Method for detecting dirt on a windshield of a vehicle

Abstract

The invention relates to a method for detecting soiling (V) on a windscreen (2) of a vehicle (3), wherein in an application method - in data of the vehicle environment recorded by an environment detection sensor (5), positions of surrounding objects (O1 to On) and distances (d, d1 to d8) of the surrounding objects (O1 to On) to the vehicle (3) are determined, - in images (B1 to Bz) of the eyes (A1, A2) of the driver (F) captured by a driver observation camera (4), the gaze direction (BR1 to BRm) of the driver (F) and the accommodation (AK) of the eyes (A1, A2) are determined, - a correspondence analysis determines which surrounding object (O1 to On) the driver (F) is looking at, and - using a calibration function (f) determined in a calibration process, a soiling-free windshield (2) is detected if a relationship between the distance (d, d1 to d8) of the respective surrounding object (O1 to On) and the accommodation (AK) of the eyes (A1, A2) of the driver (F) when looking at this surrounding object (O1 to On) lies within a predetermined target range, and soiling (V) of the windshield (2) in the field of vision of the driver (F) is detected if the relationship lies outside the predetermined target range. The invention further relates to a method for operating a cleaning device (1) for a windshield (2) of a vehicle (3).

Description

The invention relates to a method for detecting soiling on a windshield of a vehicle.The invention further relates to a method for operating a cleaning device for a windshield of a vehicle.DE 10 2020 001 344 A1 discloses a method for detecting the environment of a motor vehicle, using a camera system comprising a camera. The camera is positioned in a first position and a recording of the environment is captured by means of the camera. The camera system comprises an electronic computing device, by means of which a control signal for the camera system is generated. The camera is adjusted by means of a holding device arranged on the camera system as a function of the control signal into a second position different from the first position. The recording of the environment is analyzed by means of the electronic computing device and a contamination on a windshield of the motor vehicle is detected by means of an evaluation of the recording.Furthermore, DE 10 2009 054 194 A1 discloses a vehicle having a device for generating a head-up display, wherein a camera, comprising an image sensor and a focusing element, is introduced statically or dynamically into the beam path of the head-up display via an optical or mechanical element. In this case, a region of an outer side of the windshield is imaged in a focused manner on the image sensor or an eye region of the driver is imaged in a focused manner on the image sensor. The vehicle has assistance functions based on an evaluation of the data of the image sensor, wherein the assistance functions comprise a driver detection, a fatigue detection and a rain / dirt detection.DE 10 2021 005 212 A1 describes a method for adapting a calibration of a driver observation sensor of a motor vehicle, whereina plurality of surroundings recordings are recorded by means of a surroundings sensor,an object is detected in at least two surroundings recordings of the plurality of surroundings recordings and object tracking of the at least one detected object is carried out,determining by means of the driver observation sensor whether a driver of the motor vehicle is observing the detected object,if the driver observes the detected object, it is determined for each vision beam of a plurality of vision beams determined by means of the driver observation sensor whether there is an intersection point of the vision beam with the observed object, andthe calibration of the driver observation sensor is evaluated and adapted on the basis of the plurality of sight beams.DE 10 2015 005 382 A1 describes a method for operating a windshield wiper of a motor vehicle, comprising the steps:detecting a predetermined environmental condition outside the motor vehicle,detecting a predetermined viewing direction of a driver of the motor vehicle,activating a windshield wiper as a function of the detected predetermined viewing direction and of the detected predetermined environmental state.DE 10 2020 205 328 A1 describes a method for displaying objects in a vehicle that are superimposed by the environment, for which purpose an eye focusing plane of a person using the vehicle is detected. A selection of the displayed objects takes place depending on the eye focusing plane and a lens curvature of one eye or both eyes.DE 10 2017 103 287 A1 describes a display controller for a vehicle, wherein the display controller comprises:a display control device that displays a predetermined display image including display of a traveling state on a front windshield of the vehicle;a visibility decreasing region detecting detector that detects presence or absence of a visibility decreasing region on the front glass that decreases visibility for a driver; anda gaze detector that detects a gaze of a driver, wherein when the visibility decreasing region is detected but the display of the driving state does not overlap the visibility decreasing region and the gaze of the driver remains directed to the visibility decreasing region, the display control device displays driving direction information required for driving the vehicle on the visibility decreasing region.The prior art also discloses interval-based cleaning of a windshield based on an interaction of a driver of a vehicle with a cleaning device, in which the windshield is cleaned cyclically by means of the cleaning device on the basis of an interval predefined by the driver. In this case, planar and local non-uniform soiling is cleaned.Furthermore, a camera-assisted cleaning of a windshield of a vehicle is known from the prior art, in which an algorithm is applied by means of a front camera, which algorithm detects a degree of soiling in the visible region of the front camera and accordingly initiates a cleaning carried out by means of a cleaning device. This removes uniformly flat soiling, such as water films, from the windshield.The invention is based on the object of specifying a novel method for detecting soiling on a windshield of a vehicle and a novel method for operating a cleaning device for a windshield of a vehicle.The object is achieved according to the invention bya method for detecting soiling on a windshield of a vehicle, which has the features specified in claim 1, anda method of operating a vehicle windshield cleaner having the features recited in claim 9.Advantageous embodiments of the invention are the subject matter of the dependent claims.The method according to the invention for detecting soiling on a windshield of a vehicle is distinguished in thatin a calibration method, a calibration function is determined for a dirt-free windshield, which represents a relationship between a determined real distance of a respective surrounding object to the vehicle and a determined accommodation of eyes of a driver of the vehicle when looking at this surrounding object, andin an application methodpositions of surrounding objects and distances of the surrounding objects to the vehicle are determined in data of the vehicle environment detected by means of a surrounding environment detection sensor system,in images of the driver's eyes acquired by means of a driver observation camera, a viewing direction of the driver and accommodation of the eyes are determined,determining in a correspondence analysis the surrounding object to which the driver is looking; anda contamination-free windshield is detected using the calibration function if a relation between the distance of the respective surrounding object and the accommodation of the eyes of the driver when looking at this surrounding object is within a predefined setpoint range, and a contamination of the windshield located in the field of vision of the driver is detected if the relation is outside the predefined setpoint range.By means of the present method, it is possible, in particular compared to the camera-based methods known from the prior art, also to detect soiling outside a detection range of a front camera of a vehicle. In this case, the detection correlates with the field of view of the driver of the vehicle on the basis of a coupling of the accommodation of the eyes of the driver and surrounding objects located in the environment of the vehicle. Thus, by means of the method, it is possible to detect contaminants that occur partially on the windshield, such as, for example, salt lines. Such soiling is spatially limited and is generally not located on the entire surface of the windshield. A detection of such soiling is not possible with certainty by means of the method known from the prior art using the front camera, since the partial soiling can be located outside a detection range of the front camera, but can nevertheless lead to a restriction of the vision of the driver in the field of vision of the latter. By means of the present method, such local soiling can be detected very reliably, in particular in the field of vision of the driver, so that, when such soiling occurs, a cleaning device for cleaning the windshield can be activated automatically, for example. The cleaning can thereby comprise wetting the windshield with a cleaning liquid and activating windshield wipers of the vehicle.Thus, the method enables contamination detection based on physical driver parameters.The surrounding objects are detected by means of the surrounding environment detection sensor system, for example cameras, lidar sensors, radar sensors, etc. Since such an environment detection sensor system and also a driver observation camera are already frequently present in vehicles for operating driver assistance functions, no additional hardware is required for carrying out the method. Positions and distances of the surrounding objects to the vehicle and an object classification can be derived, for example, from a dynamic object fusion.There are further advantages of the methodin a possible recurrent activation of the same in the vehicle,in a possible derivation of the calibration function during a real operation of the vehicle,utilizing already known image evaluation methods,in real-time detection of soilingin independence from a light situation present, andin the possibility of detecting a variety of soils, such as salt, slush, sludge, etc.One possible embodiment of the method provides that in the calibration methodpositions of surrounding objects and distances of the surrounding objects to the vehicle are determined in data of the vehicle environment detected by means of the environment detection sensor system,in images of eyes of the driver of the vehicle acquired by means of the driver observation camera, a viewing direction of the driver and an accommodation of the eyes are determined,determining in a correspondence analysis the surrounding object to which the driver is looking; anddetermining the calibration function which reproduces the relationship between the determined real distance of a respective surrounding object to the vehicle and the determined accommodation of eyes of the driver when looking at this surrounding object.Such a determination of the calibration function enables a very good comparison of the relations of the accommodations and distances from the calibration function with the relations which are determined in the application method. Thus, a particularly reliable detection of soiling on the windshield of the vehicle can be realized.A further possible configuration of the method provides that in the correspondence analysis a correspondence between the viewing direction and a position of the respective surrounding object is determined by a course of viewing directions of the driver and a trajectory of the captured surrounding object. Such a determination of the correspondence when following the trajectory is particularly reliable.A further possible embodiment of the method provides that relations of accommodations of the driver's eyes are determined during the detection of the contamination in the application method for distances of different positions of a respective surrounding object along the trajectory and evaluated using the calibration function. Such a determination of the accommodations when following the trajectory enables a particularly reliable detection of soiling on the windshield of the vehicle.A further possible embodiment of the method provides that the application method is carried out again after automatic cleaning of the windshield carried out on the basis of a detection of dirt. It can thus be determined whether the cleaning was successful and, if appropriate, a cleaning of the windshield can be carried out automatically again until the contamination is removed.A further possible embodiment of the method provides that positions and distances of the surrounding objects to the vehicle and / or an object classification are derived from a dynamic object fusion in the calibration method and the application method. On the basis of data of such an object fusion, the positions, distances and the object classification can be determined very reliably.A further possible embodiment of the method provides that the determination, carried out in the correspondence analysis, on which surrounding object the driver directs his gaze is carried out on the basis of intersection recognition of the viewing directions of the driver with the surrounding objects. Such intersection detection can be carried out particularly easily and enables the correspondence analysis to be carried out exactly.A further possible embodiment of the method provides that the check as to whether the relation between the distance of the respective surrounding object and the accommodation of the driver's eyes when looking at this surrounding object is within a predefined desired range is carried out on the basis of a threshold value observation. Such a threshold value observation can be carried out particularly easily and enables an exact determination as to whether the relation lies within the predefined desired range. A corresponding threshold value can be determined realistic in test drives, applications, etc., for example.In the method according to the invention for operating a cleaning device for a windshield of a vehiclechecking whether the windshield is soiled is present in the field of vision of the driver by means of the aforementioned method, andIn the presence of such a contamination, the cleaning device is automatically activated, which carries out a cleaning of the windshield.The method enables automatic cleaning of the windshield without the need for interaction of the driver with a cleaning device. Thus, distraction of the driver from a traffic situation resulting from manual activation can be avoided and driving comfort for the driver can be increased. Also, dirt on the windshield can be quickly and reliably removed, so that a vision of the driver can be improved and traffic safety can be increased.One possible configuration of the method provides that the application method is carried out again after the automatic cleaning of the windshield.It can thus be determined whether the cleaning was successful and, if appropriate, a cleaning of the windshield can be carried out automatically again until the contamination is removed.Exemplary embodiments of the invention are explained in more detail below with reference to drawings.The following are shown: FIG. 1 schematically shows a sequence of a method for operating a cleaning device of a windshield of a vehicle, FIG. 2 schematically shows a visualization of a first method step of the method according to FIG. 1, FIG. 3 schematically shows a visualization of a second method step of the method according to FIG. 1, FIG. 4 schematically shows a visualization of a third method step of the method according to FIG. 1, FIG. 5 schematically shows a visualization of a fourth method step of the method according to FIG. 1, FIG. 6 schematically shows a visualization of a fifth method step of the method according to FIG. 1, FIG. 7 schematically shows a visualization of a sixth method step of the method according to FIG. 1, FIG. 8 schematically shows a visualization of a seventh method step of the method according to FIG. 1, FIG. 9 schematically shows a visualization of an eighth method step of the method according to FIG. 1, FIG. 10 schematically shows a visualization of a ninth method step of the method according to FIG. 1 ; and FIG. 11 schematically shows a visualization of a tenth method step of the method according to FIG. 1.Corresponding parts are provided with the same reference numerals in all figures.FIG. 1 schematically shows a sequence of a method for operating a cleaning device 1 (illustrated in FIG. 11 ) of a windshield 2 (illustrated in FIG. 4 ) of a vehicle 3 (illustrated in FIG. 4 ). FIGS. 2 to 11 show visualizations of method steps VS1 to VS10 of the method.By means of the method, it is possible to detect contaminants V, which are shown in more detail in FIG. 11 and are distributed locally on the windshield 2, by coupling a driver observation camera 4, which is shown in more detail in FIG. 4, to a detection of surrounding objects O 1 to On, which are present in a vehicle environment and are shown in more detail in FIG. 5. In this case, it is assumed that the driver observation camera 4 and a surrounding environment detection sensor system 5, which is likewise illustrated in more detail in FIG. 4, are calibrated and adjusted accordingly extrinsically and intrinsically and transformation between coordinate systems of the driver observation camera 4 and the surrounding environment detection sensor system 5 is possible.First, in a calibration method, with windshield 2 free of dirt at least in a field of view of driver F, images B 1 to Bz of a driver F of vehicle 3, in particular images B 1 to Bz of driver F, are captured by means of driver observation camera 4 in a first method step VS 1, as visualized in FIG. 2.In a second method step VS 2, as visualized in FIG. 3, a movement of eyes A 1, A 2 of the driver F is detected by evaluating the images B 1 to Bz. Here, an accommodation AK of eyes A 1, A 2 shown in more detail in FIG. 8 is determined in the images B 1 to Bz, which substantially describes a change of a lens of a respective eye A 1, A 2, whereby a refractive power of the eye A 1, A 2 is adapted such that objects are focused in space. Here, the accommodation AK correlates with a distance d from a surrounding object O 1 to On focused by the driver F. In this case, it is possible for the accommodation AK to be derived from video images by means of tracking the eyes A 1, A 2, also referred to as eye tracking.In a third method step VS 3, as visualized in FIG. 4, viewing directions BR 1 to BRm of the driver F, also referred to as gauze, are determined starting from a head position KPOS of the driver F.In a fourth method step VS 4, as visualized in FIG. 5, a three-dimensional detection of the vehicle environment is carried out by means of the environment detection sensor system 5, in which a depth of the vehicle environment in relation to a vehicle origin is determined. For this purpose, the environment detection sensor system 5 comprises appropriately designed sensors, for example cameras, lidars, radars, etc. On the basis of data detected by means of the environment detection sensor system 5, positions of environment objects O 1 to On and distances d of the environment objects O 1 to On to the vehicle 3 shown in more detail in FIG. 8 are determined. The surrounding objects O 1 to On can also be classified and thus differentiated, for example, between passenger cars, trucks, pedestrians, cyclists, etc. Furthermore, a respective trajectory of the surrounding objects O 1 to On is determined on the basis of the data detected by means of the surrounding environment detection sensor system 5.In a fifth method step VS 5, as visualized in FIG. 6, a transformation of a coordinate system KS 1 of the environment detection sensor system 5, that is to say an environment model, into a coordinate system KS 2 of the driver observation camera 4 is carried out.In a sixth method step VS 6, as visualized in FIG. 7, a correspondence analysis is carried out in which it is determined which surrounding object O 1 to On the driver F is aiming at. In this case, a correspondence between a respective viewing direction BR 1 to BRm and a position of the respective surrounding object O 1 to On is determined by a course of viewing directions BR 1 to BRm of the driver F and a trajectory of the captured surrounding object O 1 to On. That is, analysis of correspondence between an object trajectory and a gaze direction history of the driver F is performed. In this case, it is determined which of the captured surrounding objects O 1 to On the driver F focuses. This can be done, for example, on the basis of intersection recognition of the viewing directions BR 1 to BRm with the surrounding objects O 1 to On.In a seventh method step VS 7, as visualized in FIG. 8, a calibration function f is determined, which represents a relation between a determined real distance d of a respective surrounding object O 1 to On to the vehicle 3 and a determined accommodation AK of the eyes A 1, A 2 of the driver F when looking at this surrounding object O 1 to On as a function of a polar angle PW. The calibration function f is a three-dimensional function which reproduces the relation between real depth and accommodation AK of the eyes A 1, A 2 of the driver F.The calibration method described above with method steps VS 1 to VS 7 can be carried out in real operation of vehicle 3, for example in hazardous situations in which it is known that driver F focuses a surrounding object O 1 to On lying ahead. Such a hazardous situation can be, for example, emergency braking in the case of a traffic situation in a road intersection region.In an application method, i.e. for detecting soiling V during the driving operation of the vehicle 3, the aforementioned method steps VS 1 to VS 6, except for the method step VS 7, are likewise carried out. That is, itthe positions of surrounding objects O1 to On and distances d of the surrounding objects O1 to On to the vehicle 3 are determined in data of the vehicle environment detected by means of the environment detection sensor system 5,the viewing direction BR 1 to BRm of the driver F and the accommodation AK of the eyes A 1, A 2 of the driver F are determined in images B 1 to Bz of the eyes A 1, A 2 of the driver F captured by means of the driver observation camera 4; anddetermining in the correspondence analysis which surrounding object O1 to On the driver F is looking at.In the application method, a comparison between detected accommodation AK and real object depth is thus carried out in the real operation of vehicle 3.Subsequently, in an eighth method step VS 8, as visualized in FIG. 9, when a correspondence between the viewing direction BR 1 to BRm of the driver F and a position of an environmental object O 1 to On has been determined in the correspondence analysis, accommodations AK 1 to AK 8 of the eyes A 1, A 2 of the driver F along the trajectory of the corresponding environmental object O 1 to On are adjusted using the calibration function f. This means that a congruence of the relationship between the accommodations AK 1 to AK 8 of the eyes A 1, A 2 of the driver F at a respective distance d 1 to d 8 of the surrounding object O 1 to On with the relationship stored in the calibration function f is checked at such a distance d 1 to d 8.It is assumed here that, in the case of a windshield 2 which is free of dirt in the field of vision of the driver F, the accommodation AK, AK 1 to AK 8 follows the trajectory thereof at least substantially relative to the object depth, that is to say the respective distance d, d 1 to d 8 of the surrounding object O 1 to On from the vehicle 3. However, if contaminants V are present on the windshield 2 in the field of vision of the driver F, it is to be assumed that the driver F attempts to focus the surrounding object O 1 to On again. In this case, the driver F attempts to view the scene, but is at least partially limited by the contamination V, since he focuses the contamination V instead of the scene. This results in a divergence between the distance d, d 1 to D 8 derived from the accommodation AK and the distance d, d 1 to d 8 derived from the data of the environment detection sensor system 5. That is, the distance d, d 1 to d 8 derived from the accommodation AK is compared with the distance d, d 1 to d 8 obtained from the data acquired by the environment detection sensor array 5.That is, a difference ΔAK, ΔAK1 to ΔAK8 of the accommodation AK, AK1 to AK8 of the eyes A1, A2 determined in the application method and the accommodation AK, AK1 to AK8 originating from the calibration function f is determined at the same distance d, d1 to d8.In a ninth method step VS 9, as visualized in FIG. 10, a contamination-free windshield 2 is detected using the calibration function f if a relation between the distance d of the respective surrounding object O 1 to On and the accommodation AK of the eyes A 1, A 2 of the driver F when looking at this surrounding object O 1 to On is within a predefined setpoint range. A contamination V of the windshield 2 located in the field of vision of the driver F, on the other hand, is detected when the relationship lies outside the predefined setpoint range.A threshold value observation is carried out, wherein a threshold value SW of the difference ΔAK between the accommodations AK is determined from test drives, applications, etc. and is predefined accordingly. If the difference ΔAK exceeds this threshold value SW, for example in at least three successive frames of the images B 1 to Bz captured by means of the driver observation camera 4, a divergence is detected between a scenario depth, that is to say the corresponding distance d, d 1 to d 8 of the position of the respective surrounding object O 1 to On, and the accommodation AK of the eyes A 1, A 2 of the driver F determined from the images B 1 to Bz, in which the relation is outside the predefined setpoint range on account of contamination V in the field of vision of the driver F.If dirt V in the field of view of the driver F is detected in this way, the cleaning device 1, which performs cleaning of the windshield 2, is automatically activated in a tenth method step VS 10, as visualized in FIG. 11.As shown in FIG. 11, the driver F, according to the representations of the first two images BI 1, BI 2, looks through a dirt-free region of the windshield 2 at surrounding objects O 1, O 2, which are located at a distance d 1, d 2 from the vehicle 3. In this case, the difference ΔAK between the accommodations AK does not exceed the threshold value SW, so that a contamination-free region of the windshield 2 is detected in the region of the respective viewing direction BR 1 to BRm of the driver F.If, according to the representations of the images BI 3 to BI 5, the driver F looks through a region of the windshield 2 provided with a contamination V, for example through an ice covering, at surrounding objects O 3 to O 5 which are located at a distance d 3 to d 5 from the vehicle 3, the difference ΔAK of the accommodations AK exceeds the threshold value SW. Thus, the dirt V of the windshield 2 is detected in the region of the respective viewing direction BR 1 to BRm of the driver F and the cleaning device 1 is automatically activated.When the purification has been carried out, the application procedure is carried out again.

Claims

Method for detecting soiling (V) on a windshield (2) of a vehicle (3), wherein - in a calibration method, in the case of a soiling-free windshield (2), a calibration function (f) is determined which reproduces a relation between a determined real distance (d, d1 to d8) of a respective surrounding object (O1 to On) to the vehicle (3) and a determined accommodation (AK) of eyes (A1, A2) of a driver (F) of the vehicle (3) when looking at this surrounding object (O1 to On), and - in an application method - in data of the vehicle environment recorded by means of a surrounding detection sensor system (5), positions of surrounding objects (O1 to On) and distances (d, d1 to d8) of the surrounding objects (O1 to On) to the vehicle (3), - in images (B1 to Bz) of the eyes (A1, A2) of the driver (F) acquired by means of a driver observation camera (4), a viewing direction (BR1 to BRm) of the driver (F) and the accommodation (AK) of the eyes (A1, A2) are ascertained, - in a correspondence analysis, it is ascertained to which surrounding object (O1 to On) the driver (F) directs his gaze, and - using the calibration function (f), a contamination-free windshield (2) is detected if a relation between the distance (d, d, A2) is determined, d1 to d8) of the respective surrounding object (O1 to On) and the accommodation (AK) of the eyes (A1, A2) of the driver (F) when looking at this surrounding object (O1 to On) is within a predefined target range, and a contamination (V) of the windshield (2) located in the field of view of the driver (F) is detected if the relation is outside the predefined target range.Method according to Claim 1, wherein in the calibration method - positions of surrounding objects (O1 to On) and distances (d, d1 to d8) of the surrounding objects (O1 to On) to the vehicle (3) are determined in data of the vehicle environment detected by means of the environment detection sensor system (5), - a viewing direction (BR1 to BRm) of the driver (F) and an accommodation (AK) of the eyes (A1, A2) are determined in images (B1 to Bz) of the driver (F) of the vehicle (3) detected by means of the driver observation camera (4), - a correspondence analysis is determined, to which surrounding object (O1 to On) the driver (F) directs his gaze is determined, and - the calibration function (f) is determined which reproduces the relation between the determined real distance (d, d1 to d8) of a respective surrounding object (O1 to On) to the vehicle (3) and the determined accommodation (AK) of eyes (A1, A2) of the driver (F) when looking at this surrounding object (O1 to On).Method according to Claim 1 or 2, wherein in the correspondence analysis a correspondence between the viewing direction (BR1 to BRm) and a position of the respective surrounding object (O1 to On) is determined by a profile of viewing directions (BR1 to BRm) of the driver (F) and a trajectory of the captured surrounding object (O1 to On).Method according to Claim 3, wherein relations of accommodations (AK) of the eyes (A1, A2) of the driver (F) are determined during the detection of the contamination (V) in the application method for distances (d, d1 to d8) of different positions of a respective surrounding object (O1 to On) along the trajectory and are evaluated using the calibration function (f).Method according to one of the preceding claims, wherein the application method is carried out again after automatic cleaning of the windshield (2) carried out on the basis of a detection of a contamination (V).Method according to one of the preceding claims, wherein in the calibration method and the application method positions and distances (d, d1 to d8) of the surrounding objects (O1 to On) to the vehicle (3) and / or an object classification are derived from a dynamic object fusion.Method according to one of the preceding claims, wherein the determination carried out in the correspondence analysis as to which surrounding object (O1 to On) the driver (F) directs his gaze is carried out on the basis of intersection recognition of the viewing directions (BR1 to BRm) of the driver (F) with the surrounding objects (O1 to On).Method according to one of the preceding claims, wherein the check as to whether the relation between the distance (d, d1 to d8) of the respective surrounding object (O1 to On) and the accommodation (AK) of the eyes (A1, A2) of the driver (F) when looking at this surrounding object (O1 to On) is within a predefined desired range is carried out on the basis of a threshold value observation.Method for operating a cleaning device (1) for a windshield (2) of a vehicle (3), wherein - a method according to one of the preceding claims is used to check whether a contamination (V) of the windshield (2) is present in the field of vision of the driver (F), and - the cleaning device (1) which carries out a cleaning of the windshield (2) is automatically activated if such a contamination (V) is present.Method according to claim 9, characterised in that the application method is carried out again after the automatic cleaning of the windscreen (2).

Images