A vehicle positioning correction method, storage medium, and terminal device

Abstract

This invention discloses a vehicle positioning correction method, storage medium, and terminal device. The method obtains several fitted lane lines based on several map lane lines within a preset range of the vehicle's current positioning point in a high-precision map. It then synchronizes the collected sensing lane lines with the fitted lane lines in time. The method arranges and combines the fitted lane lines and the time-synchronized sensing lane lines, calculates the total matching error between the fitted and sensing lane lines for each combination, and selects the combination with the smallest total matching error as the optimal combination. Based on the coordinate data of map geometric points on the fitted lane lines and the curve coefficients of the sensing lane lines in each lane line pair of the optimal combination, the positioning coordinates of the current positioning point are corrected. This method can correct the vehicle's positioning coordinates, thereby improving the vehicle's positioning accuracy.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a vehicle positioning correction method, a computer-readable storage medium, and a terminal device. Background Technology

[0002] With the rapid development of autonomous driving technology, the requirements for vehicle positioning accuracy are also increasing. However, in vehicles that use GNSS (Global Navigation Satellite System) / INS (Inertial Navigation System) for integrated navigation and positioning, if the integrated navigation system is in a GNSS-denied condition for an extended period of time, cumulative errors will inevitably occur, thus affecting the vehicle's positioning accuracy. Summary of the Invention

[0003] The purpose of this invention is to provide a vehicle positioning correction method, a computer-readable storage medium, and a terminal device that can correct errors in the positioning coordinates of a vehicle, thereby improving the positioning accuracy of the vehicle.

[0004] To achieve the above objectives, embodiments of the present invention provide a vehicle positioning correction method, comprising:

[0005] Several fitted lane lines are obtained based on several map lane lines within a preset range of the vehicle's current location point in the high-precision map;

[0006] The sensing lane lines collected at the current location point are synchronized with the fitted lane lines in time.

[0007] The fitted lane lines and the time-synchronized sensing lane lines are arranged and combined; wherein each combination includes at least one pair of lane lines, and each pair of lane lines includes one fitted lane line and one corresponding sensing lane line.

[0008] Based on the lane matching error corresponding to each pair of lane lines in each combination, the total matching error corresponding to each combination is obtained, and the combination with the smallest total matching error is taken as the optimal combination.

[0009] The positioning coordinates of the current positioning point are corrected based on the coordinate data of the map geometric points on the fitted lane lines and the curve coefficients of the perceived lane lines in each pair of lane lines in the optimal combination method.

[0010] Furthermore, the step of time-synchronizing the several perceived lane lines collected at the current positioning point with the several fitted lane lines specifically includes:

[0011] Acquire several lane lines sensed by the vehicle within the preset range;

[0012] According to the vehicle at t hdmap to t perception The distance traveled within a time period determines the translation distance; where t hdmap This indicates the timestamp for extracting the lane lines from the map, t. perception Indicates the timestamp of the collected lane line data;

[0013] Based on the translation distance, the curve equations of the plurality of sensing lane lines are derived from t. perception The perception of time involves translating and transforming the vehicle's coordinate system to t. hdmap The map uses the vehicle's coordinate system at any given time to obtain several perceived lane lines after time synchronization.

[0014] Furthermore, the step of obtaining the total matching error corresponding to each combination method based on the lane line matching error corresponding to each pair of lane lines in each combination method specifically includes:

[0015] For each combination method, calculate the lane matching error between the fitted lane line and the perceived lane line in each lane line pair in the same combination method, and take the sum of the lane matching errors of all lane line pairs in the same combination method as the total matching error corresponding to the same combination method.

[0016] For each lane pair, the lane matching error between the fitted lane line and the perceived lane line in the same lane pair is calculated using the following formula:

[0017]

[0018] Where n represents the number of map geometric points on the fitted lane line, (x i_hdmap y i_hdmap ) represents the coordinate data of the i-th map geometric point in the map vehicle coordinate system, and c0, c1, c2 and c3 represent the cubic curve coefficients of the perceived lane line.

[0019] Further, the step of correcting the positioning coordinates of the current positioning point based on the coordinate data of the map geometric points on the fitted lane lines in each lane line pair in the optimal combination method and the curve coefficient of the perceived lane lines specifically includes:

[0020] The coordinate data of the map geometric points on the fitted lane lines and the curve coefficients of the perceived lane lines in each pair of lane lines in the optimal combination method are substituted into the optimization function to calculate the lateral error δd and the heading error δθ of the vehicle.

[0021] Based on the lateral error δd and the heading error δθ, the positioning coordinates of the current positioning point are compensated for errors to obtain the corrected positioning coordinates;

[0022] The optimization function is:

[0023]

[0024] Among them, (x i_hdmap y i_hdmap ) represents the coordinate data of the i-th map geometric point on the fitted lane line in the map vehicle coordinate system, and c0, c1, c2 and c3 represent the cubic curve coefficients of the perceived lane line.

[0025] Furthermore, obtaining several fitted lane lines based on several map lane lines within a preset range of the vehicle's current location in the high-precision map specifically includes:

[0026] Extract several map lane lines within a preset range of the vehicle's current location in the high-precision map;

[0027] Convert the latitude and longitude coordinates of the map geometric points on the several map lane lines into coordinate data in the map vehicle coordinate system;

[0028] The map lane lines are stitched together based on the coordinate data of the map geometric points, and curve fitting is performed on each stitched map lane line to obtain several fitted lane lines.

[0029] Further, the step of stitching together the plurality of map lane lines based on the coordinate data of the map geometric points, and performing curve fitting on each stitched map lane line to obtain a plurality of fitted lane lines, specifically includes:

[0030] Based on the coordinate data of the map geometric points on the several map lane lines, and combined with the preceding and succeeding relationships of lanes in the high-precision map, the several map lane lines are spliced ​​together to obtain several spliced ​​map lane lines.

[0031] The least squares method was used to perform curve fitting on each lane line of the stitched map to obtain several fitted lane lines.

[0032] Further, the step of stitching together the map lane lines based on the coordinate data of the map geometric points on the several map lane lines and the preceding and succeeding relationships of lanes in the high-precision map to obtain several stitched map lane lines specifically includes:

[0033] Based on the coordinate data of the map geometric points on the several map lane lines, and combined with the preceding and succeeding relationships of lanes in the high-precision map, it is determined whether there is a connection relationship between the several map lane lines.

[0034] Map lane lines that are connected are stitched together to obtain several stitched map lane lines. Map lane lines that are not connected to other map lane lines are directly used as the stitched map lane lines.

[0035] Furthermore, the step of using the least squares method to perform curve fitting on each lane line of the stitched map to obtain several fitted lane lines specifically includes:

[0036] Based on the coordinate data of the map geometric points on each map lane line after stitching, the least squares method is used to perform curve fitting on each map lane line after stitching to obtain several fitted lane lines.

[0037] Each fitted lane line is a cubic curve, and the cubic curve coefficients a0, a1, a2, and a3 of each fitted lane line are expressed by the formula... Calculated, (r) front r right This represents the coordinate data of map geometric points on the map lane lines in the map's vehicle coordinate system.

[0038] This invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the vehicle positioning correction method described above.

[0039] This invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the vehicle positioning correction method described above when executing the computer program.

[0040] Compared with existing technologies, embodiments of the present invention provide a vehicle positioning correction method, a computer-readable storage medium, and a terminal device. The method obtains several fitted lane lines based on several map lane lines within a preset range of the vehicle's current positioning point in a high-precision map. It then performs time synchronization processing on several perceived lane lines collected at the current positioning point and the fitted lane lines, and arranges and combines the fitted lane lines and the time-synchronized perceived lane lines. Based on the lane line matching error corresponding to each pair of lane lines in each combination, it obtains the total matching error corresponding to each combination method, and selects the combination method with the smallest total matching error as the optimal combination method. Finally, it corrects the positioning coordinates of the current positioning point based on the coordinate data of map geometric points on the fitted lane lines and the curve coefficients of the perceived lane lines in each pair of lane lines in the optimal combination method. This method can correct the vehicle's positioning coordinates, thereby improving the vehicle's positioning accuracy. Attached Figure Description

[0041] Figure 1 This is a flowchart of a preferred embodiment of a vehicle positioning correction method provided by the present invention;

[0042] Figure 2 This is a schematic diagram illustrating an application scenario of a vehicle positioning correction method provided in an embodiment of the present invention;

[0043] Figure 3 This is a structural block diagram of a preferred embodiment of a terminal device provided by the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] This invention provides a vehicle positioning correction method, see [link to relevant documentation]. Figure 1 The diagram shown is a flowchart of a preferred embodiment of a vehicle positioning correction method provided by the present invention, the method comprising steps S11 to S15:

[0046] Step S11: Obtain several fitted lane lines based on several map lane lines within a preset range of the vehicle's current location point in the high-precision map.

[0047] In practical implementation, the current location of the vehicle can be determined first using a high-precision map. Then, a preset range corresponding to the current location on the high-precision map can be determined (i.e., the area around the current location, for example, a circular area with the current location as the center and a preset value as the radius). Within the preset range, high-precision roads with a distance less than a preset distance threshold from the current location can be further determined. Based on the determined high-precision roads, several map lane lines on the high-precision roads can be extracted. Based on the extracted map lane lines, curve fitting can be performed to obtain several fitted lane lines.

[0048] It should be noted that high-precision maps are dedicated maps for autonomous driving. They consist of vector information such as lane models, road components, and road attributes containing semantic information, as well as feature layers for multi-sensor positioning. With the assistance of high-precision maps, autonomous vehicles can more easily determine their own position and also obtain information such as driving area, driving direction, and driving lane.

[0049] As a preferred embodiment, obtaining several fitted lane lines based on several map lane lines within a preset range of the vehicle's current location point in the high-precision map specifically includes:

[0050] Extract several map lane lines within a preset range of the vehicle's current location in the high-precision map;

[0051] Convert the latitude and longitude coordinates of the map geometric points on the several map lane lines into coordinate data in the map vehicle coordinate system;

[0052] The map lane lines are stitched together based on the coordinate data of the map geometric points, and curve fitting is performed on each stitched map lane line to obtain several fitted lane lines.

[0053] In practical implementation, while extracting several map lane lines within a preset range of the vehicle's current location from the high-precision map, relevant information about the map lane lines can also be obtained. For example, the latitude and longitude coordinates of the map geometric points on the map lane lines can be obtained. Then, the latitude and longitude coordinates of the map geometric points on each map lane line can be converted into coordinate data in the map's vehicle coordinate system. Based on the coordinate data of the map geometric points on several map lane lines in the map's vehicle coordinate system, combined with the lane connection relationship provided by the high-precision map, the extracted map lane lines can be stitched together to obtain several stitched map lane lines. Then, curve fitting is performed on each stitched map lane line to calculate the curve equation of several fitted lane lines in the map's vehicle coordinate system. One stitched map lane line corresponds to one fitted lane line.

[0054] It should be noted that the vehicle coordinate system of the map after high-precision map conversion is generally the front right lower coordinate system. The latitude and longitude coordinates of map geometric points on the lane lines can be converted to coordinate data in the map's vehicle coordinate system using the following formula:

[0055]

[0056] Among them, (r front r right ) represents the coordinate data of map geometric points on the map lane lines in the map's vehicle coordinate system, r front Let r be the forward coordinate in the vehicle's coordinate system. right Let R be the rightward coordinate in the vehicle coordinate system, where ψ represents the heading angle and R is the leftward coordinate. M R represents the radius of curvature of the Earth's meridian. N The radius of curvature of the Earth's geoid is represented by h, and the altitude of the current location is represented by h. Indicates the latitude of the current location point. δλ represents the difference between the latitude of the map geometric point and the latitude of the current location point, and δλ represents the difference between the longitude of the map geometric point and the longitude of the current location point.

[0057] For example, the values ​​of each parameter in the formula only need to conform to their definition. For instance, the heading angle ψ is generally defined as the range of values ​​clockwise from 0° north (-180°, 180°), i.e., 0° north, 90° east, -90° west, and 180° east. The definition of the heading angle ψ can also be adjusted according to the specific usage. For example, it can be defined as [0°, 360°] clockwise from 0° north, in which case west is 270°, and the rest of the definitions remain unchanged.

[0058] See Figure 2 The diagram shown is an application scenario illustration of a vehicle positioning correction method provided by an embodiment of the present invention. Figure 2 The image shows a typical off-ramp scenario. Figure 2 The solid lines (L11, L12, L13, L14, L15, L21, L22, L23, and L24) are lane lines or boundaries. Figure 2 The dashed line in the diagram represents the lane centerline. Assuming the current location point is at the intersection of high-precision road 1 and high-precision road 2, and within a preset range of the current location point, the two high-precision roads closest to the current location point are determined as follows: Figure 2If a vehicle is assigned a high-precision road (High-Precision Road 1) and a high-precision road (High-Precision Road 2), the high-precision map will send the lane information of High-Precision Road 1 and High-Precision Road 2 to the vehicle's positioning system. The vehicle's positioning system will then extract all the map lane lines (i.e., L11, L12, L13, L14, L15, L21, L22, L23, and L24) from High-Precision Road 1 and High-Precision Road 2, convert the latitude and longitude coordinates of the map geometric points on the map lane lines into coordinate data in the map's vehicle coordinate system, and then, based on the lane connection relationship provided by the high-precision map, adjust the high-precision road... All map lane lines in 1 and high-precision road 2 are spliced ​​together according to their connection relationships. Specifically, L11 is spliced ​​with L21, L12 with L22, L13 with L23, L14 with L24, and L15 with L24, resulting in 5 spliced ​​map lane lines (i.e., L11+L21, L12+L22, L13+L23, L14+L24, and L15+L24). Then, curve fitting is performed on these 5 map lane lines to obtain the curve equations of the 5 fitted lane lines.

[0059] It should be noted that curve fitting of the map lane lines is mainly to obtain the curve equation of the high-precision map lane lines (i.e., the curve equation of the fitted lane lines). This equation is used to find the optimal match between the map lane lines (i.e., the fitted lane lines) and the perceived lane lines. The map lane lines are then compared with the optimally matched perceived lane lines to prevent significant differences in shape and position between the map lane lines and the optimally matched perceived lane lines from resulting in an abnormal correction amount that could affect the positioning error correction. For example, as can be seen from the following embodiments, in this invention, both the map lane lines and the perceived lane lines are cubic curves. The comparison method generally involves checking whether the difference between the constant terms a0 and c0 in the cubic curve equations exceeds a threshold, and whether the difference between the quadratic coefficients a2 and c2 in the cubic curve equations exceeds a threshold. The constant term comparison checks the position, and the quadratic coefficient comparison checks the shape.

[0060] Step S12: Synchronize the several perceived lane lines collected at the current positioning point with the several fitted lane lines in time.

[0061] In practice, the collected lane lines need to be synchronized with the fitted lane lines (or map lane lines) in time. This can be done by using the difference between the timestamp of the high-precision map data and the timestamp of the perception, and by using the positioning speed at the time the high-precision map was generated to perform a translation. This transforms the curve equation of the perceived lane lines from the perception vehicle coordinate system to the map vehicle coordinate system after the high-precision map is converted, so that the matching of the fitted lane lines and the perceived lane lines is performed in the same coordinate system.

[0062] Step S13: Arrange and combine the several fitted lane lines and the several time-synchronized sensing lane lines; wherein each combination includes at least one pair of lane lines, and each pair of lane lines includes one fitted lane line and one corresponding sensing lane line.

[0063] In practice, after processing the map lane lines and the perceived lane lines, we can obtain the number of effective map lane lines (i.e., fitted lane lines) and the number of perceived lane lines (i.e., perceived lane lines after time synchronization). Generally, the number of fitted lane lines should be greater than or equal to the number of perceived lane lines. Let's assume the number of fitted lane lines is represented as n. hdmap The number of lane lines to be sensed is represented by n. perception Then for n hdmap Fitted lane lines and n perception By arranging and combining the sensing lane lines, C can be obtained. n n h p d e m rc a e p ption There are several combination methods, and in each combination method, there is at least one pair of lane lines. In each pair of lane lines, there is a fitted lane line and a sensing lane line corresponding to the fitted lane line. There is a correspondence between the fitted lane line and the sensing lane line in the same pair of lane lines.

[0064] For example, suppose n hdmap =4,n perception There are four possible combinations between 3 or 4 fitted lane lines and 3 perceived lane lines, as shown in Table 1. In Table 1, map0, map1, map2, and map3 represent the 4 fitted lane lines from right to left, and per0, per1, and per2 represent the 3 perceived lane lines from right to left. In combination 1, there are 3 pairs of lane lines: map0 and per0 form one pair (map0 and per0 have a corresponding relationship), map1 and per1 form another pair (map1 and per1 have a corresponding relationship), and map2 and per2 form another pair (map2 and per2 have a corresponding relationship). However, map3 does not form a lane line pair with any of the perceived lane lines, nor does map3 have a corresponding relationship with any of the perceived lane lines. The same applies to combinations 2 through 4, which will not be elaborated here.

[0065] Table 1. Combination methods of fitted lane lines and perceived lane lines

[0066] map0 map1 map2 map3 Combination Method 1 per0 Per1 Per2 Combination Method 2 per0 Per1 Per2 Combination method 3 per0 Per1 Per2 Combination method 4 per0 Per1 Per2

[0067] Step S14: Based on the lane line matching error corresponding to each pair of lane lines in each combination method, obtain the total matching error corresponding to each combination method, and take the combination method corresponding to the minimum total matching error as the optimal combination method.

[0068] Step S15: Correct the positioning coordinates of the current positioning point based on the coordinate data of the map geometric points on the fitted lane lines in each pair of lane lines in the optimal combination method and the curve coefficient of the perceived lane lines.

[0069] In practical implementation, after obtaining the combination of map lane lines (i.e., fitted lane lines) and perceived lane lines (i.e., perceived lane lines after time synchronization), for each pair of lane lines in each combination, the lane line matching error between the fitted lane lines and perceived lane lines in the same pair is calculated. Based on the lane line matching errors corresponding to all lane line pairs in the same combination, the total matching error corresponding to the same combination is obtained. Among the total matching errors corresponding to all combination methods, the combination with the smallest total matching error is taken as the optimal combination. Then, based on the coordinate data of the map geometric points on the fitted lane lines contained in each pair of lane lines in the optimal combination in the map's vehicle coordinate system and the curve coefficient of the curve equation of the perceived lane lines, the positioning coordinates of the vehicle's current positioning point are corrected, and the corrected positioning coordinates are obtained accordingly.

[0070] For example, as shown in Table 1, combination mode 1 includes 3 lane line pairs. Therefore, for combination mode 1, we can first calculate the lane line matching error between map0 and per0, the lane line matching error between map1 and per1, and the lane line matching error between map2 and per2. Then, we can calculate the total matching error corresponding to combination mode 1 based on the lane line matching errors corresponding to these 3 lane line pairs. The same applies to combination modes 2 to 4, which will not be elaborated here.

[0071] The vehicle positioning correction method provided in this invention can be applied to assisted driving navigation functions in scenarios such as urban expressways and highways. It utilizes the arrangement and combination of map lane lines and perceived lane lines in a high-precision map to find the optimal matching method between them, thereby obtaining the corresponding relationship between the map lane lines and the perceived lane lines. By combining the coordinate data of map geometric points on the map lane lines in the map's vehicle coordinate system and the curve coefficients of the curve equation of the perceived lane lines, the positioning coordinates are optimized and corrected. This overcomes the problem of accumulated errors that inevitably occur when the GNSS / INS integrated navigation system is under GNSS rejection conditions for extended periods. By correcting the positioning coordinates of the GNSS / INS integrated navigation system and correcting the errors in the vehicle's positioning coordinates, the vehicle's positioning accuracy is improved.

[0072] In another preferred embodiment, the step of time-synchronizing the plurality of perceived lane lines collected at the current positioning point with the plurality of fitted lane lines specifically includes:

[0073] Acquire several lane lines sensed by the vehicle within the preset range;

[0074] According to the vehicle at t hdmap to t perception The distance traveled within a time period determines the translation distance; where t hdmap This indicates the timestamp for extracting the lane lines from the map, t. perception Indicates the timestamp of the collected lane line data;

[0075] Based on the translation distance, the curve equations of the plurality of sensing lane lines are derived from t. perception The perception of time involves translating and transforming the vehicle's coordinate system to t. hdmap The map uses the vehicle's coordinate system at any given time to obtain several perceived lane lines after time synchronization.

[0076] Specifically, in conjunction with the above embodiments, when synchronizing the perceived lane lines with the fitted lane lines (or map lane lines) in time, several perceived lane lines collected by the vehicle within a preset range of the current positioning point can be acquired first. For example, several perceived lane lines can be acquired using a camera in the vehicle within a preset range of the current positioning point; then, based on the vehicle's position at time t... hdmap to t perception The distance traveled within a time period determines the translation distance, where t hdmap This indicates the timestamp for extracting the lane lines from the map, t. perception This represents the timestamp for collecting the perceived lane lines; then, based on the determined translation distance, the curve equations of several perceived lane lines are calculated from t... perception The perception of time involves translating and transforming the vehicle's coordinate system to t. hdmapThe map's vehicle coordinate system at any given time is used to obtain several perceived lane lines after time synchronization. For example, the curve equations of the perceived lane lines can be obtained from t... perception The vehicle's coordinate system is translated according to the translation distance to obtain t. hdmap The curve equation in the perceived vehicle coordinate system at any given time is obtained by performing a coordinate system transformation between the perceived vehicle coordinate system and the map vehicle coordinate system, thus yielding t. hdmap The curve equation in the vehicle coordinate system of the map at any given time is obtained, which is the perceived lane line after time synchronization.

[0077] It should be noted that, in the actual calculation of the translation distance, it is assumed that the vehicle is at t hdmap to t perception If the driving speed remains constant over a time period, then it can be determined by t. hdmap With t perception Multiply the time difference between the two points by the vehicle's speed to calculate the distance traveled. Then, directly assign the distance between the two points in time t. hdmap to t perception The driving distance within a time period is used as the translation distance to compensate for the perceived lane line.

[0078] It should be noted that when performing coordinate system transformation, the transformation formula needs to be analyzed specifically according to the defined coordinate system. For example, if the vehicle coordinate system used for perception is the front left upper coordinate system, and the coordinate system after high-precision map transformation is the front right lower coordinate system, then the formula for time compensation is as follows:

[0079]

[0080] Where c0, c1, c2, and c3 represent the cubic curve coefficients of the perceived lane line after coordinate transformation, b0, b1, b2, and b3 represent the cubic curve coefficients of the perceived lane line before coordinate transformation, f represents the forward movement distance of the vehicle, r represents the rightward movement distance of the vehicle, and v x Indicates the vehicle at t hdmap The forward velocity at time v y Indicates the vehicle at t perception The rightward velocity at time t.

[0081] In yet another preferred embodiment, obtaining the total matching error corresponding to each combination method based on the lane line matching error corresponding to each pair of lane lines in each combination method specifically includes:

[0082] For each combination method, calculate the lane matching error between the fitted lane line and the perceived lane line in each lane line pair in the same combination method, and take the sum of the lane matching errors of all lane line pairs in the same combination method as the total matching error corresponding to the same combination method.

[0083] For each lane pair, the lane matching error between the fitted lane line and the perceived lane line in the same lane pair is calculated using the following formula:

[0084]

[0085] Where n represents the number of map geometric points on the fitted lane line, (x i_hdmap y i_hdmap ) represents the coordinate data of the i-th map geometric point in the map vehicle coordinate system, and c0, c1, c2 and c3 represent the cubic curve coefficients of the perceived lane line.

[0086] Specifically, in conjunction with the above embodiments, when obtaining the total matching error corresponding to each combination method, for each combination method, the lane line matching error between the fitted lane line and the perceived lane line in each pair of lane lines in the same combination method can be calculated first. Then, the sum of the lane line matching errors corresponding to all lane line pairs in the same combination method is taken as the total matching error corresponding to the same combination method.

[0087] It should be noted that, for each lane pair in each combination method, the lane matching error between the fitted lane line and the perceived lane line in the same lane pair can be calculated using the following formula:

[0088]

[0089] Where, in the formula, n represents the number of map geometric points on the corresponding fitted lane lines in the same set of lane line pairs, n>1, (x i_hdmap y i_hdmap ) represents the coordinate data of the i-th (i = 1, 2, ..., n) map geometric point on the fitted lane line in the map vehicle coordinate system, and c0, c1, c2 and c3 represent the cubic curve coefficients of the corresponding perceived lane lines in the same set of lane line pairs.

[0090] In another preferred embodiment, the step of correcting the positioning coordinates of the current positioning point based on the coordinate data of the map geometric points on the fitted lane lines and the curve coefficients of the perceived lane lines in each lane line pair of the optimal combination method specifically includes:

[0091] The coordinate data of the map geometric points on the fitted lane lines and the curve coefficients of the perceived lane lines in each pair of lane lines in the optimal combination method are substituted into the optimization function to calculate the lateral error δd and the heading error δθ of the vehicle.

[0092] Based on the lateral error δd and the heading error δθ, the positioning coordinates of the current positioning point are compensated for errors to obtain the corrected positioning coordinates;

[0093] The optimization function is:

[0094]

[0095] Among them, (x i_hdmap y i_hdmap ) represents the coordinate data of the i-th map geometric point on the fitted lane line in the map vehicle coordinate system, and c0, c1, c2 and c3 represent the cubic curve coefficients of the perceived lane line.

[0096] Specifically, in conjunction with the above embodiments, when actually correcting the positioning coordinates of the vehicle's current location point according to the optimal combination method, for each lane line pair in the optimal combination method, the coordinate data of the map geometric points on the corresponding fitted lane lines in the same lane line pair and the curve coefficients of the curve equation of the perceived lane line are substituted into the optimization function for calculation. Based on the calculation results corresponding to each lane line pair in the optimal combination method, the lateral error δd and heading error δθ of the vehicle are obtained. Then, the positioning coordinates of the vehicle's current location point are compensated for errors based on the obtained lateral error δd and heading error δθ, and the corrected positioning coordinates are obtained accordingly.

[0097] It should be noted that the formula for the optimization function is:

[0098]

[0099] Among them, (x i_hdmap y i_hdmap ) represents the coordinate data of the i-th map geometric point on the fitted lane line with corresponding relationship in the same set of lane line pairs in the map vehicle coordinate system, and c0, c1, c2 and c3 represent the cubic curve coefficients of the perceived lane line with corresponding relationship in the same set of lane line pairs.

[0100] In another preferred embodiment, the step of stitching together the plurality of map lane lines based on the coordinate data of the map geometric points, and then performing curve fitting on each of the stitched map lane lines to obtain a plurality of fitted lane lines, specifically includes:

[0101] Based on the coordinate data of the map geometric points on the several map lane lines, and combined with the preceding and succeeding relationships of lanes in the high-precision map, the several map lane lines are spliced ​​together to obtain several spliced ​​map lane lines.

[0102] The least squares method was used to perform curve fitting on each lane line of the stitched map to obtain several fitted lane lines.

[0103] Specifically, in conjunction with the above embodiments, when stitching and curve fitting map lane lines, the coordinate data of the map geometric points on each map lane line in the map's vehicle coordinate system can be used first. Combined with the preceding and succeeding relationships of lanes in the high-precision map, the extracted map lane lines can be stitched together to obtain several stitched map lane lines. Then, the least squares method is used to perform curve fitting on each stitched map lane line to calculate the curve equation of each map lane line in the map's vehicle coordinate system, thereby obtaining the fitted lane line corresponding to each map lane line.

[0104] It should be noted that the preceding and succeeding relationships of lanes in a high-definition map mainly reflect the connectivity of lanes and are an attribute of lane elements in the high-definition map. Each lane element in a high-definition map has a globally unique ID. The aforementioned attribute includes the IDs of all lane elements connected to the current lane. Lane elements traveling in the same direction as the current vehicle are considered preceding lanes, while those traveling in the opposite direction are considered succeeding lanes. Therefore, lane stitching can be performed based on these lane element IDs, followed by map lane line stitching.

[0105] As an improvement to the above solution, the step of stitching together the map lane lines based on the coordinate data of the map geometric points on the several map lane lines and the preceding and succeeding relationships of lanes in the high-precision map to obtain several stitched map lane lines specifically includes:

[0106] Based on the coordinate data of the map geometric points on the several map lane lines, and combined with the preceding and succeeding relationships of lanes in the high-precision map, it is determined whether there is a connection relationship between the several map lane lines.

[0107] Map lane lines that are connected are stitched together to obtain several stitched map lane lines. Map lane lines that are not connected to other map lane lines are directly used as the stitched map lane lines.

[0108] Specifically, in conjunction with the above embodiments, when actually stitching map lane lines, the coordinate data of the map geometric points on each map lane line in the map's vehicle coordinate system can be used, combined with the preceding and succeeding relationships of lanes in the high-precision map, to determine whether there is a connection between map lane lines. If so, the map lane lines with a connection relationship are stitched together to obtain several stitched map lane lines. It is understandable that among the extracted map lane lines, there may also be independent map lane lines that are not connected to other map lane lines. For independent map lane lines, there is no need to stitch them together; they can be directly used as the stitched map lane lines.

[0109] Combination Figure 2 As shown, first, traverse the 5 map lane lines in high-precision road 1. For L11, find the starting point of L11. If L21 in high-precision road 2 contains a point that coincides with (or is very close to) the starting point of L11, then L11 and L21 are considered to be connected. Then, L11 and L21 are stitched together (for elements with a connection relationship in a high-precision map, there will generally be a pair of overlapping points in their geometric shapes). The stitching method for L12 to L15 is similar and will not be repeated here. Specifically, the starting points of L14 and L15 in high-precision road 1 coincide with the ending point of L24 in high-precision road 2. Therefore, L14 is stitched together with L24, and L15 is stitched together with L24.

[0110] As an improvement to the above scheme, the least squares method is used to perform curve fitting on each lane line of the stitched map to obtain several fitted lane lines, specifically including:

[0111] Based on the coordinate data of the map geometric points on each map lane line after stitching, the least squares method is used to perform curve fitting on each map lane line after stitching to obtain several fitted lane lines.

[0112] Each fitted lane line is a cubic curve, and the cubic curve coefficients a0, a1, a2, and a3 of each fitted lane line are expressed by the formula... Calculated, (r) front r right This represents the coordinate data of map geometric points on the map lane lines in the map's vehicle coordinate system.

[0113] Specifically, in conjunction with the above embodiments, when actually performing curve fitting on map lane lines, the least squares method is used to perform curve fitting on each map lane line after stitching. The curve equation of the fitted lane line is calculated to be a cubic curve, and the cubic curve coefficients a0, a1, a2, and a3 of the curve equation of each fitted lane line can be calculated using the following formula:

[0114]

[0115] Among them, (r front r right ) represents the coordinate data of map geometric points on the map lane line corresponding to the fitted lane line in the map's vehicle coordinate system. As can be seen from the above examples, r front Let r be the forward coordinate in the vehicle's coordinate system on the map. right The coordinates are the rightward coordinates in the map's vehicle coordinate system.

[0116] This invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the vehicle positioning correction method described in any of the above embodiments.

[0117] This invention also provides a terminal device, see [link to relevant documentation]. Figure 3 The diagram shown is a structural block diagram of a preferred embodiment of a terminal device provided by the present invention. The terminal device includes a processor 10, a memory 20, and a computer program stored in the memory 20 and configured to be executed by the processor 10. When the processor 10 executes the computer program, it implements the vehicle positioning correction method described in any of the above embodiments.

[0118] Preferably, the computer program can be divided into one or more modules / units (such as computer program 1, computer program 2, ...), and the one or more modules / units are stored in the memory 20 and executed by the processor 10 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0119] The processor 10 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor 10 may be any conventional processor. The processor 10 is the control center of the terminal device, connecting various parts of the terminal device through various interfaces and lines.

[0120] The memory 20 mainly includes a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc., while the data storage area can store related data, etc. Furthermore, the memory 20 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart media card (SMC), a secure digital card (SD), and a flash card, or other volatile solid-state storage devices.

[0121] It should be noted that the aforementioned terminal devices may include, but are not limited to, processors and memory, as will be understood by those skilled in the art. Figure 3 The structural block diagram is merely an example of the terminal device described above and does not constitute a limitation on the terminal device. It may include more or fewer components than shown in the diagram, or combine certain components, or use different components.

[0122] In summary, the vehicle positioning correction method, computer-readable storage medium, and terminal device provided by this invention utilize the arrangement and combination of map lane lines and perceived lane lines in a high-precision map to find the optimal matching method between map lane lines and perceived lane lines, thereby obtaining the corresponding relationship between map lane lines and perceived lane lines. By combining the coordinate data of map geometric points on the map lane lines in the map's vehicle coordinate system and the curve coefficients of the curve equation of the perceived lane lines, the positioning coordinates are optimized and corrected. This can overcome the problem of accumulated errors that inevitably occur when the GNSS / INS integrated navigation system is in a GNSS denial condition for a long time, correct the positioning coordinates of the GNSS / INS integrated navigation system, and improve the positioning accuracy of the vehicle by correcting the error of the vehicle's positioning coordinates.

[0123] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A vehicle positioning correction method, characterized in that, include: Several fitted lane lines are obtained based on several map lane lines within a preset range of the vehicle's current location point in the high-precision map, including: Extract several map lane lines within a preset range of the vehicle's current location in the high-precision map; Convert the latitude and longitude coordinates of the map geometric points on the several map lane lines into coordinate data in the map vehicle coordinate system; The map lane lines are stitched together based on the coordinate data of the map geometric points, and curve fitting is performed on each stitched map lane line to obtain several fitted lane lines, including: Based on the coordinate data of map geometric points on the plurality of map lane lines, and combined with the preceding and succeeding relationships of lanes in the high-precision map, the plurality of map lane lines are stitched together to obtain a plurality of stitched map lane lines, including: Based on the coordinate data of the map geometric points on the several map lane lines, and combined with the preceding and succeeding relationships of lanes in the high-precision map, it is determined whether there is a connection relationship between the several map lane lines. The map lane lines that are connected are stitched together to obtain several stitched map lane lines. Map lane lines that are not connected to other map lane lines are directly used as the stitched map lane lines. The least squares method was used to perform curve fitting on each lane line of the stitched map to obtain several fitted lane lines. The several perceived lane lines collected at the current positioning point are synchronized with the several fitted lane lines in time. The fitted lane lines and the time-synchronized sensing lane lines are arranged and combined; wherein each combination includes at least one pair of lane lines, and each pair of lane lines includes one fitted lane line and one corresponding sensing lane line. Based on the lane matching error corresponding to each pair of lane lines in each combination, the total matching error corresponding to each combination is obtained, and the combination with the smallest total matching error is taken as the optimal combination. The step of obtaining the total matching error for each combination method based on the lane line matching error corresponding to each pair of lane lines in each combination method specifically includes: For each combination method, calculate the lane matching error between the fitted lane line and the perceived lane line in each lane line pair in the same combination method, and take the sum of the lane matching errors of all lane line pairs in the same combination method as the total matching error corresponding to the same combination method. Specifically, for each lane pair, the lane matching error between the fitted lane line and the perceived lane line in the same lane pair is calculated using the following formula. error : ； in, n This represents the number of map geometry points along the fitted lane lines. x i_hdmap , y i_hdmap ) indicates the first i The coordinate data of a map geometric point in the map's vehicle coordinate system. c 0、 c 1. c 2 and c 3 indicates the cubic curve coefficient of the perceived lane line; Based on the coordinate data of map geometric points on the fitted lane lines and the curve coefficients of the perceived lane lines in each lane line pair of the optimal combination method, the positioning coordinates of the current positioning point are corrected, specifically including: The coordinate data of the map geometric points on the fitted lane lines and the curve coefficients of the perceived lane lines in each lane line pair of the optimal combination are substituted into the optimization function to calculate the lateral error of the vehicle. δd and heading error δθ ; According to the lateral error δd and the heading error δθ Error compensation is performed on the positioning coordinates of the current positioning point to obtain the corrected positioning coordinates; The optimization function is: ； in,( x i_hdmap , y i_hdmap ) represents the first position on the fitted lane line. i The coordinate data of a map geometric point in the map's vehicle coordinate system. c 0、 c 1. c 2 and c 3 represents the cubic curve coefficient of the perceived lane line.

2. The vehicle positioning correction method as described in claim 1, characterized in that, The step of synchronizing the time between the several perceived lane lines collected at the current positioning point and the several fitted lane lines specifically includes: Acquire several lane lines sensed by the vehicle within the preset range; According to the vehicle in t hdmap arrive t perception The distance traveled within a time period determines the translation distance; among which, t hdmap This indicates the timestamp used to extract the lane lines from the map. t perception Indicates the timestamp of the collected lane line data; Based on the translation distance, the curve equations of the plurality of sensing lane lines are derived from... t perception The perception of time involves translating and transforming the vehicle's coordinate system. t hdmap The map uses the vehicle's coordinate system at any given time to obtain several perceived lane lines after time synchronization.

3. The vehicle positioning correction method as described in claim 1, characterized in that, The least squares method is used to perform curve fitting on each lane line of the stitched map to obtain several fitted lane lines, specifically including: Based on the coordinate data of the map geometric points on each map lane line after stitching, the least squares method is used to perform curve fitting on each map lane line after stitching to obtain several fitted lane lines. Each fitted lane line is a cubic curve, and the cubic curve coefficients of each fitted lane line are... a 0、 a 1. a 2 and a 3. Through formula Calculated, ( r front , r right () represents the coordinate data of map geometric points on the map lane lines in the map's vehicle coordinate system.

4. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program; wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the vehicle positioning correction method as described in any one of claims 1 to 3.

5. A terminal device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the vehicle positioning correction method as described in any one of claims 1 to 3.

Images