory point is most likely to be located, thereby improving the accuracy of the matching result, by considering the distance between the target trajectory point and the candidate road and using the Gaussian distribution function to quantify the mapping probability.

[0112] In the above Figure 3 In some optional implementations of the process shown, the target trajectory point is mapped to the target mapping point of the target candidate road, which can be determined based on the following steps:

[0113] In some examples, when the foot of a perpendicular line from a target trajectory point to a target candidate road is located on the target candidate road, the foot of the perpendicular line may be used as a target mapping point of the target trajectory point on the target candidate road.

[0114] Alternatively or additionally, when the foot of a perpendicular line from the target trajectory point to the target candidate road is located on an extension line of the target candidate road, an end point of the target candidate road adjacent to the foot of the perpendicular line may be used as a target mapping point of the target trajectory point on the target candidate road.

[0115] In some optional implementations of the above embodiments, the above Figure 1 The state transition probability between the generated target mapping point and the next target mapping point in can be expressed as follows: Figure 4 The flowchart shown in is implemented. Figure 4 The present invention is a flowchart of a method for generating state transition probabilities between a target mapping point and a target next mapping point according to an embodiment of the present disclosure.

[0116] like Figure 4 As shown, the process includes the following steps:

[0117] Step S401, determining a first distance between a target track point and a next target track point.

[0118] In this embodiment, the target trajectory point is the point of current interest on the target trajectory, which may be the current position of a vehicle, robot, or other moving object. The next target trajectory point is the point immediately following the target trajectory point on the target trajectory, indicating the position that the object is about to reach. The first distance may be the distance between the two points. The distance may be calculated using a method of calculating distance in the prior art or in a technology to be developed in the future, and this application does not limit this. The first distance may reflect the distance that the vehicle actually moves in physical space.

[0119] Step S402: determining a second distance between the target mapping point and the next target mapping point.

[0120] In this embodiment, the target mapping point may be a corresponding point in the target candidate road mapped from the target trajectory point. The target next mapping point, corresponding to the target mapping point, is a corresponding point in the target candidate road mapped from the target next trajectory point. The second distance may be a distance calculated between the two mapping points. The distance may be calculated in the same manner as the first distance.

[0121] Step S403: determine the minimum value and the maximum value of the first distance and the second distance respectively.

[0122] Step S404: taking the ratio of the minimum value to the maximum value as the state transition probability between the target mapping point and the target next mapping point.

[0123] In this implementation, the state transition probability from the current mapping point to the next mapping point is estimated by comparing the ratio of the minimum value to the maximum value. Compared with the usual case where the closer the first distance is to the second distance, the greater the state transition probability is, the state transition probability is determined by using the ratio of the minimum value to the maximum value, which enlarges the difference between different state transition probabilities, that is, the difference between the spacing between the target trajectory points and the spacing between the target mapping points is enlarged, thereby increasing the distinction between the state transition probabilities between different mapping points.

[0124] In a specific example, the state transition probability between the target mapping point and the target next mapping point can be generated by the following steps:

[0125] The state transition probability can be measured by the difference between the distance between the target trajectory points and the distance between the target mapping points. The calculation formula of the state transition probability, the numerator can be the minimum value of the distance between the target mapping points and the distance between the target trajectory points, and the denominator can be the maximum value of the distance between the target mapping points and the distance between the target trajectory points:

[0126]

[0127] Where: p(q i →q j ) means the target mapping point q i To the next mapping point q j The state transition probability between i →q j ) means the target mapping point q i To the next mapping point q j The distance is taken as the length of the shortest path on the road network, which can be calculated using the classic shortest path calculation algorithm Dijkstra algorithm; dis(pi →p j ) means the target trajectory point p i To the next trajectory point p j The distance is calculated as follows:

[0128] dis(q i →q j )=dijsktra(q i ,q j )

[0129] dis(p i →p j )=|mileage(p i )-mileage(p j )|

[0130] mileage(p i ) can be the target trajectory point p i The mileage reported by the vehicle. The mileage is the additional uploaded information in the track information and is directly provided by the vehicle system. If the vehicle cannot provide the mileage, dis(p i →p j ) can be replaced by p i to p j The Euclidean distance.

[0131] The method for generating the state transition probability between the target mapping point and the next target mapping point in this implementation determines the state transition probability by the ratio of the minimum value and the maximum value in the first distance between the target trajectory points and the second distance between the target mapping points, thereby improving the distinction of the state transition probabilities between different mapping points.

[0132] In some optional implementations of the above embodiments, the above Figure 1 According to the observation probability set and the state transition probability set, the path likelihood probability of the target candidate road is generated. From the candidate road set, the target path corresponding to the path likelihood probability can be determined. Figure 5 The flowchart shown in is implemented. Figure 5 The present invention is a flowchart of a method for determining a target path corresponding to a path likelihood probability according to an embodiment of the present disclosure.

[0133] like Figure 5 As shown, the process includes the following steps:

[0134] Step S501 : Generate the product of the observation probability of mapping the target trajectory point to the target mapping point of the target candidate road and the state transition probability between the target mapping point and the target next mapping point.

[0135] In this implementation, for each point in the target trajectory, the observation probability of it being mapped to the corresponding point (target mapping point) on the target candidate road can be calculated. The observation probability reflects the possibility of the trajectory point appearing at that position on the target candidate road.

[0136] After that, the state transition probability between the target mapping point and the target next mapping point can be calculated. The state transition probability reflects the possibility of moving from the current mapping point to the next mapping point.

[0137] Afterwards, the observation probability and the state transition probability can be multiplied to obtain the probability contribution of each trajectory point in the target trajectory mapped to the candidate road.

[0138] Step S502, for all trajectory points in the target trajectory, the sum of the products is calculated as the path likelihood probability of the target candidate road.

[0139] In this implementation, the probability contributions of all trajectory points are added together to obtain the path likelihood probability of the candidate road. The path likelihood probability reflects the matching degree between the entire target trajectory and the target candidate road.

[0140] Step S503 : selecting a candidate road with the largest path likelihood probability from the candidate road set as the target path of the representative trajectory in the candidate road set.

[0141] In this implementation, from the candidate road set including all candidate roads, the candidate road with the largest path likelihood probability is selected as the target path, indicating that the target path has the highest matching degree with the target trajectory and is most likely to be the true path of the target trajectory.

[0142] In a specific example, the target path can be determined by the following steps:

[0143] For each possible candidate path r: r1->r2->....->r n , the path likelihood probability can be calculated based on the observation probability and state transition probability: The maximum value is selected from the most likely path sequence, and the most likely path sequence is used as the target path. The dynamic programming algorithm can be used to accelerate the calculation of the maximum likelihood probability among all paths.

[0144] The method for determining the target path in this implementation adopts the sum of the product of the observation probability and the state transition probability of each trajectory point as the path likelihood probability, and then takes the candidate road with the largest path likelihood probability as the target path. In this process, the matching degree of the entire trajectory and the candidate road is referred to, and the order and relationship between the trajectory points are also considered, thereby improving the accuracy of the determined target path.

[0145] Figure 6It is a flowchart of a specific application scenario of the vehicle model coordinate system calibration method according to an embodiment of the present disclosure.

[0146] like Figure 6 As shown, Figure 6 The coordinate system of the vehicle model can be determined by the vehicle model coordinate system calibration method disclosed in the present invention. If successful, a vehicle model coordinate system calibration result table is obtained; if unsuccessful, the coordinate system is converted and the coordinate system of the vehicle model is determined again. The data for determining the coordinate system of the vehicle model comes from the representative trajectory of the vehicle model obtained by selecting the trajectory and the candidate road set in the road network data set.

[0147] In this embodiment, a vehicle coordinate system calibration device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0148] This embodiment provides a vehicle model coordinate system calibration device, such as Figure 7 As shown, the present disclosure provides a vehicle coordinate system calibration device, the device comprising: a trajectory acquisition module 701, used to acquire a target trajectory, the target trajectory being each of a plurality of representative trajectories of the vehicle model; a probability generation module 702, used to generate, for a target trajectory point in the target trajectory, an observation probability of mapping the target trajectory point to a target mapping point of a target candidate road, to obtain an observation probability set corresponding to the target trajectory, and to generate a state transition probability between the target mapping point and the next mapping point of the target, to obtain a state transition probability set of the target candidate roads; wherein the target trajectory point is each trajectory point in the target trajectory, the target candidate road is each candidate road in the candidate road set, and the target The next mapping point is the mapping point of the target next trajectory point on the target candidate road, and the target next trajectory point is the next trajectory point of the target trajectory point; the path determination module 703 is used to generate the path likelihood probability of the target candidate road according to the observation probability set and the state transition probability set, and determine the target path corresponding to the path likelihood probability from the candidate road set; the coordinate system determination module 704 is used to determine the positioning trajectory point among all trajectory points of the target trajectory, and determine the coordinate system of the vehicle type based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in multiple representative trajectories; wherein the distance between the positioning trajectory point and its mapping point on the target path is less than the distance threshold.

[0149] In some optional implementations of the present embodiment, the multiple representative trajectories of the vehicle model in the trajectory acquisition module 701 are determined based on the following sub-units: an area determination sub-unit, which is used to determine the candidate trajectory sample area based on at least one of the following: the distribution of vehicle model data of vehicles traveling in the area meets the distribution requirements; the network infrastructure in the area meets the facility requirements; the road network data in the area meets the road network requirements; a vehicle determination sub-unit, which is used to determine a preset number of target vehicles corresponding to the vehicle model from the candidate trajectory sample area; a trajectory selection sub-unit, which is used to select a predetermined number of target driving trajectories for each target vehicle to obtain multiple representative trajectories of the vehicle model; wherein the target driving trajectory meets at least one of the following: includes a preset number of trajectory points; the driving speed of each trajectory point is greater than the preset speed.

[0150] In some optional implementations of the present embodiment, the probability generation module 702 includes: a set selection subunit, which is used to select a set of candidate roads whose distance from the target trajectory point is less than a given radius in the target road network data set; a distance determination subunit, which is used to determine the target distance between the target trajectory point and the target mapping point mapped to the target candidate road; and a probability generation subunit, which is used to generate a probability density function of mapping the target trajectory point to the target mapping point according to the target distance using a Gaussian distribution function as the observation probability of the target trajectory point being mapped to the target mapping point.

[0151] In some optional implementations of the present embodiment, the target trajectory point in the distance determination subunit is mapped to the target mapping point of the target candidate road based on at least one of the following determinations: when the foot of a perpendicular line from the target trajectory point to the target candidate road is located on the target candidate road, the foot of the perpendicular is used as the target mapping point of the target trajectory point on the target candidate road; when the foot of the perpendicular line from the target trajectory point to the target candidate road is located on the extension line of the target candidate road, the endpoint of the target candidate road adjacent to the foot of the perpendicular is used as the target mapping point of the target trajectory point on the target candidate road.

[0152] In some optional implementations of this embodiment, the probability generation module 702 includes: a first distance determination subunit, used to determine the first distance between the target trajectory point and the target next trajectory point, and a second distance determination subunit, used to determine the second distance between the target mapping point and the target next mapping point; an extreme value determination subunit, used to determine the minimum value and the maximum value of the first distance and the second distance respectively; a probability determination subunit, used to use the ratio of the minimum value to the maximum value as the state transition probability between the target mapping point and the target next mapping point.

[0153] In some optional implementations of the present embodiment, the path determination module 703 includes: a product generation subunit, which is used to generate the product of the observation probability of mapping the target trajectory point to the target mapping point of the target candidate road and the state transition probability between the target mapping point and the target next mapping point; a sum calculation subunit, which is used to calculate the sum of the products for all trajectory points in the target trajectory as the path likelihood probability of the target candidate road; and a path selection subunit, which is used to select the candidate road with the largest path likelihood probability from the candidate road set as the target path of the representative trajectory in the candidate road set.

[0154] In some optional implementations of this embodiment, the coordinate system determination module 704 includes: a ratio determination subunit, which is used to determine the ratio of positioning trajectory points based on the number of positioning trajectory points in multiple representative trajectories and the number of all trajectory points in multiple representative trajectories; a coordinate system determination subunit, which is used to determine the coordinate system of the vehicle model based on the ratio of positioning trajectory points.

[0155] In some optional implementations of the present embodiment, the coordinate system determination subunit is further used for at least one of the following: when the ratio of the positioning trajectory points is greater than or equal to the ratio threshold, determining the coordinate system of the vehicle model to be the same coordinate system as the coordinate system of the candidate road set; when the ratio of the positioning trajectory points is less than the ratio threshold, obtaining the target trajectory, the target trajectory being each trajectory of the multiple representative trajectories of the vehicle model, including: obtaining the target trajectory, the multiple representative trajectories to which the target trajectory belongs are generated based on the following steps: selecting a target coordinate system conversion function from a coordinate system conversion function set, and using the target coordinate system conversion function to update the multiple representative trajectories to multiple representative trajectories that describe their positions in the coordinate system of the candidate road set.

[0156] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0157] The vehicle coordinate system calibration device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0158] See also Figure 8 , Figure 81 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present disclosure, and the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0159] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0160] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0161] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0162] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0163] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means.

[0164] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0165] The embodiments of the present disclosure also provide a computer-readable storage medium. The above-mentioned method according to the embodiments of the present disclosure can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium and downloaded through a network, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0166] A part of the present disclosure may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present disclosure through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0167] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A vehicle model coordinate system calibration method, characterized in that: The method comprises: Obtaining a target trajectory, wherein the target trajectory is each trajectory of a plurality of representative trajectories of the vehicle model; For a target trajectory point in a target trajectory, generating an observation probability of mapping the target trajectory point to a target mapping point of a target candidate road, obtaining an observation probability set corresponding to the target trajectory, and generating a state transition probability between the target mapping point and a target next mapping point, obtaining a state transition probability set of the target candidate road; The target trajectory point is each trajectory point in the target trajectory, the target candidate road is each candidate road in the candidate road set, the target next mapping point is the mapping point of the target next trajectory point on the target candidate road, and the target next trajectory point is the next trajectory point of the target trajectory point; Generate a path likelihood probability of the target candidate road according to the observation probability set and the state transition probability set, and determine a target path corresponding to the path likelihood probability from the candidate road set; Among all the trajectory points of the target trajectory, a positioning trajectory point is determined, and based on the number of the positioning trajectory points in the multiple representative trajectories and the number of all trajectory points in the multiple representative trajectories, a coordinate system of the vehicle model is determined; wherein the distance between the positioning trajectory point and its mapping point on the target path is less than a distance threshold.

2. The method according to claim 1, characterized in that: Multiple representative trajectories of the vehicle model are determined based on the following steps: The candidate trajectory sample area is determined based on at least one of the following: the distribution of vehicle model data of vehicles traveling in the area meets the distribution requirements; the network infrastructure in the area meets the facility requirements; the road network data in the area meets the road network requirements; Determining a preset number of target vehicles corresponding to the vehicle type from the candidate trajectory sample area; For each of the target vehicles, a predetermined number of target driving trajectories are selected to obtain multiple representative trajectories of the vehicle model; wherein the target driving trajectory meets at least one of the following conditions: including a preset number of trajectory points; and the driving speed of each trajectory point is greater than a preset speed.

3. The method according to claim 1, characterized in that: The generating of the observation probability of the target trajectory point being mapped to the target candidate road target mapping point includes: In the target road network data set, a candidate road set whose distance from the target trajectory point is less than a given radius is selected; Determine a target distance between the target trajectory point and a target mapping point where the target trajectory point is mapped to a target candidate road; A Gaussian distribution function is used to generate a probability density function of mapping the target trajectory point to the target mapping point according to the target distance as an observation probability of mapping the target trajectory point to the target mapping point.

4. The method according to claim 3, characterized in that: The target trajectory point is mapped to the target mapping point of the target candidate road, and is determined based on at least one of the following: When the foot of a perpendicular line from the target trajectory point to the target candidate road is located on the target candidate road, the foot of the perpendicular line is used as a target mapping point of the target trajectory point on the target candidate road; When the foot of a perpendicular line from the target trajectory point to the target candidate road is located on the extension line of the target candidate road, an end point of the target candidate road adjacent to the foot of the perpendicular line is used as a target mapping point of the target trajectory point on the target candidate road.

5. The method according to claim 3, characterized in that: The generating the state transition probability between the target mapping point and the target next mapping point includes: determining a first distance between the target trajectory point and the target next trajectory point, and Determining a second distance between the target mapping point and the target next mapping point; Determine a minimum value and a maximum value of the first distance and the second distance respectively; The ratio of the minimum value to the maximum value is used as the state transition probability between the target mapping point and the target next mapping point.

6. The method according to claim 1, characterized in that: The step of generating the path likelihood probability of the target candidate road according to the observation probability set and the state transition probability set, and determining the target path corresponding to the path likelihood probability from the candidate road set comprises: Generate an observation probability product of mapping the target trajectory point to a target mapping point of a target candidate road and a state transition probability product between the target mapping point and a target next mapping point; For all trajectory points in the target trajectory, the sum of the products is calculated as the path likelihood probability of the target candidate road; From the candidate road set, a candidate road with the largest path likelihood probability is selected as the target path of the representative trajectory in the candidate road set.

7. The method according to claim 1, characterized in that: The determining the coordinate system of the vehicle type based on the number of the positioning trajectory points in the plurality of representative trajectories and the number of all trajectory points in the plurality of representative trajectories includes: Determining a positioning trajectory point ratio based on the number of the positioning trajectory points in the plurality of representative trajectories and the number of all trajectory points in the plurality of representative trajectories; Based on the ratio of the positioning trajectory points, a coordinate system of the vehicle model is determined.

8. The method according to claim 7, characterized in that The determining of the coordinate system of the vehicle model based on the ratio of the positioning trajectory points includes at least one of the following: When the ratio of the positioning trajectory points is greater than or equal to a ratio threshold, determining that the coordinate system of the vehicle type is the same as the coordinate system of the candidate road set; When the ratio of the positioning trajectory points is less than a ratio threshold, the target trajectory is obtained, and the target trajectory is each trajectory of a plurality of representative trajectories of the vehicle model, including: obtaining the target trajectory, and the plurality of representative trajectories to which the target trajectory belongs are generated based on the following steps: selecting a target coordinate system conversion function from a set of coordinate system conversion functions, and using the target coordinate system conversion function to update the plurality of representative trajectories into a plurality of representative trajectories that describe positions in the coordinate system of the candidate road set.

9. A vehicle type coordinate system calibration device, characterized in that: The device comprises: A trajectory acquisition module, used to acquire a target trajectory, wherein the target trajectory is each trajectory of a plurality of representative trajectories of the vehicle model; A probability generation module is used to generate, for a target trajectory point in a target trajectory, an observation probability of mapping the target trajectory point to a target mapping point of a target candidate road, to obtain an observation probability set corresponding to the target trajectory, and to generate a state transition probability between the target mapping point and a target next mapping point, to obtain a state transition probability set of the target candidate road; The target trajectory point is each trajectory point in the target trajectory, the target candidate road is each candidate road in the candidate road set, the target next mapping point is the mapping point of the target next trajectory point on the target candidate road, and the target next trajectory point is the next trajectory point of the target trajectory point; A path determination module, used to generate a path likelihood probability of the target candidate road according to the observation probability set and the state transition probability set, and determine a target path corresponding to the path likelihood probability from the candidate road set; A coordinate system determination module is used to determine a positioning trajectory point among all trajectory points of the target trajectory, and determine the coordinate system of the vehicle model based on the number of the positioning trajectory points in the multiple representative trajectories and the number of all trajectory points in the multiple representative trajectories; wherein the distance between the positioning trajectory point and its mapping point on the target path is less than a distance threshold.

10. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the vehicle model coordinate system calibration method according to any one of claims 1 to 8 by executing the computer instructions.