TEST DEVICE FOR A HEAD-UP DISPLAY (HUD)

DE502019014469D1Active Publication Date: 2026-03-26SAINT GOBAIN SEKURIT FRANCE
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Filing Date
2019-06-28
Publication Date
2026-03-26

AI Technical Summary

Technical Problem

Conventional testing devices struggle to capture the entire image sharply due to the limited depth of field and curvature of the windshield in contact-analog HUDs, leading to blurred and distorted images, especially with larger projection distances and areas.

Method used

Incorporation of an optical element, such as a curved mirror, to redirect radiation and compensate for image curvature and astigmatism, ensuring the entire virtual image lies within the camera's depth of field.

Benefits of technology

Enables sharp capture and evaluation of the entire HUD image, reducing space requirements and improving measurement accuracy by aligning the image within the camera's depth of field, suitable for contact-analog HUDs with large projection distances.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a test device for a head-up display (HUD) for a contact-analog HUD, and a method for testing such a HUD.

[0002] Modern automobiles are increasingly equipped with so-called head-up displays (HUDs). A projector, located, for example, in the dashboard or roof area, projects images onto the windshield, where they are reflected and perceived by the driver as a virtual image (from their perspective) behind the windshield. This allows important information to be projected into the driver's field of vision, such as current speed, navigation instructions, or warnings, which the driver can perceive without taking their eyes off the road. Head-up displays can thus significantly contribute to improving road safety. The projection distance (distance between the virtual image and the driver) of a classic HUD for displaying static information is typically around 2 meters.

[0003] A newer variant of the HUD is called a contact-analog HUD or Augmented Reality These are referred to as HUDs (Head-Up Displays). These HUDs are characterized by a larger HUD area (projection surface on the windshield) and a significantly greater projection distance of at least 5 meters, typically even over 7 meters. Contact-analog HUDs offer the possibility of not only projecting static, readable information onto the windshield, but also using the optical information to highlight elements of the vehicle's real-world surroundings. Examples of applications include the optical marking of lane boundaries, the visual highlighting of pedestrians at the roadside, navigation instructions directly on the road, or the marking of vehicles detected by the driver assistance system. The greater projection distance is achieved by a longer optical path length of the rays within the projector, for example, through additional mirrors and a larger volume.Contact analog HUDs are known, for example, from DE102014001710A1, WO2014079567A1, US2013249942A1, US2014354692A1, US2014375816A1 and WO2013136374A1.

[0004] When designing a head-up display (HUD), it is standard practice to examine the optical quality of the image in a test setup. The windshield is illuminated by an imaging unit, generating a virtual image corresponding to the HUD image, typically in the form of a characteristic pattern of points and lines. This image is then captured through the windshield by a camera unit from a variety of possible driver eye positions. Based on these images, the occurrence of distortions, aberrations, ghosting due to multiple reflections, and other optical effects can then be evaluated in detail for different eye positions.

[0005] Due to the curved geometry of conventional windshields and the associated field curvature, the virtual HUD image is not arranged in a plane but is also curved in space. This poses no problem when testing conventional HUDs because the image has a relatively small area and can therefore be captured sharply by the camera unit despite the curvature. However, with contact-analog HUDs, which have a significantly larger image size, the image curvature can lead to parts of the image being blurred due to the limited depth of field of the camera unit, making evaluation difficult or impossible. Further image errors result from astigmatism, since the light rays strike the windshield at an angle to the optical axis.

[0006] In Alexander Neumann's dissertation, "Simulation-Based Measurement Technology for Testing Head-up Displays," January 1, 2012 (2012-01-01), XP055558288, a test setup for examining a head-up display on a windshield master is shown. The use of a curved mirror to compensate for image errors caused by the curvature of the windshield is described in WO 2017 / 138430 A1, US 4790613 A, and US 9551867 B1.

[0007] The invention is based on the objective of providing an improved testing device for a head-up display (HUD) which is applicable to contact-analog HUDs.

[0008] This problem is solved according to the invention by a testing device according to claim 1. Preferred embodiments are described in the dependent claims.

[0009] Typical test devices comprise an imaging unit, a positioning device, and a camera unit. The test device according to the invention for a head-up display (HUD) additionally comprises an optical element. The imaging unit emits radiation to generate a virtual image after reflection from a windshield under test, the positioning and extent of which correspond to the subsequent HUD image. However, the windshield is not directly illuminated by the imaging unit, but indirectly via the optical element. The optical element is suitable and designed to redirect the radiation from the imaging unit towards the windshield and thereby illuminate the HUD area of ​​the windshield to generate the virtual image. The positioning device is suitable and designed to fix the windshield in a defined arrangement relative to the optical element and the imaging unit.The windshield should be illuminated by the imaging unit in the same way as it will later be by the HUD projector in the vehicle, so that the image generated in the test setup is a good model for the subsequent HUD projection. The radiation from the imaging unit should therefore strike essentially the same area of ​​the windshield with the same angle of incidence and aperture as the radiation from the HUD projector will later. The camera unit is suitable and designed to capture the virtual image through the windshield from various eye positions. Eye positions refer to the possible positions of the eyes of a future viewer / user of the HUD.

[0010] The optical element reduces the curvature of the virtual image in space (field curvature), enabling the entire image to be captured sharply by the camera unit, which is a significant advantage of the present invention. The optical element is designed such that the entire virtual image lies within the depth of field of the camera unit. More precisely, this means that the extent of the image along the optical axis (running between the camera unit and the center of the image) is less than or equal to the extent of the depth of field of the camera unit, making it possible to focus the camera unit on the image in such a way that the entire image lies within this depth of field and is thus displayed sharply. Strictly speaking, the image comprises the superposition of the main image and the ghost image, since both are to be displayed sharply.Ideally, the main image is arranged in a single plane onto which the camera unit can focus. However, a slight deviation from this ideal planarity, i.e., a slight residual curvature of the image, is acceptable. The degree of acceptable curvature depends on the camera unit used and its depth of field. In addition to field curvature, astigmatism can also be compensated for by the optical element according to the invention.

[0011] The invention aims to shorten the image extent, thereby enabling the entire image to be captured within the camera's depth of field. This problem arises in connection with contact-analog HUDs. Without the optical element, the virtual image would not lie entirely within the camera unit's depth of field. In other words, without the optical element, the image extent along the optical axis is greater than the camera unit's depth of field.

[0012] Another advantage of the invention is that the beam path is bent through the optical element, which makes the testing device significantly more space-saving, which is particularly advantageous in connection with contact-analog HUDs and their large projection distances.

[0013] The invention also includes a method for testing a HUD with a test device according to the invention, comprising the following method steps: (a) Positioning a windshield with a HUD area by means of a positioning device in a defined arrangement relative to an optical element, (b) irradiating the optical element by an imaging unit, wherein the radiation is deflected by the optical element towards the windshield and the HUD area is irradiated, generating a virtual image, (c) capturing the virtual image by means of a camera unit from different eye positions, wherein the optical element (13) is designed such that the entire virtual image (7) lies within the depth of field range (Δd) of the camera unit (12).

[0014] The invention is described in detail below, with preferred embodiments relating equally to the device and the method.

[0015] The optical element is a curved mirror. By appropriately designing the mirror's curvature, it is possible to compensate for unwanted image field curvature of the virtual image and any astigmatism. For example, the mirror's radius of curvature can change from the center towards the edge to influence the positioning of the image portion generated by the respective mirror areas. The geometry of the disk being tested is crucial in the mirror's design, as is the position of the camera unit. The required design, particularly the mirror's curvature, can be determined using standard methods, such as the well-known Coddington equations, especially with the application of ray tracing techniques. The optical element is preferably a highly polished metal mirror, which can be manufactured with high precision.Alternatively, other optical elements are also conceivable, for example lenses with a suitable curvature profile.

[0016] In a preferred embodiment, the optical element, particularly when designed as a mirror, is not rotatable but statically mounted. This allows for a simpler and less error-prone design, enabling higher measurement accuracy and faster measurement. The optical element must then be designed such that the entire eyebox is illuminated simultaneously. This distinguishes the optical element according to the invention, in particular, from a mirror within a HUD projector, which can only illuminate a limited portion of the eyebox window and must be rotated to cover the entire eyebox.

[0017] The test device according to the invention is particularly advantageously suited for testing contact-analog HUDs (so-called Augmented Reality Head-Up-Displays (AR-HUD)). These have significantly greater projection distances (distance between image and viewer) than classic HUDs and significantly larger projected images. As a result, the curvature effect of the projected image during recording is more pronounced, potentially making it impossible to capture the entire image sharply with conventional testing devices, since parts of the image extend outside the depth of field of the camera unit. This effect is reduced by the device according to the invention. Classic HUDs have projection distances of less than 3 m, while contact-analog HUDs have projection distances of (sometimes significantly) more than 3 m. The virtual image is therefore generated with a projection distance of more than 3 m, preferably more than 4 m, and particularly preferably more than 5 m.

[0018] A projection arrangement for a head-up display (HUD) comprises at least a vehicle windshield (in particular, that of a motor vehicle, for example, a passenger car) and a projector. The projector illuminates an area of ​​the windshield where the radiation is reflected towards the viewer (driver), thereby creating a virtual image that the viewer in the vehicle perceives as if behind the windshield. The area of ​​the windshield that can be illuminated by the projector is called the HUD area. The projector is directed at the HUD area. The direction of the projector's beam can typically be varied by mirrors, especially vertically, to adjust the projection to the viewer's height. The area in which the viewer's eyes must be located for a given mirror position is called the eyebox window.This eyebox window can be moved vertically by adjusting the mirrors, with the entire accessible area (that is, the superposition of all possible eyebox windows) being referred to as the eyebox. A viewer located within the eyebox can perceive the virtual image. This means, of course, that the viewer's eyes must be within the eyebox, not their entire body. The eyebox is, in effect, the sum of all possible positions of a viewer's eyes, typically referred to as eye positions. The beam that runs between the projector and the center of the eyebox is commonly called the center beam. It is a characteristic reference beam for the design of a HUD projection setup.

[0019] The technical terms used here from the field of HUDs are generally known to experts. For a detailed explanation, please refer to the dissertation "Simulation-Based Measurement Technology for Testing Head-Up Displays" by Alexander Neumann at the Institute of Computer Science of the Technical University of Munich (Munich: University Library of the Technical University of Munich, 2012), mentioned at the beginning, in particular to Chapter 2 "The Head-Up Display".

[0020] The perception of the HUD projection depends on the eye position. A HUD is typically optimized for the center of the eyebox, and at other eye positions, distortions, stronger ghosting, or other undesirable optical effects can sometimes occur. The device according to the invention is intended to verify this.

[0021] The HUD area of ​​the windshield is illuminated by the imaging unit to generate a virtual test image. The test image is a pattern of dots and / or lines. Such patterns can subsequently be easily evaluated quantitatively with regard to optical criteria. The pattern can have a pre-distortion so that any distortion unavoidably caused by the windshield is compensated for and the pattern appears regular in the virtual image (at least with respect to the central beam). According to the invention, the imaging unit comprises a plate and a planar light source that back-illuminates the plate. The pattern is incorporated into the plate in the form of perforations similar to a perforated plate. The plate can be made of, for example, metal or plastic. Alternatively, the imaging unit can also be implemented, for example, as a display (screen), such as an LED, LCD, or DLP display.

[0022] The radiation from the imaging unit strikes the optical element and is projected by it onto the HUD area of ​​the windshield, which is fixed in the positioning device. The positioning device is, for example, a holder into which the windshield is clamped to ensure its position is reproducible. The relative arrangement of the windshield, optical element, and imaging unit is chosen so that the radiation strikes the windshield at the same angle and with the same extent as the radiation from the HUD projector later does, thus ensuring that the test image is a realistic model for the HUD projection.

[0023] The resulting virtual test image is captured by the camera unit from different eye positions. A single camera can be used, mounted in a movable manner, for example on a robot arm, so that it can be moved between the different eye positions. Alternatively, a multitude of cameras can be used, each assigned to a specific eye position. In this case, movement during the test is not required, and the cameras are therefore preferably mounted in a static position.

[0024] Using standard image processing software, the images captured by the camera unit are subsequently analyzed. This allows for the examination of the occurrence and intensity of ghost images, distortions (such as rotational or trapezoidal distortions), and other optical effects, depending on the eye position.

[0025] A windshield typically comprises an outer pane and an inner pane bonded together by a thermoplastic interlayer. The windshield is designed to separate the interior of a vehicle from the external environment within a window opening. For the purposes of this invention, the term "inner pane" refers to the pane of the composite glass facing the interior (vehicle interior). The term "outer pane" refers to the pane facing the external environment. Windshields are typically curved, with common radii of curvature ranging from approximately 10 cm to approximately 40 m. In the head-up display (HUD) area, radii of curvature typically range from 2 m to 20 m. The interior surface of the inner pane is essentially concave, while the exterior surface of the outer pane is convex. The interior surface is defined as the surface facing the vehicle's interior when the windshield is installed.The term "outer surface" refers to the surface that faces the external environment when the product is installed.

[0026] Just as in the HUD, the windshield in the test device is irradiated from the inside. The HUD projector and the imaging unit are therefore positioned on the inside of the windshield, so that its inner pane faces them. The windshield is thus irradiated via the inner surface of the inner pane.

[0027] Head-up Displays (HUDs) often suffer from the problem of so-called ghost images. The projector is located inside the vehicle, and its beam strikes the interior surface of the windshield, where a portion is reflected towards the driver's eyes, creating the desired virtual image. From the driver's perspective, this image appears to be behind the windshield, i.e., on the outside. The majority of the beam passes through the windshield, but some is reflected again from the outer surface. This secondary reflection creates another HUD image—the ghost image—which the driver perceives as slightly offset from the main image and with lower intensity, thus causing distraction.Typical windshields have a wedge shape, so that the inner surface of the inner pane and the outer surface of the outer pane are angled relative to each other in order to superimpose the ghost image onto the main image or at least reduce the distance between them, making the ghost image less noticeable. However, this compensation is often not ideal and also depends on the eye's position.

[0028] The wedge shape of the windshield is typically achieved by using a wedge-shaped thermoplastic interlayer. The thickness of this interlayer varies, at least in sections, along the vertical path between the top and bottom edges of the windshield. "Sectionally" here means that the vertical path between the top and bottom edges includes at least one section where the thickness of the interlayer changes depending on the location, thus exhibiting a wedge angle. The thickness of the interlayer varies, at least in the area covered by the windshield (HUD). However, the thickness can also change in several sections or increase essentially continuously along the entire vertical path, for example, from the bottom to the top edge. "Vertical path" refers to the path between the top and bottom edges with a direction essentially perpendicular to the top edge.Since the upper edge of windshields can deviate significantly from a straight line, the vertical profile, more precisely defined according to the invention, is aligned perpendicular to the line connecting the corners of the upper edge. The intermediate layer exhibits a finite wedge angle, at least in sections, i.e., a wedge angle greater than 0°, namely in the section where the thickness varies. The wedge angle is defined as the angle between the two surfaces of the intermediate layer. If the wedge angle is not constant, the tangents to the surfaces are used to measure it at a point. Typical wedge angles are in the range of 0.2 mrad to 1 mrad, particularly from 0.3 mrad to 0.7 mrad. The wedge angle can be constant along the vertical profile, resulting in a linear change in the thickness of the intermediate layer, with the thickness typically and preferably increasing from bottom to top.The direction "from bottom to top" refers to the direction from the bottom edge to the top edge. However, more complex thickness profiles are also possible, in which the wedge angle varies from bottom to top (i.e., is location-dependent along the vertical path), either linearly or non-linearly.

[0029] In the case of a contact-analog HUD or Augmented Reality The HUD does not merely project information onto a limited area of ​​the windshield, but incorporates elements of the external environment into the display. Examples include marking a pedestrian, displaying the distance to a vehicle ahead, or projecting navigation instructions directly onto the road, for example, to indicate the lane to be selected. The contact-analog HUD differs from a classic, static HUD in that the projection distance is greater than 3 m, preferably greater than 4 m, and typically even more than 5 m. With a static HUD, the projection distance is significantly shorter, typically around 2 m. For the purposes of this invention, projection distance refers to the distance between the virtual image and the viewer, usually the driver's head. The projection distance is preferably at least 7 m. The projection distance is preferably at most 15 m.

[0030] In a contact-analog HUD projection setup, the projection distance is essentially constant for all projected images. Even projections that the viewer is meant to perceive as being at different distances actually have essentially the same projection distance. The subjective impression of a different distance is achieved through geometric optical effects.

[0031] The distance between the windshield and the virtual image is usually referred to as the image distance. Since the driver's head is typically about one meter away from the windshield, the image distance is approximately 1 meter less than the projection distance. Therefore, the image distance can be used as a criterion with sufficient accuracy as an alternative to the projection distance. Accordingly, the image distance is preferably more than 2 meters, particularly preferably more than 3 meters, most preferably more than at least 4 meters, and preferably at most 14 meters.

[0032] The HUD area of ​​a contact-based HUD is typically larger than that of a conventional static HUD. In a preferred embodiment, the area of ​​the HUD area is at least 7% of the windshield area, and particularly preferably at least 8%. The area of ​​the HUD area of ​​a static HUD is typically at most 4-5% of the windshield area.

[0033] The inner and outer panes are preferably made of glass, particularly soda-lime glass, which has proven effective for window glass. However, the panes can also be made of other types of glass, such as borosilicate glass or aluminosilicate glass. Alternatively, the panes can be made of plastic, especially polycarbonate or PMMA. The thicknesses of the outer and inner panes can generally be freely selected within the range of standard industry values. For conventional windshields, thicknesses of the individual panes in the range of 1 mm to 5 mm, particularly 1.2 mm to 3 mm, are common. Standard pane thicknesses are, for example, 2.1 mm or 1.6 mm. The outer pane, the inner pane, and the thermoplastic interlayer can be clear and colorless, but also tinted or colored. In a preferred embodiment, the total transmission through the laminated glass is greater than 70%, particularly when the laminated glass is a windshield.The term total transmission refers to the procedure for testing the light transmittance of motor vehicle windows as laid down by ECE-R 43, Annex 3, § 9.1.

[0034] The intermediate layer is typically formed by at least one thermoplastic film, preferably based on polyvinyl butyral (PVB), ethylene vinyl acetate (EVA), polyurethane (PU), or mixtures, copolymers, or derivatives thereof, particularly preferably PVB. The thickness of the intermediate layer, measured at its thin edge in the case of a wedge film, is typically in the range of 0.2 mm to 2 mm, particularly 0.5 mm to 1 mm.

[0035] The installation angle of the windshield is typically in the range of 55° to 75° to the horizontal, particularly 60° to 70°, for example approximately 65°.

[0036] Also shown is a test setup for a HUD, comprising the test device according to the invention and a windshield fixed in the positioning device.

[0037] The HUD tested according to the invention is preferably used in a vehicle, particularly preferably in a motor vehicle, and most preferably in a passenger car.

[0038] Also shown is the use of an optical element for deflecting the radiation of an imaging unit onto a HUD area of ​​a windshield, the arrangement of which to the optical element is fixed by a positioning device, wherein a virtual image is generated which is recorded through the windshield for testing purposes by means of a camera unit from different eye positions, wherein the optical element is designed such that the entire virtual image lies within the depth of field of the camera unit.

[0039] The invention is explained in more detail below with reference to a drawing and exemplary embodiments. The drawing is a schematic representation and not to scale. The drawing does not limit the invention in any way.

[0040] They show: Fig. 1 a top view of the windshield for a HUD, Fig. 2 a cross-section through the windshield made of Fig. 1 as a projection surface of a HUD, Fig. 3 a side view through a conventional testing device for HUDs, Fig. 4 a side view through a testing device for HUDs according to the invention, Fig. 5 simulated example images of a conventional testing device according to Fig. 3 and a test device according to the invention Fig. 4 and Fig. 6 a schematic illustration of the extent of the virtual image and the depth of field of the camera unit.

[0041] Fig. 1 Figure 1 shows a top view of a windshield 1 of a HUD projection system. The windshield 1 has a top edge O, a bottom edge U, and two connecting side edges. In its installed position, the top edge O points upwards towards the vehicle roof (roof edge), and the bottom edge U points downwards towards the engine compartment (engine edge). The windshield 1 has a HUD area B, which, in its installed position, can be illuminated by the HUD projector and is illuminated during operation.

[0042] Fig. 2 shows a cross-section through a HUD projection arrangement, encompassing the windshield 1 from Fig. 1 and a HUD projector 5. The windshield 1 is cut by the section line AA'. The windshield 1 consists of an outer pane 2 and an inner pane 3, which are bonded together via a thermoplastic intermediate layer 4. The windshield 1 separates the vehicle interior from the external environment, with the outer pane 2 facing the external environment in its installed position and the inner pane 3 facing the vehicle interior. The installation angle β to the horizontal is, for example, 65°.

[0043] The outer pane 2 and the inner pane 3 are made of non-tempered soda-lime glass, for example. The outer pane 2 has a thickness of 2.1 mm and the inner pane 3 a thickness of 1.6 mm. These panes are commonly used for windshields. The thickness of the intermediate layer 4 increases steadily vertically from the lower edge U to the upper edge O, with a substantially constant wedge angle α between the two surfaces. The intermediate layer 4 consists of a single sheet of PVB. The thickness of the intermediate layer 4 at the upper edge O is, for example, 1.0 mm and at the lower edge U, for example, 0.76 mm. Due to the wedge shape of the intermediate layer 4, the two images generated by reflection of the projector image on the two surfaces of the outer pane 2 and the inner pane 3 facing away from the intermediate layer 4 are superimposed. Therefore, distracting ghost images occur to a lesser degree.

[0044] Projector 5 is directed at HUD area B. Images are to be generated in this area by projector 5. The projected image is reflected off the windshield 1 towards the viewer 6 (vehicle driver). This creates the virtual image 7, which the viewer 6, located inside the vehicle, perceives as being behind the windshield 1. The distance between the viewer 6 and the virtual image 7 is called the projection distance d. The distance between the windshield 1 and the virtual image 7 is called the image distance w.

[0045] The projection setup is a so-called contact-analog HUD or Augmented Reality-HUD, which is characterized by a large projection distance d of, for example, 10 m. This allows the surroundings to be incorporated into the visual display, so that, for example, the lane to be selected can be seemingly projected directly onto the road as a navigational aid for the viewer 6. In addition to the larger projection distance d, the contact-analog HUD also differs from the classic HUD by a larger HUD area B, the surface area of ​​which is, for example, 9% of the area of ​​the windshield 1.

[0046] The area within which the viewer's eyes must be positioned to perceive the virtual image is called the eyebox window. The eyebox window is vertically adjustable by mirrors in the projector to adapt the HUD to viewers of different heights and seating positions. The entire accessible area within which the eyebox window can be moved is called the eyebox E.

[0047] A number of undesirable optical effects can occur with a HUD. These include distortions of the display, such as rotations or trapezoidal distortions. Furthermore, the projector beam is reflected off both outer surfaces of the windshield 1, resulting in a slightly offset ghost image in addition to the main image. The wedge angle α is intended to superimpose the main image and ghost image as much as possible, with the central beam (the beam between projector 5 and the center of the eyebox E) usually being the basis for the wedge angle's design. However, the occurrence of these optical effects is highly dependent on the eye position, i.e., the viewer's precise positioning within the eyebox E. Therefore, the occurrence of these effects as a function of eye position should be tested, for which purpose test devices are the subject of the present invention.

[0048] Fig. 3 Figure 1 shows a generic test device for HUDs according to the prior art. A windshield 1 for a HUD is fixed in a positioning device 11, which determines its arrangement relative to an imaging unit 8. The imaging unit comprises a plate 9 with a characteristic pattern of through-holes in the form of points and / or lines, and also a planar light source 10, which back-illuminates the plate 9. The relative arrangement of the windshield 1 and the imaging unit 8 corresponds to the relative arrangement of the windshield 1 and the HUD projector 5 in the planned HUD setup at the deployment site. The HUD area B of the windshield 1 is illuminated by the imaging unit 8, generating a virtual image 7 of the characteristic pattern of points and / or lines behind the windshield 1. The radiation is represented by a gray block arrow.The virtual image 7 is captured from various eye positions within the eyebox E using a camera unit 12, for example, a camera mounted on a robot arm. The images from camera unit 12 can then be analyzed in detail using image processing software. This allows for the quantitative evaluation of undesirable optical effects, such as ghosting or distortion, depending on the eye position, in order to draw conclusions about the quality of the HUD display.

[0049] As indicated in the illustration, the virtual image 7 is not flat, but curved in space, primarily due to the curvature of the screen. Particularly with larger images 7, such as those found in contact-analog HUDs, problems can arise during recording. The camera unit 12 has a limited depth of field, so it is sometimes not possible to capture the entire image in focus. Instead, at each focal plane, there are areas of the image 7 that extend outside the depth of field and are therefore captured out of focus. This impairs the evaluation of the recordings.

[0050] Fig. 4 In contrast, a test device for HUDs according to the invention is shown. Unlike the conventional test device made of Fig. 3 The windshield 1 is not directly illuminated by the imaging unit 8. Instead, the radiation is deflected towards the windshield 1 by an optical element 13. The radiation is represented by gray block arrows. The optical element 13 can influence the virtual image 7, transforming it so that it is approximately arranged in a plane, as indicated in the illustration. At a minimum, the curvature of the image 7 is reduced to such an extent that its extent along the optical axis is at most as large as the depth of field of the camera unit 12. The camera unit 12 can then be focused so that the entire image 7 is captured sharply, allowing for easy evaluation.

[0051] The optical element 13 is designed, for example, as a curved mirror made of a highly polished metal plate, wherein the inventive influence on the curvature of the image 7 is achieved by a suitable curvature profile of the mirror, in which the radius of curvature changes appropriately from the center of the mirror towards the side edges.

[0052] Fig. 5 shows simulated recordings of an example of the present invention and a comparative example. Fig. 5 (a) A conventional test device was used according to Fig. 3 The scenario used is one in which the windshield 1 was directly illuminated by the imaging unit 8 from the position of the HUD projector. It is clearly visible that areas of the image are blurred because they are not within the depth of field of the camera unit 12. In contrast, in Fig. 5 (b) a test device according to the invention Fig. 4 Based on an aspherical mirror as optical element 13. The extent of the image 7 along the optical axis can be reduced by the optical element 13, so that the image 7 lies completely within the depth of field and is therefore sharply imaged.

[0053] In the simulations, a backlit perforated plate with a two-dimensional pattern of holes was assumed as the imaging unit, a projection distance of 10 m and an image size of 1600 mm x 800 mm (measured in the plane of the virtual image at a distance of 10 meters from the central camera (eyepoint)). Fig. 5 (a) The distance between the perforated plate and the composite panel is 10 m. Fig. 5 (b) The distance between the aspherical mirror and the composite disc is 300 mm.

[0054] Fig. 6 This illustrates the desired effect of the present invention. The camera unit 12 has a limited depth of field Δd. For a given focus setting, this is the area between the far point df (the point furthest from the camera unit 12 that is in focus) and the near point dn (the point closest to the camera unit 12 that is in focus) along the optical axis (the axis connecting the camera unit 12 and the center of the image 12). The virtual image 7 is curved in space, primarily due to the curvature of the windshield 1. This results in an extension Δx of the image 7 along the optical axis. If the extension Δx is greater than the depth of field Δd, there is no focal plane with which the entire image 7 can be in focus – it always extends partially outside the depth of field Δd. Fig. 6a The optical element 13 according to the invention reduces the extent Δx so that it is smaller than the depth of field Δd. A focal plane can now be selected for which the entire image 7 is displayed sharply and can subsequently be analyzed ( Fig. 6b In order to enable the possibility of an overall sharply displayed image 7, the following must generally apply: Δx ≤ Δd

[0055] For simplicity, image 7 is represented here as a single line. In reality, the dimension Δx refers to the entirety of the main image and the ghost image, since both should be displayed sharply for analysis. Reference symbol list:

[0056] 1 Windshield 2 Outer pane 3 Inner pane 4 Thermoplastic interlayer 5 Projector 6 Viewer / Driver 7 Virtual image 8 Imaging unit 9 (Metal) plate with pattern of feedthroughs 10 Planar light source 11 Windshield positioning device 12 Camera unit 13 Optical element OUpper edge of the windshield 1 ULower edge of the windshield 1 BHUD area of ​​the windshield 1 α Wedge angle of the intermediate layer 4 β Installation angle from 1 to the horizontal d Projection distance / distance between 6 and 7 w Image distance / distance between 1 and 7 dn Near point of camera unit 12 df Far point of camera unit 12 Δd Depth of field of camera unit 12 Δx Extent of image 7 along the optical axis (E)Eyebox A-A'vertical section line

Claims

1. Testing device for a head-up display (HUD), at least comprising - an imaging unit (8), - an optical element (13), arranged for deflecting the radiation of the imaging unit (8) in the direction of a windshield (1) and thereby irradiating an HUD region (B) of the windshield (1) in order to generate a virtual image (7), wherein the HUD region (B) of the windshield (1) has radii of curvature ranging from 2 m to 20 m, - a positioning device (11), arranged for fixing the windshield (1) in a defined arrangement relative to the optical element (13), and - a camera unit (12), arranged for capturing the virtual image (7) through the windshield (1) from different eye positions, wherein the optical element (13) is a curved mirror or a lens and is configured such that a curvature of the virtual image due to the windshield (1) is reduced in space, such that the entire virtual image (7) lies within the depth of field range (Δd) of the camera unit (12), wherein the expansion (Δx) of the virtual image (7) along the optical axis is less than or equal to the depth of field range (Δd) of the camera unit (12), while the expansion (Δx) of the virtual image (7) along the optical axis without using the optical element (13) is greater than the depth of field range (Δd) of the camera unit (12), wherein the testing device is configured to generate the virtual image (7) with a projection distance (d) of more than 3 m, wherein the imaging unit (8) includes a plate (9) with a pattern of perforations, backlit by a flat light source (10), or is implemented as a display.

2. Testing device according to claim 1, which is suitable for generating the virtual image (7) with a projection distance (d) of more than 4 m, preferably more than 5 m.

3. Testing device according to one of claims 1 through 2, wherein the optical element (13) is a highly polished metal mirror.

4. Testing device according to one of claims 1 through 3, wherein the radius of curvature of the mirror or the lens is variable from the centre toward the edges.

5. Testing device according to one of claims 1 through 4, wherein the optical element (13) is mounted non-rotatably.

6. Testing device according to one of claims 1 through 5, wherein the camera unit (12) is implemented as a single, movably mounted camera or as a plurality of statically mounted cameras.

7. Method for testing a head-up display (HUD) with the testing device according to one of claims 1 through 6, comprising the following steps: (a) Positioning a windshield (1) with an HUD region (B) by means of a positioning device (11) in a defined arrangement relative to an optical element (13), (b) Irradiating the optical element (13) by an imaging unit (8), wherein the radiation is deflected by the optical element (13) in the direction of the windshield (1), and the HUD region (B) is irradiated, with a virtual image (7) generated thereby, (c) Capturing the virtual image (7) by means of a camera unit (12) from different eye positions, wherein the optical element (13) is configured such that a curvature of the virtual image due to the windshield (1) is reduced in space, such that the entire virtual image (7) lies within the depth of field range (Δd) of the camera unit (12) wherein the expansion (Δx) of the image (7) along the optical axis is less than or equal to the depth of field range (Δd) of the camera unit (12), while the expansion (Δx) of the image (7) along the optical axis without using the optical element (13) is greater than the depth of field range (Δd) of the camera unit (12), wherein the virtual image (7) is generated with a projection distance (d) of more than 3 m.

8. Method according to claim 7, wherein the virtual image (7) is generated with a projection distance (d) of more than 4 m, preferably more than 5 m.

9. Method according to claim 8, wherein the HUD region (B) makes up at least 7% of the area of the windshield (1), preferably at least 8%.

10. Method according to one of claims 7 through 9, wherein the windshield (1) comprises an outer pane (2) and an inner pane (3) that are joined to one another by means of a thermoplastic intermediate layer (4), wherein the intermediate layer (4) is wedge-shaped at least in the HUD region (B).

11. Method according to claim 10, wherein the windshield (1) is irradiated via the inner pane (3).