A windshield inspection optimization method that meets HUD accuracy requirements

Abstract

The present invention discloses a windshield glass detection optimization method that meets the HUD precision requirements, establishes a detection reference plane of the HUD windshield glass, determines the reference plane and reference point of the detection area, determines the detection area and detection point of the HUD windshield glass, measures the curvature of the detection area of ​​the HUD windshield glass, that is, measures the minimum curvature radius Rmin and the maximum curvature radius Rmax of each detection point, calculates and optimizes. The present invention can be analyzed after the first version of the modeling is produced, and the analysis cycle is short. After the analysis is confirmed to be OK, the glass surface data can be locked, thereby ensuring the locking nodes of the large surface data of the whole vehicle modeling.

Description

Technical Field

[0001] The invention belongs to the technical field of HUD windshield detection, and in particular relates to a windshield detection optimization method that meets HUD accuracy requirements. Background Art

[0002] As the sense of automobile technology improves, more and more cars are equipped with HUD, or head-up display system, which projects important driving information such as speed and navigation onto the windshield in front of the driver, so that the driver can see important driving information such as speed and navigation without lowering or turning his head. The emergence of HUD technology has higher and higher requirements for the surface shape of glass. If the curvature of the glass surface does not meet the requirements, HUD will have display problems such as imaging ghosting, rotation, and mutation.

[0003] The Chinese invention patent application (application publication number CN112595257A, application publication date 2021.4.2) discloses a windshield surface detection method for HUD display, modeling and simulation - building a test environment - image source preprocessing - image acquisition - image comparison - surface determination; the image captured by the image acquisition device is compared with the theoretical image, and the difference between the image reflected by the measured windshield and the theoretical image at the position viewed by the human eye is output, so as to determine whether the measured windshield can meet the HUD display requirements. Its essence is to determine whether the glass surface meets the HUD projection quality through image difference analysis in a simulation way. However, the detailed design of the HUD needs to be initially completed, and simulation analysis is carried out after detailed data and optical path data are available. The simulation modeling and analysis cycle is long, which makes the confirmation of the glass surface quality lag behind. Summary of the invention

[0004] The purpose of the present invention is to provide a windshield inspection optimization method that has a short analysis cycle, ensures that the glass surface quality is not delayed, and meets the HUD accuracy requirements in view of the above technical defects.

[0005] To achieve the above objectives, the windshield glass detection optimization method that meets the HUD accuracy requirements is specifically as follows:

[0006] 1) Establish the detection reference surface of HUD windshield;

[0007] 2) Determine the reference plane and reference point of the detection area on the detection reference plane;

[0008] 3) Determine the HUD windshield detection area and detection points:

[0009] Orthogonally project the reference plane of the detection area onto the windshield to obtain the HUD windshield detection area, and orthogonally project all reference points onto the windshield to obtain the detection points;

[0010] 4) Measure the curvature of the HUD windshield detection area:

[0011] Measure the minimum value Rmin and the maximum value Rmax of the radius of curvature of each detection point;

[0012] 5) Calculate and optimize

[0013] The obtained minimum curvature radius Rmin and maximum curvature radius Rmax are compared with the set values ​​and the curvature radius of the windshield is adjusted.

[0014] Furthermore, the specific process of comparing the obtained minimum curvature radius Rmin and the maximum curvature radius Rmax with the set value and adjusting the curvature radius of the windshield is as follows:

[0015] 5a) calculating the average value of the minimum value of the radius of curvature, that is, the average value Rmin; if the average value Rmin is ≥ the first set value, executing step 5c); if the average value Rmin is < the first set value, executing step 5b);

[0016] 5b) increasing the curvature radius of the windshield in the direction of the minimum curvature radius, and then jumping to step 4);

[0017] 5c) calculating the average value of the maximum value of the curvature radius, that is, the average value Rmax; if the average value Rmax is ≥ the second set value, executing step 5e); if the average value Rmax is < the second set value, executing step 5d);

[0018] 5d) increasing the curvature radius of the windshield in the direction of the maximum curvature radius, and then jumping to step 4);

[0019] 5e) If the average value of Rmax / the average value of Rmin=the third setting range, execute step 5h); if the average value of Rmax / the average value of Rmin<the lower limit of the third setting range, execute step 5f); if the average value of Rmax / the average value of Rmin>the upper limit of the third setting range, execute step 5g);

[0020] 5f) increasing the curvature radius of the windshield along the direction of the maximum curvature radius, and gradually increasing the curvature radius, and jumping to step 4) each time the curvature radius increases;

[0021] 5g) increasing the curvature radius of the windshield along the direction of the minimum curvature radius, and gradually increasing the curvature radius, and jumping to step 4) each time the curvature radius is increased;

[0022] 5h) calculating the absolute deviation rate of the maximum value of the radius of curvature, that is, the absolute deviation rate Rmax; if the absolute deviation rate Rmax is ≤ the fourth set value, executing step 5j); if the absolute deviation rate Rmax is > the fourth set value, executing step 5i);

[0023] 5i) comparing the maximum values ​​of the two boundary curvature radii in the HUD windshield detection area along the direction of the maximum value of the curvature radius, taking the smaller one of the boundary curvature radii as a reference, gradually increasing the curvature radius of the windshield along the direction of the maximum value of the curvature radius by 3 to 5% of the reference, and jumping to step 4) every time the curvature radius is increased;

[0024] 5j) calculating the absolute deviation rate of the minimum value of the radius of curvature, that is, the absolute deviation rate Rmin; if the absolute deviation rate Rmin is ≤ the fifth set value, the detection optimization is completed; if the absolute deviation rate Rmin is > the fifth set value, executing step 5k);

[0025] 5k) Compare the minimum values ​​of the two boundary curvature radii in the HUD windshield detection area along the direction of the minimum curvature radius, take the one with the smaller minimum curvature radius as the reference, gradually increase the curvature radius of the windshield along the direction of the minimum curvature radius by 3 to 5% of the reference, and jump to step 4) each time the curvature radius is increased.

[0026] Furthermore, the specific process of establishing the detection reference surface of the HUD windshield in step 1) is as follows:

[0027] 1a) The XZ plane where point R is located is the symmetry plane of the human eye, point R is the seat reference point, and X and Z are the X and Z axes of the vehicle coordinate system;

[0028] 1b) On the symmetry plane of the human eye, make a lower sight line through point V0. The angle between the lower sight line and the X-axis is the HUD design angle. Point V0 is the middle eye point between the upper eye point V1 and the lower eye point V2. The point where the lower sight line intersects with the windshield is point O. Make a tangent reference plane of the windshield through point O. The tangent reference plane is the detection reference plane.

[0029] Furthermore, the specific process of determining the detection area reference plane and reference point on the detection reference plane in step 2) is as follows:

[0030] Make a square on the detection reference plane, and then divide the square into n×n squares to determine the detection area reference plane, 3≤n≤8, and the center point of the square is point O, and one side of the square is perpendicular to the lower line of sight;

[0031] Take two reference points on each square to obtain a total of 2n 2 There are reference points, and the line connecting the two reference points on each square passes through the center point of the square. The lines connecting the two reference points on all squares are in the same direction. At the same time, the distance from each reference point to the two perpendicular sides of the square where the reference point is located is equal.

[0032] Furthermore, in the step 5b), the increased curvature radius = (first set value - Rmin average value) + 100 to 200 mm.

[0033] Furthermore, in the step 5d), the increased curvature radius = (the second set value - the average value of Rmax) + 100 to 200 mm.

[0034] Furthermore, in step 5f), the curvature radius is gradually increased according to the increase of the average value of Rmax×(3-5%), and the process jumps to step 4) every time the radius increases.

[0035] Furthermore, in step 5g), the radius of curvature is gradually increased according to the increase of the average value of Rmin×(3-5%), and the process jumps to step 4) after each increase.

[0036] Furthermore, the Rmax absolute deviation rate=Rmax absolute deviation value×100% / Rmax average value; Rmin absolute deviation rate=Rmin absolute deviation value×100% / Rmin average value;

[0037] Where: Rmax absolute deviation value = [(Rmax-1-Rmax average value) + (Rmax-2-Rmax average value) + ... + (Rmax-2n 2 -Rmax average value)] / 2n 2

[0038] Rmin absolute deviation value = [(Rmin-1-Rmin average value) + (Rmin-2-Rmin average value) + ... + (Rmin-2n 2 -Rmin average value)] / 2n 2 .

[0039] Compared with the prior art, the present invention has the following advantages: the present invention can be analyzed after the initial version of the modeling is completed, the analysis cycle is short, and the glass surface data can be locked after the analysis is confirmed to be OK, thereby ensuring the locking nodes of the large surface data of the entire vehicle modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram for determining the detection reference surface according to the present invention. DETAILED DESCRIPTION

[0041] The present invention is further described in detail below in conjunction with specific embodiments so that those skilled in the art can understand. First, two technical terms are briefly described as follows:

[0042] Point R: Seat reference point, which is determined by the design point specified for each seating position in vehicle manufacturing relative to the three-dimensional coordinate system.

[0043] Point V: In the passenger compartment, the longitudinal plumb plane passing through the center line of the front outer seating position is related to the position of point R. This point is used to check whether the vehicle's field of view meets the requirements.

[0044] The specific process of the windshield glass detection optimization method of the present invention that meets the HUD accuracy requirements is as follows:

[0045] 1) Establish the detection reference surface of HUD windshield

[0046] 1a) The XZ plane where point R is located is the symmetry plane of the human eye, point R is the seat reference point, and X and Z are the X and Z axes of the vehicle coordinate system;

[0047] 1b) On the symmetry plane of the human eye, make a lower sight line 1 through point V0. The angle between the lower sight line 1 and the X-axis is the HUD design angle. Point V0 is the middle eye point between the upper eye point V1 and the lower eye point V2. The point where the lower sight line intersects with the windshield 2 is point O. Make a tangent reference plane of the windshield 2 through point O. The tangent reference plane is the detection reference plane 3. Figure 1 As shown;

[0048] 2) Determine the reference plane and reference point of the detection area on the detection reference plane

[0049] Make a square (such as 30mm×30mm) on the detection reference plane, and then divide the square into n×n (such as 4×4) squares to determine the detection area reference plane, 3≤n≤8, and the center point of the square is point O, and one side of the square is perpendicular to the lower line of sight;

[0050] Take two reference points on each square to obtain a total of 2n 2 (e.g. 16) reference points, and the line connecting the two reference points on each square passes through the center of the square, and the lines connecting the two reference points on all squares are in the same direction. At the same time, the distance from each reference point to the two perpendicular sides of the square where the reference point is located is equal (e.g. 10mm);

[0051] 3) Determine the HUD windshield detection area and detection points

[0052] Orthogonally project the reference plane of the detection area onto the windshield to obtain the HUD windshield detection area, and orthogonally project all reference points onto the windshield to obtain 2n 2 Testing points;

[0053] 4) Measure the curvature of the HUD windshield detection area

[0054] Measure the minimum value Rmin and the maximum value Rmax of the radius of curvature of each detection point, that is, Rmin-1, Rmin-2, ..., Rmin-2n 2 , Rmax-1, Rmax-2,..., Rmax-2n 2 ;

[0055] 5) Calculate and optimize

[0056] 5a) Calculate the average value of the minimum value of the radius of curvature, that is, the average value Rmin. If the average value Rmin is ≥ the first set value 3000 mm, execute step 5c); if the average value Rmin is < the first set value 3000 mm, execute step 5b);

[0057] 5b) Increase the curvature radius of the windshield in the direction of the minimum curvature radius, and then jump to step 4); the increased curvature radius = (3000-Rmin average value) + 100-200mm;

[0058] 5c) calculating the average value of the maximum value of the radius of curvature, that is, the average value Rmax; if the average value Rmax is ≥ the second set value 6000 mm, executing step 5e); if the average value Rmax is < the second set value 6000 mm, executing step 5d);

[0059] 5d) Increase the curvature radius of the windshield along the direction of the maximum curvature radius, and then jump to step 4); the increased curvature radius = (6000-Rmax average value) + 100 to 200 mm;

[0060] 5e) If the average value of Rmax / the average value of Rmin=the third setting range 5, execute step 5h); if the average value of Rmax / the average value of Rmin<1.5, execute step 5f); if the average value of Rmax / the average value of Rmin>2.5, execute step 5g);

[0061] 5f) increasing the curvature radius of the windshield along the direction of the maximum curvature radius, gradually increasing the curvature radius by an increase of the average value of Rmax × (3-5%), and jumping to step 4) each time the curvature radius increases;

[0062] 5g) increasing the curvature radius of the windshield along the direction of the minimum curvature radius, gradually increasing the curvature radius by an increase of the average value of Rmin × (3-5%), and jumping to step 4) each time the curvature radius increases;

[0063] 5h) calculating the absolute deviation rate of the maximum value of the radius of curvature, that is, the absolute deviation rate Rmax; if the absolute deviation rate Rmax is ≤ the fourth set value 3%, executing step 5j); if the absolute deviation rate Rmax is > the fourth set value 3%, executing step 5i);

[0064] 5i) comparing the maximum values ​​of the two boundary curvature radii in the HUD windshield detection area along the direction of the maximum value of the curvature radius, taking the smaller one of the boundary curvature radii as a reference, gradually increasing the curvature radius of the windshield along the direction of the maximum value of the curvature radius by 3 to 5% of the reference, and jumping to step 4) every time the curvature radius is increased;

[0065] 5j) calculating the absolute deviation rate of the minimum value of the radius of curvature, that is, the absolute deviation rate Rmin; if the absolute deviation rate Rmin is ≤ the fifth set value 3%, the detection optimization is completed; if the absolute deviation rate Rmin is > the fifth set value 3%, executing step 5k);

[0066] 5k) Compare the minimum values ​​of the two boundary curvature radii in the HUD windshield detection area along the direction of the minimum curvature radius, take the one with the smaller minimum curvature radius as the reference, gradually increase the curvature radius of the windshield along the direction of the minimum curvature radius by 3 to 5% of the reference, and jump to step 4) each time the curvature radius is increased.

[0067] In the above:

[0068] Rmax absolute deviation rate = Rmax absolute deviation value × 100% / Rmax average value

[0069] Rmin absolute deviation rate = Rmin absolute deviation value × 100% / Rmin average value

[0070] Rmax absolute deviation value = [(Rmax-1-Rmax average value) + (Rmax-2-Rmax average value) + ... + (Rmax-2n 2 -Rmax average value)] / 2n 2

[0071] Rmin absolute deviation value = [(Rmin-1-Rmin average value) + (Rmin-2-Rmin average value) + ... + (Rmin-2n 2 -Rmin average value)] / 2n 2

[0072] By detecting the glass surface through the detection method of the present invention, it is possible to identify in advance whether the glass surface can meet the HUD display accuracy requirement; and a glass surface optimization method is provided, and the unsatisfactory items detected by the optimization method of the present invention are gradually optimized until the corresponding requirements are met, so as to achieve the glass surface curvature requirement that meets the HUD display accuracy, thereby avoiding the problem of unqualified display quality caused by the glass surface not meeting the requirements after the physical sample is produced in the later stage, and then changing the glass surface to incur cycle and cost.

[0073] The existing simulation analysis can confirm whether the glass surface can meet the HUD realistic accuracy, but the simulation analysis cycle is long and can only analyze whether it meets the HUD display requirements. There is no clear simulation optimization method, and there are repeated changes and repeated confirmations. The present invention can perform analysis after the first version of the modeling is completed, and the analysis cycle is short. After the analysis is confirmed to be OK, the glass surface data can be locked, thereby ensuring the locking nodes of the large surface data of the whole vehicle modeling.

Claims

1. A windshield inspection optimization method that meets HUD accuracy requirements, Features: The detection optimization method is as follows: 1) Establish the detection reference surface of HUD windshield; 2) Determine the reference plane and reference point of the detection area on the detection reference plane; 3) Determine the HUD windshield detection area and detection points: Orthogonally project the reference plane of the detection area onto the windshield to obtain the HUD windshield detection area, and orthogonally project all reference points onto the windshield to obtain the detection points; 4) Measure the curvature of the HUD windshield detection area: Measure the minimum value Rmin and the maximum value Rmax of the radius of curvature of each detection point; 5) Calculate and optimize The obtained minimum curvature radius Rmin and maximum curvature radius Rmax are compared with the set values ​​and the curvature radius of the windshield is adjusted.

2. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 1, Features: The specific process of comparing the obtained minimum curvature radius Rmin and the maximum curvature radius Rmax with the set value and adjusting the curvature radius of the windshield is as follows: 5a) calculating the average value of the minimum value of the radius of curvature, that is, the average value Rmin; if the average value Rmin is ≥ the first set value, executing step 5c); if the average value Rmin is < the first set value, executing step 5b); 5b) increasing the curvature radius of the windshield in the direction of the minimum curvature radius, and then jumping to step 4); 5c) calculating the average value of the maximum value of the curvature radius, that is, the average value Rmax; if the average value Rmax is ≥ the second set value, executing step 5e); if the average value Rmax is < the second set value, executing step 5d); 5d) increasing the curvature radius of the windshield in the direction of the maximum curvature radius, and then jumping to step 4); 5e) If the average value of Rmax / the average value of Rmin=the third setting range, execute step 5h); if the average value of Rmax / the average value of Rmin<the lower limit of the third setting range, execute step 5f); if the average value of Rmax / the average value of Rmin>the upper limit of the third setting range, execute step 5g); 5f) increasing the curvature radius of the windshield along the direction of the maximum curvature radius, and gradually increasing the curvature radius, and jumping to step 4) each time the curvature radius increases; 5g) increasing the curvature radius of the windshield along the direction of the minimum curvature radius, and gradually increasing the curvature radius, and jumping to step 4) each time the curvature radius is increased; 5h) calculating the absolute deviation rate of the maximum value of the radius of curvature, that is, the absolute deviation rate Rmax; if the absolute deviation rate Rmax is ≤ the fourth set value, executing step 5j); if the absolute deviation rate Rmax is > the fourth set value, executing step 5i); 5i) comparing the maximum values ​​of the two boundary curvature radii in the HUD windshield detection area along the direction of the maximum value of the curvature radius, taking the smaller one of the boundary curvature radii as a reference, gradually increasing the curvature radius of the windshield along the direction of the maximum value of the curvature radius by 3 to 5% of the reference, and jumping to step 4) every time the curvature radius is increased; 5j) calculating the absolute deviation rate of the minimum value of the radius of curvature, that is, the absolute deviation rate Rmin; if the absolute deviation rate Rmin is ≤ the fifth set value, the detection optimization is completed; if the absolute deviation rate Rmin is > the fifth set value, executing step 5k); 5k) Compare the minimum values ​​of the two boundary curvature radii in the HUD windshield detection area along the direction of the minimum curvature radius, take the one with the smaller minimum curvature radius as the reference, gradually increase the curvature radius of the windshield along the direction of the minimum curvature radius by 3 to 5% of the reference, and jump to step 4) each time the curvature radius is increased.

3. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 2, Features: The specific process of establishing the detection reference surface of the HUD windshield in step 1) is as follows: 1a) The XZ plane where point R is located is the symmetry plane of the human eye, point R is the seat reference point, and X and Z are the X and Z axes of the vehicle coordinate system; 1b) On the symmetry plane of the human eye, make a downward sight line through point V0. The angle between the downward sight line and the X-axis is the HUD design angle. Point V0 is the middle eye point between the upper eye point V1 and the lower eye point V2. The point where the lower sight line intersects the windshield is point O. The tangent reference plane of the windshield is made through point O, and the tangent reference plane is the detection reference plane.

4. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 3, Features: The specific process of determining the reference plane and reference point of the detection area on the detection reference plane in step 2) is as follows: Make a square on the detection reference plane, and then divide the square into n×n squares to determine the detection area reference plane, 3≤n≤8, and the center point of the square is point O, and one side of the square is perpendicular to the lower line of sight; Take two reference points on each square to obtain a total of 2n 2 There are reference points, and the line connecting the two reference points on each square passes through the center point of the square. The lines connecting the two reference points on all squares are in the same direction. At the same time, the distance from each reference point to the two perpendicular sides of the square where the reference point is located is equal.

5. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 2, Features: In the step 5b), the increased curvature radius = (first set value - Rmin average value) + 100 to 200 mm.

6. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 2, Features: In the step 5d), the increased radius of curvature = (the second set value - the average value of Rmax) + 100 to 200 mm.

7. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 2, Features: In the step 5f), the curvature radius is gradually increased according to the increase of the average value of Rmax×(3-5%), and the process jumps to step 4) after each increase.

8. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 2, Features: In the step 5g), the curvature radius is gradually increased according to the increase of the average value of Rmin×(3-5%), and the process jumps to step 4) after each increase.

9. The windshield detection optimization method that meets the HUD accuracy requirements according to claim 4, Features: The Rmax absolute deviation rate = Rmax absolute deviation value × 100% / Rmax average value; Rmin absolute deviation rate = Rmin absolute deviation value × 100% / Rmin average value; Where: Rmax absolute deviation value = [(Rmax-1-Rmax average value) + (Rmax-2-Rmax average value) + ... + (Rmax-2n 2 -Rmax average value)] / 2n 2 Rmin absolute deviation value = [(Rmin-1-Rmin average value) + (Rmin-2-Rmin average value) + ... + (Rmin-2n 2 -Rmin average value)] / 2n 2 .

Images