Vehicle positioning method, system and device and storage medium

By combining a probability model of multiple positioning information in the navigation map, the problems of lane-level positioning accuracy and stability in complex road scenarios are solved, and stable and accurate lane-level positioning based on navigation maps are achieved.

CN118936487APending Publication Date: 2024-11-12ZHEJIANG SMART INTELLIGENCE TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410998064.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In the complex road scenarios, the lane-level positioning accuracy and stability based on satellite navigation maps are difficult to ensure, especially in situations such as signal occlusion or reflection.

Method used

By combining static reference objects, dynamic obstacles and historical motion information in the navigation map, a variety of positioning sources are fused, and maximum likelihood estimation is used to achieve stable and accurate lane-level lateral positioning.

Benefits of technology

The stability and accuracy of lane-level positioning are achieved in the navigation map, avoiding dependence on high-precision maps and external high-precision positioning sources, and improving vehicle positioning accuracy in complex road scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118936487A_ABST
    Figure CN118936487A_ABST
Patent Text Reader

Abstract

The invention provides a vehicle positioning method, system and device and a storage medium, and relates to the technical field of vehicle intelligent driving. The vehicle positioning method comprises the following steps: acquiring a vehicle position, and correspondingly loading a navigation map; generating initial road information in a preset positioning range of the navigation map; based on the reference object position information and the initial road information, determining first probability distribution of the vehicle in a navigation map; determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information; based on the historical motion information and the initial road information of the vehicle, determining third probability distribution of the vehicle in the navigation map; fusing the first probability distribution, the second probability distribution and the third probability distribution to obtain fusion probability distribution; and obtaining and updating the location of the vehicle in the navigation map based on the fusion probability distribution. According to the vehicle positioning method, information such as reference objects on a lane, dynamic obstacle observation and dead reckoning is fused, and lane-level positioning independent of a high-precision map is obtained.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of vehicle intelligent driving technology, and in particular to a vehicle positioning method, system, device and storage medium. Background Art

[0002] High-precision positioning usually relies on high-precision maps to provide vehicles with high-precision position information on high-precision maps. However, due to the high production cost and long update cycle of high-precision maps, it is difficult to mass-produce them on a large scale. The industry generally abandons or reduces the use of high-precision maps and adopts the so-called "light map" technology route. "Light map" means using SD map (navigation map) to provide only low-precision map navigation information. Correspondingly, the reliance on high-precision positioning is also reduced. However, it is still necessary to determine the lane-level positioning information of the vehicle in the SD map.

[0003] Compared with road-level positioning, lane-level positioning improves the positioning accuracy of vehicles within the lane. It is a necessary module for applications such as lane-level navigation, ADAS, and L2 / L2+ intelligent driving. At present, lane-level positioning generally relies on high-precision GNSS positioning. However, in scenes such as tunnels and elevated roads, GNSS positioning accuracy cannot be guaranteed due to signal obstruction, reflection, etc., and will degenerate into road-level positioning.

[0004] Therefore, it is necessary to provide a vehicle positioning method, system, device and storage medium that can achieve high-precision and stable lane-level positioning in the navigation map, thereby improving the above-mentioned problems. Summary of the invention

[0005] The object of the present invention is to provide a vehicle positioning method, system, device and storage medium to improve the technical problem of reduced positioning accuracy and stability of existing lane-level positioning based on satellite navigation maps in complex road scenes.

[0006] To achieve the above-mentioned object and other related objects, in a first aspect, the present invention provides a vehicle positioning method, which comprises the following steps:

[0007] Obtaining the vehicle position, and loading a navigation map corresponding to the vehicle position;

[0008] Based on the navigation map, extracting initial road information within a preset positioning range around the vehicle position;

[0009] Extracting the measured reference object position information within the preset positioning range, and determining a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information;

[0010] Extracting the measured dynamic obstacle position information within the preset positioning range, and determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information;

[0011] Determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information;

[0012] Merging the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution;

[0013] The location of the vehicle in the navigation map is obtained and updated based on the fused probability distribution.

[0014] In a second aspect, the present invention provides a vehicle positioning system, the vehicle positioning system comprising:

[0015] A preliminary positioning unit, used to obtain the vehicle position and load a navigation map corresponding to the vehicle position;

[0016] A road initialization unit, which extracts initial road information within a preset positioning range around the vehicle position based on the navigation map;

[0017] a first prediction unit, which extracts the measured reference object position information within the preset positioning range, and determines a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information;

[0018] a second prediction unit, extracting the measured dynamic obstacle position information within the preset positioning range, and determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information;

[0019] a third prediction unit, determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information;

[0020] A probability fusion unit, which fuses the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution;

[0021] A positioning update unit is used to obtain and update the positioning information of the vehicle in the navigation map based on the fused probability distribution.

[0022] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned vehicle positioning method when executing the computer program.

[0023] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned vehicle positioning method are implemented.

[0024] The above-mentioned vehicle positioning method, system, device and storage medium, based on the navigation map, converts the observation of static reference objects, dynamic obstacle observation and self-vehicle dead reckoning on the lane into multiple probability distributions based on a unified probability model and fuses them to obtain a fused probability distribution, and performs maximum likelihood estimation on the fused probability distribution to provide stable and accurate lane-level lateral positioning in the navigation map. The vehicle positioning method does not rely on high-precision maps or external high-precision positioning sources such as RTK. By relying on observations from multiple positioning sources, it can stably achieve lane-level positioning based on the navigation map. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The features and advantages of the present invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the present invention in any way. In the accompanying drawings:

[0026] Figure 1 Shown is a schematic flow chart of the steps of a vehicle positioning method according to an embodiment of the present invention;

[0027] Figure 2 Shown is a schematic diagram of the application architecture of a vehicle positioning system according to an embodiment of the present invention;

[0028] Figure 3 It is a schematic diagram of the process of step S2 in one embodiment of the present invention;

[0029] Figure 4 Shown is a schematic diagram of establishing a road coordinate system in a navigation map in one embodiment of the present invention;

[0030] Figure 5 Shown is a schematic diagram of extracting initial road information from a navigation map in one embodiment of the present invention;

[0031] Figure 6 It is a schematic diagram of the process of step S3 in one embodiment of the present invention;

[0032] Figure 7 Shown is a schematic diagram of converting reference object observation in a navigation map into reference object position information in one embodiment of the present invention;

[0033] Figure 8 It is a schematic diagram of the process of step S4 in one embodiment of the present invention;

[0034] Fig. 9 Shown is a schematic diagram of converting dynamic obstacle observations in a navigation map into dynamic obstacle position information in one embodiment of the present invention;

[0035] Fig.10 It is a schematic diagram of the process of step S5 in one embodiment of the present invention;

[0036] Fig.11 Shown is a schematic diagram of converting vehicle dead reckoning into vehicle dead position trajectory information in a navigation map in one embodiment of the present invention;

[0037] Fig.12 Shown is a structural block diagram of a vehicle positioning system in one embodiment of the present invention;

[0038] Fig.13 Shown is a schematic diagram of the structure of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] See also Figures 1 to 13 It should be noted that the illustrations provided in this embodiment are only used to illustrate the basic concept of the present invention in a schematic manner, and the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0041] See also Figure 1 and Figure 2 The present invention provides a vehicle positioning method, system, device and storage medium. The vehicle positioning method converts the observation of static reference objects, dynamic obstacle observation and self-vehicle dead reckoning on the lane into multiple probability distributions based on a unified probability model on the basis of a navigation map and fuses them to obtain a fused probability distribution, and performs maximum likelihood estimation on the fused probability distribution to provide stable and accurate lane-level lateral positioning in the navigation map.

[0042] See also Figure 2, a vehicle positioning method and system provided by the present invention are applied in a vehicle-mounted controller. The vehicle-mounted controller interacts with the Global Navigation Satellite System (GNSS) through a built-in navigation and positioning module, and uses satellite positioning technology to determine the vehicle position; the vehicle-mounted controller loads a navigation map within a certain range based on the vehicle position and centered on the vehicle position. The vehicle-mounted controller constructs an initial probability model based on the navigation map, and substitutes the reference objects and dynamic obstacle positions measured by the detection equipment around the vehicle body (such as cameras, millimeter-wave radars, lidars, etc.) and the dead position trajectory calculated by the vehicle into the initial probability model for matching, so that the above three types of positioning information are converted into multiple probability distributions. Finally, the vehicle-mounted controller uses a built-in fusion module to obtain fused probabilities from multiple probability distributions, and performs maximum likelihood estimation on the fused probability distribution to provide stable and accurate lane-level lateral positioning in the navigation map.

[0043] See also Figure 1 and Figure 2 The above vehicle positioning method comprises the following steps:

[0044] S1. Obtain a vehicle position, and load a navigation map corresponding to the vehicle position.

[0045] In step S1, the vehicle controller interacts with the GNSS to obtain the vehicle position through satellite positioning. The vehicle position includes the longitude, latitude and altitude of the vehicle on the earth and the corresponding confidence level (i.e., the standard deviation of the corresponding longitude, latitude and altitude), such as precise single point positioning (PGNSS) including longitude, latitude and altitude; and then the vehicle controller loads a navigation map around the vehicle position, specifically loading a navigation map with a preset radius (e.g., 100 meters) centered on the vehicle position.

[0046] It should be noted that GNSS positioning is based on satellite positioning technology and is divided into single-point positioning, differential GPS positioning, and RTK (Real-time kinematic) GPS positioning. Single-point positioning provides 3-10 meters of positioning accuracy, differential GPS provides 0.5-2 meters of positioning accuracy, and RTK GPS provides centimeter-level positioning accuracy. The limitation of relying solely on GNSS navigation positioning is that it relies on positioning facilities and is affected by signal obstruction and reflection, and fails in scenes such as tunnels and elevated roads.

[0047] Next, step S2 is executed to extract initial road information within a preset positioning range around the vehicle position based on the navigation map.

[0048] Specifically, Figure 3 As shown, in some embodiments, step S2 includes:

[0049] S21. In the navigation map, establish a road coordinate system around the vehicle position;

[0050] In step S21, taking the vehicle position as the origin, convert the navigation map loaded in step S1 from the world coordinate system to the local coordinate system, and then convert the navigation map in the local coordinate system to the road coordinate system; for example, taking the vehicle position as the origin, convert the navigation map from the Earth-Centered, Earth-Fixed (ECEF) coordinate system to the East-North-Up (ENU) coordinate system, and convert the navigation map in the ENU coordinate system to the road coordinate system. As shown, in the road coordinate system, the vehicle position is the origin, the y-axis points to the road forward direction, and the x-axis points to the right side of the road. Figure 4 As shown, in the road coordinate system, the vehicle position is the origin, the y-axis points to the road forward direction, and the x-axis points to the right side of the road.

[0051] S22. In the road coordinate system, perform discrete sampling within a preset positioning range according to a preset rule to generate initial road information.

[0052] As Figure 4 and Figure 5 shown, in step S22, in the road coordinate system, within a preset positioning range (such as -a < x < a, -b < y < b) around the origin, perform discrete sampling at a preset discrete sampling rate (sx, sy) according to a preset rule, and use the discrete sampling result as the obtained initial road information. Specifically, divide the preset positioning range into n×m sampling areas, each sampling area has a size of (sx, sy), sample each sampling area within the preset positioning range one by one according to a preset rule. If the sampling area is within the lane range, set the value of this sampling area to 1, otherwise set the value of this sampling area to 0; traverse all sampling areas one by one to obtain the initial road information. As Figure 5 shown, the sampling result is represented by gray filling as 1 and white filling as 0 in the figure. Construct an initial road matrix R with n rows and m columns according to the arrangement of the sampling areas as the initial road information. Among them, the preset positioning range should be greater than the detection range of the on-vehicle detection device and the road width. Specifically, a and b are adjustable parameters. a is determined according to the observation range of the positioning source and the road width, and needs to be greater than half of the positioning source observation range and the road width. b is取值 according to the observation range of the positioning source and needs to be greater than the positioning source observation range; the discrete sampling rate (sx, sy) can be adjusted according to requirements. The smaller the sampling rate, the higher the positioning accuracy. In one example, considering the balance between efficiency and accuracy, the discrete sampling rate is taken as (0.2m, 0.2m).

[0053] Next, perform step S3, extract the measured reference object position information within the preset positioning range, and determine the first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information.

[0054] In step S3, within the preset positioning range, match the position of the reference object measured by the detection device with the navigation map to rely on the characteristics of the surrounding reference objects for positioning and obtain the first probability distribution of the vehicle in the navigation map. Specifically, as Figure 6 shown, in some embodiments, step S3 includes:

[0055] S31. Obtain the position of the reference object measured around the vehicle, where the reference object position is the position of the reference object relative to the vehicle;

[0056] As Figure 7 shown, the vehicle is equipped with detection devices such as cameras, lidar, and millimeter-wave radars. Through the above detection devices, obtain the road observation information around the vehicle, and use sensing technology to obtain the reference object observation information around the vehicle through detection, segmentation, and recognition methods. In step S31, based on the conversion relationship between the detection device and the vehicle coordinate system (i.e., the extrinsic parameter T BS ), convert the reference object to the vehicle coordinate system to obtain the position of the reference object relative to the vehicle. For example, the observation information of the observed lane line of detection device A is P A , and the extrinsic parameter of detection device A is T BA , then the position of the reference object in the vehicle coordinate system is P B = T BA × P A . Among them, the lane reference object can be one or more road elements such as a road edge, a guardrail, a pole sign, a lane line, etc. In one example, the reference object is a lane line.

[0057] S32. Extract the reference object position information within the preset positioning range according to the reference object position;

[0058] As Figure 7 shown, in step S32, in the vehicle coordinate system, perform discrete sampling around the origin at the reference object positions within the preset positioning range, and use the discrete sampling result as the obtained reference object position information. Specifically, divide the preset positioning range (such as -a < x < a, -b < y < b) into n × m sampling areas, the size of each sampling area is (sx, sy), sample each sampling area within the preset positioning range one by one. If there is a reference object in the sampling area, make the value of this sampling area 1, otherwise make the value of this sampling area 0; traverse all sampling areas according to the above rules to obtain the discrete sampling result. The discrete sampling result is, for example Figure 7As shown, the sampling results are filled with gray to represent 1 and filled with white to represent 0 in the figure, and an n-row and m-column reference object observation matrix L is constructed according to the sampling area arrangement as the reference object position information. It should be noted that in order to facilitate the cross-correlation operation and probability fusion between different position information, the preset positioning range and discrete sampling rate in step S32 are the same as those in step S2.

[0059] S33, performing cross-correlation processing on the reference object position information and the initial road information to obtain first cross-correlation information;

[0060] In step S33, the reference object observation matrix L and the initial road matrix R are cross-correlated along the row direction to obtain a first cross-correlation matrix H1. The specific operation method is as follows:

[0061]

[0062] S34. Normalize the first interrelated information to obtain a first probability distribution of vehicles in the navigation map.

[0063] In step S34, the first mutual correlation matrix H1 is normalized along the lateral direction to obtain a first probability distribution of the vehicle in the lateral direction of the road, specifically:

[0064] The first cross-correlation matrix H1 is accumulated column by column to obtain the first row vector h1. The specific operation method is as follows:

[0065]

[0066] The first row vector h1 is normalized to obtain the first probability distribution P1 of the vehicle in the lateral direction of the road. The specific calculation method is as follows:

[0067]

[0068] The first probability P1 reflects the probability value of each position of the vehicle in the lateral direction of the road obtained based on the reference object observation information.

[0069] Next, step S4 is executed to extract the measured dynamic obstacle position information within the preset positioning range, and determine a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information.

[0070] In step S4, within the preset positioning range, the dynamic obstacle position measured by the detection device is matched with the navigation map, so as to obtain a second probability distribution of the vehicle in the navigation map by relying on the characteristic positioning of the surrounding dynamic obstacles. Figure 8 As shown, in some embodiments, step S3 includes:

[0071] S41. Obtain the positions of dynamic obstacles measured around the vehicle, where the positions of the dynamic obstacles are the positions of the dynamic obstacles relative to the vehicle;

[0072] As Fig. 9 shown, the vehicle is equipped with detection devices such as cameras, lidars, and millimeter-wave radars. The road observation information around the vehicle is obtained through the above detection devices, and the observation information of dynamic obstacles around the vehicle is obtained by using sensing technology through detection, segmentation, and recognition methods. The dynamic obstacles can be other vehicles on the lane. In step S41, based on the conversion relationship between the detection device and the vehicle coordinate system (i.e., the extrinsic parameter T BS ), the dynamic obstacles are transformed into the vehicle coordinate system to obtain the positions of the dynamic obstacles relative to the vehicle. For example, the observation information of other vehicles around the observation of detection device A is P A , and the extrinsic parameter of detection device A is T BA , then the position of the dynamic obstacle in the vehicle coordinate system is P B = T BA × P A .

[0073] S42. Extract the dynamic obstacle position information within the preset positioning range according to the positions of the dynamic obstacles;

[0074] As Fig. 9 shown, in step S42, in the vehicle coordinate system, discrete sampling is performed on the positions of obstacles within the preset positioning range around the origin, and the discrete sampling result is used as the obtained dynamic obstacle position information. Specifically, the preset positioning range (such as -a < x < a, -b < y < b) is divided into n × m sampling regions, and the size of each sampling region is (sx, sy). Each sampling region within the preset positioning range is sampled one by one. If there is a dynamic obstacle in the sampling region, the value of the sampling region is set to 1, otherwise the value of the sampling region is set to 0; traversing all sampling regions according to the above rules can obtain the discrete sampling result. The discrete sampling result is, for example, Fig. 9 shown. In the figure, the sampling result is represented by gray filling for 1 and white filling for 0. A dynamic obstacle observation matrix D with n rows and m columns is constructed according to the arrangement of the sampling regions as the dynamic obstacle position information. It should be noted that for the convenience of cross-correlation operation and probability fusion between different position information, the preset positioning range and discrete sampling rate in step S42 are the same as those in step S2.

[0075] S43. Perform cross-correlation processing on the dynamic obstacle position information and the initial road information to obtain the second cross-correlation information;

[0076] In step S43, the dynamic obstacle observation matrix D and the initial road matrix R are cross-correlated in the row direction to obtain a second cross-correlation matrix H2. The specific operation method is as follows:

[0077]

[0078] S44: normalize the second mutual correlation information to obtain a second probability distribution of vehicles in the navigation map.

[0079] In step S44, the second mutual correlation matrix H2 is normalized along the lateral direction to obtain a second probability distribution of the vehicle in the lateral direction of the road, specifically:

[0080] The second cross-correlation matrix H2 is accumulated column by column to obtain the second row vector h2. The specific operation method is as follows:

[0081]

[0082] The second row vector h2 is normalized to obtain the second probability distribution P2 of the vehicle in the lateral direction of the road. The specific calculation method is as follows:

[0083]

[0084] The second probability distribution P2 reflects the probability value of each position of the vehicle in the lateral direction of the road based on the dynamic obstacle observation information.

[0085] Next, executing step S5, determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information;

[0086] like Fig.10 As shown, in some embodiments, step S5 includes:

[0087] S51, obtaining a dead track of the vehicle based on historical movement information of the vehicle;

[0088] In step S51, the vehicle controller can obtain the historical motion information of the vehicle in the past based on sensors such as the Inertial Measurement Unit (IMU), wheel speedometer, and vehicle speedometer equipped on the vehicle. The historical motion information can include the vehicle's traveling speed and direction. The vehicle controller uses the Dead Reckoning (DR) technology to calculate the dead reckoning trajectory of the vehicle within a preset time period in the past based on the historical motion information. Among them, dead reckoning starts from the position at the previous moment and calculates the position at the next moment according to the motion information of sensors such as the IMU, vehicle speedometer, and wheel speedometer, thereby providing relative positioning information during vehicle movement. Specifically, a dead reckoning coordinate system is constructed with the vehicle pose corresponding to the initial moment of the preset time period as the origin. Based on the historical motion information, the vehicle poses at each moment within the preset time period are calculated using dead reckoning technology to obtain the vehicle pose sequence SET TrjLocal ={T DB1 , T DB2 , T DB3 ……T DBn}; First, take the inverse of each vehicle pose T DBn in the vehicle pose sequence to get T BnD . Then, convert the vehicle pose sequence SET TrjLocal in the dead reckoning coordinate system into a vehicle trajectory sequence SET TrjBody ={BP1, BP2, BP3……BPn} in the vehicle coordinate system. Use the vehicle trajectory sequence SET TrjBody in the vehicle coordinate system as the dead reckoning trajectory of the vehicle. Among them, BP1 = T BnB1 , BP2 = T BnB2 , BP3 = T BnB3 ……BPn = T BnBn .

[0089] S52. With the current position of the vehicle as the origin, extract the vehicle dead reckoning trajectory information within the preset positioning range according to the dead reckoning trajectory;

[0090] As Fig.11 shown, in step S52, in the vehicle coordinate system, discrete sampling is performed on the dead reckoning trajectory within the preset positioning range around the origin, and the discrete sampling result is used as the obtained vehicle dead reckoning trajectory information. Specifically, the preset positioning range (such as -a < x < a, -b < y < b) is divided into n×m sampling areas, and the size of each sampling area is (sx, sy). Each sampling area within the preset positioning range is sampled one by one. If there is a dead reckoning trajectory in the sampling area, the value of the sampling area is set to 1, otherwise the value of the sampling area is set to 0. Traversing all sampling areas according to the above rules can obtain the discrete sampling result. The discrete sampling result is, for example Fig.11As shown, the sampling results are filled with gray to represent 1 and filled with white to represent 0 in the figure, and a vehicle position trajectory matrix E with n rows and m columns is constructed according to the arrangement of the sampling areas as the vehicle position trajectory information. It should be noted that in order to facilitate the cross-correlation operation and probability fusion between different position information, the preset positioning range and discrete sampling rate in step S52 are the same as those in step S2.

[0091] S53, performing cross-correlation processing on the vehicle dead position trajectory information and the initial road information to obtain third cross-correlation information;

[0092] In step S53, the vehicle position trajectory matrix E and the initial road matrix R are cross-correlated along the row direction to obtain a third cross-correlation matrix H3. The specific operation method is as follows:

[0093]

[0094] S54: Normalize the third mutual correlation information to obtain a third probability distribution of the vehicle in the navigation map.

[0095] In step S44, the third mutual correlation matrix H2 is normalized along the lateral direction to obtain a third probability distribution of the vehicle in the lateral direction of the road, specifically:

[0096] The third cross-correlation matrix H3 is accumulated column by column to obtain the third row vector h3. The specific operation method is as follows:

[0097]

[0098] The third row vector h3 is normalized to obtain the third probability distribution P3 of the vehicle in the lateral direction of the road. The specific calculation method is as follows:

[0099]

[0100] The third probability P3 reflects the probability value of each position of the vehicle in the lateral direction of the road obtained based on the vehicle position trajectory.

[0101] Next, step S6 is executed to fuse the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution.

[0102] In step S6, the first probability distribution P1, the second probability distribution P2 and the third probability distribution P3 are summed to obtain a fused probability distribution P, that is, P[j]=P1[j]+P2[j]+P3[j]; wherein the fused probability distribution P reflects the probability value of each position of the vehicle in the lateral direction of the road obtained by integrating the probability distributions of multiple positioning methods.

[0103] Next, step S7 is executed to obtain and update the location of the vehicle in the navigation map based on the fused probability distribution.

[0104] In step S7, the position corresponding to the maximum probability in the fused probability distribution P is used as the positioning position to update the position of the vehicle in the navigation map.

[0105] like Fig.12 As shown, in some embodiments, the present invention provides a vehicle positioning system 10, which is applied in a vehicle controller, and the vehicle positioning system 10 corresponds to the vehicle positioning method in the above embodiment. The vehicle positioning system 10 includes a preliminary positioning unit 11, a road initialization unit 12, a first prediction unit 13, a second prediction unit 14, a third prediction unit 15, a probability fusion unit 16 and a positioning update unit 17. Each functional module is described in detail as follows:

[0106] A preliminary positioning unit 11 is used to obtain the vehicle position and load a navigation map corresponding to the vehicle position;

[0107] A road initialization unit 12 extracts initial road information within a preset positioning range around the vehicle position based on the navigation map;

[0108] The first prediction unit 13 extracts the measured reference object position information within the preset positioning range, and determines a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information;

[0109] The second prediction unit 14 extracts the measured dynamic obstacle position information within the preset positioning range, and determines a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information;

[0110] A third prediction unit 15 determines a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information;

[0111] A probability fusion unit 16, which fuses the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution;

[0112] The positioning updating unit 17 acquires and updates the positioning information of the vehicle in the navigation map based on the fused probability distribution.

[0113] In some embodiments, the road initialization unit 12 is specifically configured to:

[0114] In the navigation map, a road coordinate system is established around the vehicle position;

[0115] In the road coordinate system, discrete sampling is performed within a preset positioning range according to preset rules to generate initial road information.

[0116] In some embodiments, the first prediction unit 13 is specifically configured to:

[0117] Acquire a reference object position measured around the vehicle, where the reference object position is a position of the reference object relative to the vehicle;

[0118] According to the position of the reference object, extracting the reference object position information within the preset positioning range;

[0119] Performing cross-correlation processing on the reference object position information and the initial road information to obtain first cross-correlation information;

[0120] The first intercorrelated information is normalized to obtain a first probability distribution of vehicles in the navigation map.

[0121] In some embodiments, the second prediction unit 14 is specifically configured to:

[0122] Acquire a dynamic obstacle position measured around the vehicle, where the dynamic obstacle position is a position of the dynamic obstacle relative to the vehicle;

[0123] According to the dynamic obstacle position, extracting dynamic obstacle position information within the preset positioning range;

[0124] Performing cross-correlation processing on the dynamic obstacle position information and the initial road information to obtain second cross-correlation information;

[0125] The second mutually correlated information is normalized to obtain a second probability distribution of vehicles in the navigation map.

[0126] In some embodiments, the third prediction unit 15 is specifically configured to:

[0127] Based on the historical movement information of the vehicle, obtain the vehicle's dead position trajectory;

[0128] Taking the current position of the vehicle as the origin, extracting the vehicle dead track information within the preset positioning range according to the dead track;

[0129] Performing cross-correlation processing on the vehicle dead position trajectory information and the initial road information to obtain third cross-correlation information;

[0130] The third mutual correlation information is normalized to obtain a third probability distribution of the vehicle in the navigation map.

[0131] In some embodiments, the probability fusion unit 16 is specifically configured to:

[0132] The first probability distribution, the second probability distribution and the third probability distribution are summed to obtain a fused probability distribution.

[0133] In some embodiments, the location update unit 17 is specifically configured to:

[0134] The position corresponding to the maximum probability in the fused probability distribution is used as the positioning position to update the position of the vehicle in the navigation map.

[0135] In some embodiments, a computer device is provided, the internal structure diagram of which can be as follows: Fig.13 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer program is executed by the processor, the functions or steps of a vehicle positioning method are implemented.

[0136] In some embodiments, a computer device is provided, the computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the following steps are implemented:

[0137] Obtaining the vehicle position, and loading a navigation map corresponding to the vehicle position;

[0138] Based on the navigation map, extracting initial road information within a preset positioning range around the vehicle position;

[0139] Extracting the measured reference object position information within the preset positioning range, and determining a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information;

[0140] Extracting the measured dynamic obstacle position information within the preset positioning range, and determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information;

[0141] Determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information;

[0142] Merging the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution;

[0143] The location of the vehicle in the navigation map is obtained and updated based on the fused probability distribution.

[0144] In some embodiments, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0145] Obtaining the vehicle position, and loading a navigation map corresponding to the vehicle position;

[0146] Based on the navigation map, extracting initial road information within a preset positioning range around the vehicle position;

[0147] Extracting the measured reference object position information within the preset positioning range, and determining a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information;

[0148] Extracting the measured dynamic obstacle position information within the preset positioning range, and determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information;

[0149] Determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information;

[0150] Merging the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution;

[0151] The location of the vehicle in the navigation map is obtained and updated based on the fused probability distribution.

[0152] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0153] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink), DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0154] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0155] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A vehicle positioning method, characterized in that: include: Obtaining the vehicle position, and loading a navigation map corresponding to the vehicle position; Based on the navigation map, extracting initial road information within a preset positioning range around the vehicle position; Extracting the measured reference object position information within the preset positioning range, and determining a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information; Extracting the measured dynamic obstacle position information within the preset positioning range, and determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information; Determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information; Merging the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution; The location of the vehicle in the navigation map is obtained and updated based on the fused probability distribution.

2. The vehicle positioning method according to claim 1, characterized in that: Extracting initial road information within a preset positioning range around the vehicle position based on the navigation map includes: In the navigation map, a road coordinate system is established around the vehicle position; In the road coordinate system, discrete sampling is performed within a preset positioning range according to preset rules to generate initial road information.

3. The vehicle positioning method according to claim 1, characterized in that: Extracting the measured reference object position information within the preset positioning range, and determining a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information, including: Acquire a reference object position measured around the vehicle, where the reference object position is a position of the reference object relative to the vehicle; According to the position of the reference object, extracting the reference object position information within the preset positioning range; Performing cross-correlation processing on the reference object position information and the initial road information to obtain first cross-correlation information; The first intercorrelated information is normalized to obtain a first probability distribution of vehicles in the navigation map.

4. The vehicle positioning method according to claim 1, characterized in that: Extracting the measured dynamic obstacle position information within the preset positioning range, and determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information, including: Acquire a dynamic obstacle position measured around the vehicle, where the dynamic obstacle position is a position of the dynamic obstacle relative to the vehicle; According to the dynamic obstacle position, extracting dynamic obstacle position information within the preset positioning range; Performing cross-correlation processing on the dynamic obstacle position information and the initial road information to obtain second cross-correlation information; The second mutually correlated information is normalized to obtain a second probability distribution of vehicles in the navigation map.

5. The vehicle positioning method according to claim 1, characterized in that: Determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information includes: Based on the historical movement information of the vehicle, obtain the vehicle's dead position trajectory; Taking the current position of the vehicle as the origin, extracting the vehicle dead track information within the preset positioning range according to the dead track; Performing cross-correlation processing on the vehicle dead position trajectory information and the initial road information to obtain third cross-correlation information; The third mutual correlation information is normalized to obtain a third probability distribution of the vehicle in the navigation map.

6. The vehicle positioning method according to claim 1, characterized in that: The first probability distribution, the second probability distribution and the third probability distribution are combined to obtain a combined probability distribution, including: The first probability distribution, the second probability distribution and the third probability distribution are summed to obtain a fused probability distribution.

7. The vehicle positioning method according to claim 1, characterized in that: Acquiring and updating the positioning information of the vehicle in the navigation map based on the fused probability distribution includes: The position corresponding to the maximum probability in the fused probability distribution is used as the positioning position to update the position of the vehicle in the navigation map.

8. A vehicle positioning system, characterized in that: include: A preliminary positioning unit, used to obtain the vehicle position and load a navigation map corresponding to the vehicle position; A road initialization unit, which extracts initial road information within a preset positioning range around the vehicle position based on the navigation map; a first prediction unit, which extracts the measured reference object position information within the preset positioning range, and determines a first probability distribution of the vehicle in the navigation map based on the reference object position information and the initial road information; a second prediction unit, extracting the measured dynamic obstacle position information within the preset positioning range, and determining a second probability distribution of the vehicle in the navigation map based on the dynamic obstacle position information and the initial road information; a third prediction unit, determining a third probability distribution of the vehicle in the navigation map based on the historical movement information of the vehicle and the initial road information; A probability fusion unit, which fuses the first probability distribution, the second probability distribution and the third probability distribution to obtain a fused probability distribution; A positioning update unit is used to obtain and update the positioning information of the vehicle in the navigation map based on the fused probability distribution.

9. A computer device, characterized in that: include: Processor and memory; The memory is used to store computer programs; The processor is connected to the memory, and the processor is used to execute the computer program stored in the memory, so that the computer device executes the vehicle positioning method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the vehicle positioning method according to any one of claims 1 to 7 is implemented.