Road geometric parameter fitting method based on vehicle track

Abstract

The invention discloses a road geometric parameter fitting method based on a vehicle track, which comprises the following steps of: acquiring geographic space information of a vehicle driving track, and converting the geographic space information of the vehicle driving track into a local projection coordinate system of a road; acquiring a tangential course angle of the vehicle based on the vehicle track information in the projection coordinates; according to the characteristics of the tangential course angle on different road line elements, establishing the relationship between the tangential course angle and the travel mileage on different road line elements, and obtaining a parameter expression of a tangential course angle equation; based on the parameter expression of the tangential course angle equation, performing optimization fitting on the demarcation points of the road line elements, and determining the demarcation points of different line elements of the road; and performing parameter compliance check on the road line elements to obtain road parameters. The method can reduce the labor cost, improves the calculation and analysis precision, helps a road management and maintenance department to obtain in-service road alignment information, and provides a theoretical basis for road reconstruction and extension and safety evaluation.

Description

Technical Field

[0001] The invention belongs to the field of road safety evaluation and reconstruction and expansion, and in particular relates to a road geometric parameter fitting method based on vehicle trajectory. Background Art

[0002] Road alignment geometry is an important parameter for road design. With the economic and social development of my country, existing roads are facing a large number of upgrading, reconstruction and expansion tasks. In the process of road reconstruction and expansion, it is necessary to accurately obtain the geometric alignment parameters of existing roads. At the same time, road alignment geometry is also an important parameter for road safety evaluation. Local area corrections during road construction lead to certain differences between road design documents and actual alignment parameters. At present, the determination of existing road alignment parameters is mostly based on manual measurement, satellite image recognition and other methods, but these methods are cumbersome and labor-intensive, and are easily affected by image accuracy. In the current research on vehicle trajectory characteristics, more attention is paid to the study of vehicle driving characteristics on road alignment design and traffic safety evaluation. The vehicle trajectory is used to obtain the road alignment geometry parameters. Generally, an association model between parameters such as vehicle trajectory curvature and road alignment geometry parameters is established to determine the road geometry parameters. The current method has the following shortcomings:

[0003] (1) Traditional manual measurement methods are susceptible to manual reading errors and require a lot of manual labor, resulting in low work efficiency;

[0004] (2) When using image recognition algorithms to calculate road linear parameters, lighting, stains, resolution, etc. will lead to poor image quality, affecting the accuracy of the results. At the same time, high-precision images have high requirements for equipment and increase costs.

[0005] (3) Currently, driving trajectories are more used in the study of driving characteristics. In the acquisition of road linear parameters, parameters such as trajectory curvature are used, which are easily affected by measurement errors and affect the analysis accuracy.

[0006] Therefore, a road geometry parameter fitting method based on vehicle trajectory is urgently needed. Summary of the invention

[0007] In order to solve the above technical problems, the present invention proposes a road geometry parameter fitting method based on vehicle trajectory, which can reduce labor costs while improving the accuracy of calculation and analysis, helping road maintenance departments to obtain the linear information of in-service roads, and providing a theoretical basis for road reconstruction and safety evaluation.

[0008] To achieve the above object, the present invention provides a road geometry parameter fitting method based on vehicle trajectory, comprising:

[0009] Acquire vehicle driving track geospatial information, and convert the vehicle driving track geospatial information into a local projection coordinate system;

[0010] Based on the trajectory data in the projection coordinates, obtaining a tangential heading angle of the vehicle;

[0011] According to the characteristics of the tangential heading angle at different road line elements, a relationship model between the tangential heading angle and the mileage on the road line element is established to obtain a parameter model of the tangential heading angle equation;

[0012] Based on the parameter model of the tangent heading angle equation, the road line element demarcation points are optimized and fitted to obtain the demarcation points of different road line elements;

[0013] Perform a parameter compliance check on the road line element to obtain road parameters.

[0014] Optionally, based on the projection coordinates, obtaining the tangential heading angle of the vehicle includes:

[0015] Based on the trajectory point of the vehicle in the projection coordinates; calculating the center coordinates of the trajectory point and the curvature at the trajectory point;

[0016] The tangential heading angle is obtained based on the circle center coordinates and the curvature.

[0017] Optionally, the road line elements include: straight lines, circular curves and transition curves.

[0018] Optionally, the characteristics of the tangential heading angle at different road line elements include:

[0019] In the straight line, the tangential heading angle of the vehicle is the same as the direction of travel of the vehicle;

[0020] In the circular curve, the tangential heading angle changes linearly with the increase of the travel mileage;

[0021] In the relaxation curve, the tangential heading angle presents a nonlinear increasing trend as the mileage increases.

[0022] Optionally, establishing the relationship between the tangent heading angle and the mileage on the road plane alignment includes:

[0023] If the road plane line shape is a straight line, the relationship model between the tangent heading angle and the travel mileage on the straight line is:

[0024]

[0025] If the road plane line shape is a transition curve, the relationship model between the tangent heading angle on the transition curve and the travel mileage is:

[0026]

[0027] If the road plane line shape is a circular curve, the relationship between the tangent heading angle on the circular curve and the travel mileage is:

[0028] θ Ci =b i+2 l+c i+2

[0029] in, is the tangent heading angle on the straight line, is the tangent heading angle on the transition curve, θ Ci is the tangent heading angle on the circular curve, l is the mileage, and a, b, and c are parameters.

[0030] Optionally, the parameter model for obtaining the tangential heading angle equation includes:

[0031] Set the constraint condition of the tangential heading angle at the boundary of the basic line element, and use the constraint condition to obtain the parameter model.

[0032] Optionally, based on the parameter model of the tangent heading angle equation, optimizing and fitting the road line element demarcation points to obtain the demarcation points of different road line elements includes:

[0033] Obtaining the basic linear shape of the road, and combining the basic line elements to obtain a number of line element boundary points;

[0034] Using the parameter expression, obtaining an error value;

[0035] The line element demarcation points are optimized using the error values ​​to obtain different line element demarcation points of the road.

[0036] Optionally, optimizing the line element demarcation point by using the error value to obtain the road line element parameter includes:

[0037] S1, moving the first line element demarcation point between the starting point and the second line element demarcation point, each time moving a preset distance, obtaining an error value based on the preset distance, and determining the first line element demarcation point when the error value is the smallest;

[0038] S2, moving the second line element demarcation point between the first line element demarcation point and the third line element demarcation point by a preset distance each time, obtaining an error value based on the preset distance, and determining the second line element demarcation point when the error value is the smallest;

[0039] S3, moving the nth line element demarcation point between the n-1th line element demarcation point and the end point, each time by a preset distance, obtaining an error value based on the preset distance, and determining the nth line element demarcation point when the error value is the smallest;

[0040] S4, updating the line element demarcation point, determining the line element demarcation point corresponding to the maximum error value, re-dividing the line elements adjacent to the line element demarcation point, and obtaining a plurality of line element demarcation points;

[0041] S5. Repeat steps S1-S4 until the maximum error value meets the preset value, completing the line element demarcation point optimization and obtaining different line element demarcation points of the road.

[0042] Optionally, a parameter compliance check is performed on the road line element, and the parameter compliance check includes:

[0043] If the road line element does not meet the standard value, the basic line element is eliminated to obtain a new road line element;

[0044] The new road line element dividing point is re-optimized and fitted to obtain the road parameters. Compared with the prior art, the present invention has the following advantages and technical effects:

[0045] The present invention uses vehicle GPS data to obtain road linear geometry parameters, which does not require a lot of manual work, and the tangential heading angle measurement results are relatively stable and less affected by measurement errors. While reducing labor costs, it can improve the accuracy of calculation and analysis, help road maintenance departments obtain in-service road linear information, and provide a theoretical basis for road reconstruction and safety evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0047] Figure 1 is a flow chart of a road geometry parameter fitting method based on vehicle trajectory according to an embodiment of the present invention;

[0048] Figure 2 A schematic diagram of a circle center calculation method according to an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of an experimental road section of an embodiment of the present invention;

[0050] Figure 4 It is the original data display of the embodiment of the present invention;

[0051] Figure 5 This is a heading angle data diagram of an embodiment of the present invention;

[0052] Figure 6 This is a curvature data diagram of an embodiment of the present invention;

[0053] Figure 7 This is a comparison diagram of heading angle data and azimuth angle according to an embodiment of the present invention;

[0054] Figure 8 This is a diagram showing the iterative effect of an embodiment of the present invention;

[0055] Fig. 9 Schematic diagram of the final segmentation result of an embodiment of the present invention. DETAILED DESCRIPTION

[0056] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0057] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0058] The present invention proposes a road geometry parameter fitting method based on vehicle trajectory. For two-lane roads, the latitude and longitude information of the vehicle position during driving is collected. After coordinate system conversion, the driving heading angle information is calculated using the adjacent position information, and the tangential heading angle and driving mileage are used as a bridge to obtain the road linear geometry parameters. Figure 1 As shown, the specific steps include:

[0059] Step 1: Obtain the vehicle driving trajectory geographic information and convert the vehicle driving trajectory geographic information into a local projection coordinate system.

[0060] Specifically, in order to more accurately process and analyze vehicle driving trajectory data, the present invention first performs projection coordinate conversion on the vehicle driving trajectory geographic information. Specifically, the geographic coordinate system coordinates (such as WGS 84, EPSG:4326) collected by the global positioning system (GPS) are converted into a local projection coordinate system suitable for calculation. The conversion method uses the "pyproj" function package in the Python language, and the function call method is:

[0061] p1=pyproj.Proj(init='epsg:old_epsg_value')

[0062] p2=pyproj.Proj(init='epsg:new_epsg_value')

[0063] new_lon,new_lat=pyproj.transform(p1,p2,old_lon,old_lat)

[0064] Among them, p1 is the measurement reference coordinate system, p2 is the coordinate system used for data analysis, new_lon, new_lat, old_lon, old_lat are the latitude and longitude of the analysis reference coordinate system and the measurement reference coordinate system respectively.

[0065] Step 2: Based on the vehicle trajectory information in the projection coordinates, obtain the vehicle's tangential heading angle;

[0066] Further, based on the projection coordinates, obtaining the tangential heading angle of the vehicle includes:

[0067] Based on the projection coordinate system, obtain the trajectory points of the vehicle;

[0068] Calculate the coordinates of the center of the trajectory point and the curvature at the trajectory point;

[0069] Based on the center coordinates and curvature of the circle, get the tangential heading angle.

[0070] Specifically, after obtaining the vehicle driving trajectory points in the calculation reference coordinate system, the present invention fits a circle based on three consecutive data points to calculate key indicators such as the tangential heading angle and curvature of the vehicle at each point. The specific method is as follows:

[0071] (1) Data accuracy adjustment. According to the data analysis accuracy requirements, the trajectory resolution is determined according to the driving speed and the sampling frequency. The trajectory information that meets the accuracy is averaged and aggregated to eliminate random errors and improve the analysis accuracy.

[0072] (2) Curvature and tangent heading angle traversal solution: In this step, the curvature and tangent heading angle at each trajectory point are calculated in chronological order according to the measured data sequence. Assuming that three adjacent trajectory measurement points are located on a circular curve, due to the geometric properties of the circle, the perpendicular bisector of any chord will pass through the center of the circle. Figure 2 In the equation, the perpendicular bisectors of the line segments ab and bc formed by the adjacent trajectory points intersect at point M, which is the center of the circle. The calculation method of the center coordinates, the curvature γ at the trajectory point and the tangent heading angle θ is as follows:

[0073] Calculation of the coordinates of the center M (XM, YM): The coordinates of the center M can be solved by analytical geometry, that is, by solving the equations of the two perpendicular bisectors simultaneously. In the following formula, the subscripts p and q represent the midpoints of ab and bc respectively.

[0074]

[0075] Calculation of the curvature γ of point b: After obtaining the coordinates of the center M, the distance between the two points is calculated based on the coordinates of the center and point b, which is the radius of the curve. The reciprocal of the radius is taken to obtain the curvature at point b. The curvature γ is the reciprocal of the radius r (γ = 1 / r), which reflects the degree of curvature of the curve at this point. It should be noted that when the three points are almost collinear, the radius r will tend to infinity. At this time, the curvature γ defaults to 0, indicating that the curve near this point is close to a straight line. Once the coordinates of the center M are estimated, this embodiment can use the distance from the center to point b (i.e., the radius r) to calculate the curvature γ at point b.

[0076]

[0077] Calculation of the tangent heading angle θ: The present invention defines the angle of the vehicle in the tangent direction of the current trajectory point as the tangent heading angle. Based on the coordinates of the center point of the circle, the radius vector between any trajectory point and the center of the circle can be calculated. Based on the geometric properties of the circular curve, the tangent at the trajectory point is perpendicular to the radius, and the magnitude of the tangent heading angle is the radius azimuth ±90°, and its direction is determined according to the direction of vehicle travel. The present invention uses the tangent heading angle as the driving heading angle, which can more accurately characterize the actual direction of travel of the vehicle compared to the traditional azimuth method, thereby improving the analysis accuracy.

[0078] θ=[arctan2(x M -x b ,y M -y b )±90]mod360 (5)

[0079] When the three points are almost collinear, the traditional tangent direction calculation may no longer be applicable, and it is more reasonable to directly use the direction of the line connecting adjacent points as the approximate driving direction. Therefore, in the present invention, the driving direction θ is directly set to the direction of ac (or the direction of bc, depending on the definition and context).

[0080] The driving direction θ is the tangent direction of the vehicle at point b. First, the direction of the vector Mb from the center of the circle M to point b needs to be determined. Then, according to the properties of the circle, the tangent is perpendicular to the radius at the tangent point, so the tangent direction can be obtained by rotating the Mb direction 90 degrees. The choice of rotation direction is based on the actual path of the vehicle: since the vehicle's driving direction is usually closer to the direction of bc, this embodiment selects a rotation method that makes the tangent direction closer to the bc direction (+90 degrees or -90 degrees). Use the vector dot product or cross product to determine which rotation direction makes the angle between the tangent direction and the bc direction smaller.

[0081] This method uses the tangent direction when calculating the heading angle. Compared with the direction of the line connecting the front and rear points (i.e., the azimuth) used in the traditional method, it can more accurately reflect the actual direction of the vehicle's travel, thereby improving the accuracy and reliability of the identification of road linear geometric parameters.

[0082] Step 3: Based on the vehicle's tangential heading angle, obtain the parameter expression of the tangential heading angle and mileage.

[0083] Furthermore, the road plane alignment includes: straight lines, circular curves and transition curves;

[0084] In the straight line segment, the vehicle's tangential heading angle is the same as the vehicle's travel direction;

[0085] In the circular curve segment, the tangential heading angle changes linearly with the increase of mileage;

[0086] In the transition curve section, the tangent heading angle shows a nonlinear increasing trend with the increase of mileage.

[0087] Specifically, the basic elements of road planar alignment include straight lines, circular curves, and transition curves. The vehicle trajectory is the basis of road alignment design, and the vehicle trajectory has different characteristics in different line elements. In a straight line segment, the radius is infinite, and the vehicle's tangent heading angle is the same as the vehicle's driving direction, that is, the direction of the line connecting the two points; on a circular curve, the tangent heading angle changes linearly with the increase in mileage; in a transition curve segment, the tangent heading angle shows a nonlinear increasing trend with the increase in mileage. The following describes the relationship between the tangent heading angle θ and the mileage l on a straight line, circular curve, and transition curve.

[0088] Straight line: When driving on a straight line, the tangential heading angle θ remains constant, that is, θ=c, where c is a constant.

[0089] Easing curve: Easing curve is often described by geometric clothoid, and its characteristics can be expressed by rl = A 2 Description, where r is the radius of curvature, l is the distance traveled, and A is the clothoid parameter. In differential geometry, the curvature γ is related to the rate of change of the azimuth angle ɑ, that is, (ɑ represents the azimuth angle, to avoid confusion with the tangential heading angle θ). Substituting into the above formula, we get: Solving this differential equation yields: If the heading angle at the starting point is known (assuming θ0), and the tangent direction angle at the starting point is also ɑ0=0. (That is, the curve coincides with a fixed direction at the starting point), then ɑ and θ can be considered the same (in a two-dimensional plane and when there is no sudden change in direction). Therefore, the relationship between the tangent heading angle θ and the mileage l can be expressed as:

[0090] θ=al 2 +c (6)

[0091] Circular curve: On a circular curve, the curvature γ is a constant, and the change in heading angle is proportional to the mileage, which can be expressed as

[0092] θ=bl+c (7)

[0093] For the basic road line shape (straight line 1-transition curve segment 1-circular curve segment-transition curve segment 2-straight line segment 2), where the transition curve is represented by a clothoid, there is a relationship model between the cumulative tangential heading angle θ(l) and the mileage l:

[0094] Line segment 1:

[0095]

[0096] Transition curve segment 1:

[0097]

[0098] Circular curve:

[0099] θ C =b3l+c3 (10)

[0100] Transition curve segment 2:

[0101]

[0102] Line segment 2:

[0103]

[0104] Furthermore, the method for establishing the relationship between the tangential heading angle and the mileage on the basic element is:

[0105] In a straight line segment, the relationship between the tangent heading angle and the mileage is:

[0106]

[0107] In the transition curve section, the relationship between the tangent heading angle and the mileage is:

[0108]

[0109] In the circular curve segment, the relationship between the tangent heading angle and the mileage is:

[0110] θ Ci =b i+2 l+c i+2 (15)

[0111] in, is the tangent heading angle in the straight line segment, is the tangential heading angle in the transition curve segment, θ Ci is the tangential heading angle in the circular curve segment, l is the mileage, and a, b, and c are parameters.

[0112] Furthermore, the parameter expression for obtaining the tangential heading angle equation includes:

[0113] Set the constraint conditions of the tangent heading angle at the boundary of the basic line element, and use the constraint conditions to obtain the parameter expression.

[0114] Specifically, based on the continuity characteristics of the vehicle's driving trajectory, the vehicle trajectory and tangential heading angle should satisfy the following two continuity conditions: (a) the vehicle's cumulative tangential heading angle is equal at different line element boundaries; (b) the first-order derivative of the vehicle trajectory is continuous. Therefore, the vehicle's tangential heading angle equation should satisfy the following equation at different line element boundaries:

[0115]

[0116] Based on the continuity equation, we can get the analytical expressions of the parameters of the tangent heading angle equation corresponding to different line elements. For a single basic line shape, the expressions of each parameter are as follows:

[0117]

[0118] Among them, l1 to l4 are the dividing points of straight line segment 1-transition curve segment 1, transition curve segment 1-circular curve, circular curve-transition curve segment 2, and transition curve segment 2-straight line segment 2 in the basic linear combination.

[0119] For a continuous route, assuming that the route is composed of n basic linear segments, according to the continuity condition of the cumulative tangential heading angle, the analytical expression of the equation parameters can be obtained as shown below:

[0120]

[0121] (4) Calculate the error function:

[0122]

[0123] Among them, i is the number of original data points, θ n (l) is the predicted value of the tangential heading angle at each point of the fitting curve, θ original Calculate the initial tangential heading angle.

[0124] Step 4: Based on the parameter expression of the tangent heading angle equation, optimize the fitting of the road line element dividing points.

[0125] Specifically, obtaining road line elements includes:

[0126] Obtain the basic linear shape of the road, combine the basic line elements, and obtain several line element boundary points;

[0127] Use parameter expressions to obtain error values;

[0128] The error value is used to optimize the line element dividing point to obtain the road line element.

[0129] Furthermore, the line element demarcation point is optimized using the error value, and the road line element is obtained including:

[0130] S1, moving the first line element demarcation point between the starting point and the second line element demarcation point, each time moving a preset distance, obtaining an error value based on the preset distance, and determining the first line element demarcation point when the error value is the smallest;

[0131] S2, moving the second line element demarcation point between the first line element demarcation point and the third line element demarcation point, each time moving a preset distance, obtaining an error value based on the preset distance, and determining the second line element demarcation point when the error value is the smallest;

[0132] S3, moving the nth line element demarcation point between the n-1th line element demarcation point and the end point, each time moving a preset distance, obtaining an error value based on the preset distance, and determining the nth line element demarcation point when the error value is the smallest;

[0133] S4, updating the line element demarcation point, determining the line element demarcation point corresponding to the maximum error value, re-dividing the line elements adjacent to the line element demarcation point, and obtaining a number of line element demarcation points;

[0134] S5. Repeat steps S1-S4 until the maximum error value meets the preset value, completing the line element demarcation point optimization and obtaining the road line element.

[0135] Specifically, (1) Assuming that the route is basically linear, there are four line element demarcation points: l1, l2, l3, l4

[0136] (2) Randomly assign initial values ​​to them (using equally divided points) to calculate the values ​​of a2, c2, b3, c3, a4, c4, and calculate the total error E.

[0137] (3) Iterative optimization of line element demarcation points:

[0138] ① Let l1 move between the starting point and l2. According to the distance between the starting point and l2, the iterative step length Δx is determined by the equal division method, and the optimal position of l1 is determined based on the principle of minimum total error;

[0139] ② Let l2 move between l1 and l3, moving a distance of Δx each time, and find the value of l2 when the total error is minimized.

[0140] ③ Let l3 move between l2 and l4, moving a distance of Δx each time, and find the value of l3 when the total error is minimized.

[0141] ④ Let l4 move between l3 and the end point, moving a distance of Δx each time, and find the value of l4 when the total error is minimized.

[0142] ⑤Update the values ​​of l1, l2, l3, l4, and repeat steps ①-④ until the values ​​of l1, l2, l3, l4 are no longer updated.

[0143] ⑥ Find the single maximum error point e ln =θ n (l)-θ original , add 4 segmentation points near it (for example, l n -10,l n -5,l n +5,l n +10).

[0144] ⑦ Update l1-l8 in sequence. The position close to the starting point is l1, and the position close to the end point is l8. Repeat steps ①-⑤ to update the positions of l1-l8 in sequence.

[0145] ⑧Repeat ⑥-⑦ until e ln Less than a certain value (adjusted according to the precision of the data, such as 1, 2, 3, etc.).

[0146] Step 5: Normalize the parameters of road line elements.

[0147] Obtaining road parameters includes:

[0148] If the road line element does not meet the standard value, the basic line element is eliminated to obtain a new road line element;

[0149] Re-optimize and fit the new road line element dividing points to obtain road parameters.

[0150] Specifically, according to the route design specifications, the shortest length of the straight segment, curve segment and transition curve segment in the road alignment is determined as the control index, and the length of each line element is calculated. When the line element length is less than the standard value, the adjacent line elements of the point are merged. For the group with too short transition curves, the transition curve is deleted, and the line shape becomes a simple line shape combination of straight segment 1-circular curve-straight segment 2, and each group becomes 4 points. The corresponding formula becomes:

[0151] Line segment 1:

[0152]

[0153] Circular curve:

[0154] θ C =b2 l+c2 (33)

[0155] Line segment 2:

[0156]

[0157] The corresponding analytical expression is:

[0158]

[0159] Recalculate the position of each segmentation point according to the previous method of iterative optimization of line element dividing points.

[0160] Step 6: Calculate the parameters of each road section.

[0161] For straight line segments, the length of each segment is l n+1 -l n , slope and other parameter indicators.

[0162] For circular curves:

[0163] θ C =b2 l+c2 (37)

[0164] Where b is the curvature.

[0165] For spirals:

[0166]

[0167] Among them, the transition curve parameter A = 1 / 2a.

[0168] This embodiment was field tested on Liuxin Road, Bulaotun, a district in Beijing. Liuxin Road is a Class IV highway with two lanes in both directions, and the experimental section is from K53+340 to K57+240. The driving speed of the experimental vehicle was set to 40km / h, the accuracy of GPS data was about 10cm, and the sampling frequency was as high as 100Hz. In order to ensure high accuracy and high reliability of the data, a highly sensitive GPS receiver and acceleration sensor module were deployed on the vehicle, combined with a 4G data transmission module. Such high-frequency data acquisition significantly improves the accuracy and reliability of subsequent data processing, and provides a solid foundation for the precise reconstruction of subsequent road linear parameters.

[0169] Figure 3 The data visualization results after road section collection are shown. Figure 4 The collected raw data is displayed. The data used for analysis in this embodiment include key parameters such as longitude and latitude, speed and time.

[0170] Optimization of heading angle calculation: Based on the GPS point coordinate sequence, the optimization algorithm proposed in this embodiment is used to calculate the heading angle. The experimental results show that the algorithm effectively reduces noise interference during the processing and smoothes the changes in the heading angle, so that the calculation results more accurately reflect the actual driving direction of the vehicle. Compared with traditional methods, this optimization algorithm shows significant advantages in the calculation of heading angle.

[0171] Figure 5 and Figure 6The comparison between the heading angle data and the curvature data used in this embodiment is shown, and the results show that the noise in the heading angle data is significantly lower than that in the curvature data, and has higher reliability. Figure 7 The comparative analysis of the tangential heading angle used in this embodiment and the traditional azimuth angle is shown. The results show that the tangential heading angle is more advantageous in reflecting the real driving direction of the vehicle, with less noise and higher accuracy.

[0172] Road alignment data generation: After obtaining a continuous and accurate heading angle sequence, this embodiment converts the heading angle data into a geometric description of the road alignment through the heuristic iteration method proposed in this article, including parameter generation of straight segments, arc segments, and complex curve segments. This method makes the geometric description of the road alignment more accurate and detailed through iterative optimization. Figure 8 The effect comparison of different numbers of iterations is shown. Fig. 9 The final calculated piecewise fitting results are shown.

[0173] Data storage and sharing: The processed road alignment data is stored in a unified standard format to facilitate subsequent data analysis, visualization, and sharing. This standardized data storage method not only improves the availability of data, but also facilitates multi-party collaboration and data sharing. Table 1 shows the results of the fitting, indicating that this method can produce good fitting results in most cases, providing a reliable data basis for further research and application.

[0174] Table 1

[0175]

[0176] Through the implementation of the above optimization measures, this embodiment can efficiently collect and process vehicle driving data, generate high-precision road linear data, and provide strong support for traffic planning, autonomous driving, intelligent navigation and other fields.

[0177] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A road geometry parameter fitting method based on vehicle trajectory, characterized in that: include: Acquire vehicle driving track geospatial information, and convert the vehicle driving track geospatial information into a local projection coordinate system; Based on the trajectory data in the projection coordinates, obtaining a tangential heading angle of the vehicle; According to the characteristics of the tangential heading angle at different road line elements, a relationship model between the tangential heading angle and the mileage on the road line element is established to obtain a parameter model of the tangential heading angle equation; Based on the parameter model of the tangent heading angle equation, the road line element demarcation points are optimized and fitted to obtain the demarcation points of different road line elements; Perform a parameter compliance check on the road line element to obtain road parameters.

2. A road geometry parameter fitting method based on vehicle trajectory according to claim 1, characterized in that: Based on the projection coordinates, obtaining the tangential heading angle of the vehicle includes: Based on the trajectory point of the vehicle in the projection coordinates; calculating the center coordinates of the trajectory point and the curvature at the trajectory point; The tangential heading angle is obtained based on the circle center coordinates and the curvature.

3. A road geometry parameter fitting method based on vehicle trajectory according to claim 2, characterized in that: The road line elements include: straight lines, circular curves and transition curves.

4. A road geometry parameter fitting method based on vehicle trajectory according to claim 3, characterized in that: The characteristics of the tangential heading angle at different road line elements include: In the straight line, the tangential heading angle of the vehicle is the same as the direction of travel of the vehicle; In the circular curve, the tangential heading angle changes linearly with the increase of the travel mileage; In the relaxation curve, the tangential heading angle presents a nonlinear increasing trend as the mileage increases.

5. A road geometry parameter fitting method based on vehicle trajectory according to claim 4, characterized in that: Establishing the relationship between the tangent heading angle and the traveled mileage on the road plane alignment includes: If the road plane line shape is a straight line, the relationship model between the tangent heading angle and the travel mileage on the straight line is: If the road plane line shape is a transition curve, the relationship model between the tangent heading angle on the transition curve and the travel mileage is: If the road plane line shape is a circular curve, the relationship between the tangent heading angle on the circular curve and the travel mileage is: i Ci =b i+2 l+c i+2 in, is the tangent heading angle on the straight line, is the tangent heading angle on the transition curve, θ Ci is the tangent heading angle on the circular curve, l is the mileage, and a, b, and c are parameters.

6. A road geometry parameter fitting method based on vehicle trajectory according to claim 5, characterized in that: The parameter model for obtaining the tangential heading angle equation includes: The constraint condition of the tangential heading angle at the boundary of the basic line element is set, and the parameter model is obtained by using the constraint condition.

7. A road geometry parameter fitting method based on vehicle trajectory according to claim 6, characterized in that: Based on the parameter model of the tangent heading angle equation, the road line element demarcation points are optimized and fitted to obtain the demarcation points of different road line elements, including: Obtaining the basic linear shape of the road, and combining the basic line elements to obtain a number of line element boundary points; Using the parameter model, obtaining an error value; The line element demarcation points are optimized using the error values ​​to obtain different line element demarcation points of the road.

8. The method for fitting road geometry parameters based on vehicle trajectory according to claim 7, characterized in that: Optimizing the line element demarcation point by using the error value to obtain the demarcation points of different line elements of the road includes: S1, moving the first line element demarcation point between the starting point and the second line element demarcation point, each time moving a preset distance, obtaining an error value based on the preset distance, and determining the first line element demarcation point when the error value is the smallest; S2, moving the second line element demarcation point between the first line element demarcation point and the third line element demarcation point by a preset distance each time, obtaining an error value based on the preset distance, and determining the second line element demarcation point when the error value is the smallest; S3, moving the nth line element demarcation point between the n-1th line element demarcation point and the end point, each time by a preset distance, obtaining an error value based on the preset distance, and determining the nth line element demarcation point when the error value is the smallest; S4, updating the line element demarcation point, determining the line element demarcation point corresponding to the maximum error value, re-dividing the line elements adjacent to the line element demarcation point, and obtaining a plurality of line element demarcation points; S5. Repeat steps S1-S4 until the maximum error value meets the preset value, completing the line element demarcation point optimization and obtaining different line element demarcation points of the road.

9. The method for fitting road geometry parameters based on vehicle trajectory according to claim 7, characterized in that: Performing a parameter compliance check on the road line element, the parameter compliance check comprising: If the road line element does not meet the standard value, the basic line element is eliminated to obtain a new road line element; The new road line element dividing point is re-optimized and fitted to obtain the road parameters.

Images