DE102017222183A1 - Method and system for monitoring a plurality of environmental sensors of a motor vehicle - Google Patents

Abstract

The invention relates to a method (10) for monitoring a plurality of environmental sensors (US1, US2, USn) of a motor vehicle (12). At first, at least one object in an environment of the motor vehicle (12) is detected with an environmental sensor of the plurality of environmental sensors (US1, US2, USn). Subsequently, the environmental sensor is assigned to a detected from the at least one object feature as the at least one object detecting environmental sensor from the plurality of environmental sensors (US1, US2, USn). In a further method step, the motor vehicle (12) is localized on the basis of the at least one detected object in a digital map (14), wherein the at least one object is assigned a reference object on the digital map (14), and a decalibration of the environmental sensor of the plurality of environmental sensors (US1, US2, USn) by evaluating a Dekalibrierungskritium taking into account the detected at least one object by the detected object associated with the environmental sensor and the reference object determined.

Description

The invention relates to a method for monitoring a plurality of environmental sensors of a motor vehicle. In the method, at least one object in an environment of the motor vehicle is detected with an environmental sensor of the plurality of environmental sensors. In other words, by means of the environmental sensor, sensor data from the surroundings of the motor vehicle, that is to say the vehicle surroundings, can be recorded for at least one object. The invention also relates to a system for monitoring a plurality of environmental sensors of a motor vehicle.

Current driver assistance systems and, in particular, autonomous driving depend very much on a large number of different sensors for sensing the vehicle environment. Deviation from the expected function of these sensors often leads to disruptions of the respective driver assistance functions. This deviation can be caused, for example, during driving of the motor vehicle by vibrations. The disturbance of the respective driver assistance function manifests itself either in a degradation, a complete failure of the driver assistance function or even a malfunction of the driver assistance function. A malfunction of the driver assistance function, for example, in the case of an emergency brake assist as a driver assistance system lead to an accident when the actual obstacle was perceived in another position. Such perception errors can be caused, for example, by a decalibration of the environment sensors or environmental sensors.

From the general state of the art, different methods for improving a calibration of a sensor of a motor vehicle or locating the motor vehicle are known.

For example, describes the DE 10 2015 113 911 A1 a system for calibrating an on-board sensor of a motor vehicle. The motor vehicle has a control system capable of performing at least some of the tasks related to the calibration procedure. To calibrate the sensor, vehicle state data is collected and navigation map data is provided during operation of the motor vehicle. A current calibration factor is calculated for the on-board sensor based on the vehicle condition data and the navigation map data. The on-board sensor may be calibrated in response to the current calibration factor.

From the DE 10 2010 018 088 A1 is a procedure for checking a calibration. In the context of a car-to-car communication, measurement data from surroundings sensors of a vehicle that relate to a specific object are transmitted to another vehicle. There, environmental sensors also detect the object and a comparison is possible on the basis of which it can be determined whether the calibration in the vehicle is correct. Thereupon, a warning signal can be emitted or even a new calibration based on the measured data.

Such methods for checking a calibration of environmental sensors are particularly expensive to operate. In addition, the corresponding methods provide a rough estimate of a correct calibration of the corresponding environment sensors, whereby a degree of reliability of these methods is particularly low.

Instead of calibrating sensors for safe operation of the motor vehicle, the US 5,394,333 A a method for correcting a GPS position in a hybrid navigation system. A hybrid vehicle navigation system uses both GPS and dead center sensors that operate in parallel to produce a detected and a calculated position. The detected and calculated positions are compared to a road database to determine separate correction factors for the correction of the detected and calculated positions. The system uses either the sensed position or the calculated position to navigate, depending on which position has the smaller correction factor.

Such a system is particularly expensive to operate and depends on the reliability of the sensors as well as the road database.

The object of the present invention is therefore to provide a method for monitoring a plurality of environmental sensors of a motor vehicle, which can be operated particularly easily and yet particularly reliably, in particular with regard to detecting a decalibration of the environmental sensors.

This object is achieved by a method and a system for monitoring a plurality of environmental sensors having the features of the independent patent claims. Advantageous embodiments with expedient and non-trivial developments of the invention are specified in the dependent claims.

In the method for monitoring a plurality of environmental sensors of a motor vehicle, at least one object in an environment of the motor vehicle is first detected with an environmental sensor of the plurality of environmental sensors. In other words, the environment can be shared by more than one Environmental sensors or environment sensors are perceived, wherein at least one environmental sensor detects the at least one object. Particularly preferably, several objects can also be detected by the environmental sensor.

When detecting the at least one object, the environmental sensor can record at least one feature of the object, in particular for identifying the object. Preferably, the data acquired by the environmental sensors are processed. The environmental sensors may, for example, comprise one or more cameras and / or one or more radars and / or one or more lidars and / or one or more laser scanners and / or one or more ultrasound sensors.

In particular, the object may be a static or local object, such as a lane marker or guardrail, or a traffic light or a bollard, or a shield post or lamp post. By means of the plurality of environmental sensors, a plurality of, in particular different, objects can be detected or perceived. In other words, with the aid of the environmental sensors and the interpretation of the sensor data of the environmental sensors, the vehicle environment can be measured and / or descriptions of the objects, in particular by their respective detected characteristics, can be generated.

Among other things, the idea underlying the invention consists, inter alia, in that on the one hand certain objects can only be perceived and / or classified by one sensor, in particular environmental sensor, and on the other hand any object contains information - a detected feature of the object - from which sensor it was perceived. Among other things, the invention is characterized in that the environmental sensor is assigned from the plurality of environmental sensors on the basis of a feature detected with respect to the at least one object as the environmental sensor detecting the at least one object. In other words, by the detected feature of the object, the corresponding environmental sensor is identified as the environment sensor detecting the object.

In a further method step, the motor vehicle is localized on the basis of the at least one detected object in a digital map, wherein the at least one object is assigned a reference object on the digital map. In other words, the at least one detected object can form a basis for locating or positioning the motor vehicle in the digital map. For example, the detected or perceived object and, in particular, information derived therefrom and the reference object from the digital map and, in particular, information of the digital map derived therefrom, are compared with one another for positioning. Based on the detected object and the reference object assigned to the detected object, the motor vehicle is positioned in particular relative to the digital map, in particular the reference object of the digital map. The more objects are detected by the environmental sensor, the more accurate the placement or positioning or localization of the motor vehicle in the digital map can be done.

First, a rough estimate of a current position of the motor vehicle with the aid of current GPS data of the motor vehicle can be made. For a more accurate estimation of the position of the motor vehicle, the localization of the motor vehicle to the detected object on a digital map can take place with the aid of the assignment of the reference object.

Finally, in the method, a decalibration of the environmental sensor of the plurality of environmental sensors is determined by evaluating a decalibration criterion taking into account the detected at least one object by the environmental sensor and the reference object assigned to the detected object. In other words, to detect a decalibration of the environmental sensor, the information of the digital map, particularly a high accuracy map, is needed because the object detected by the environmental sensor or the respective objects detected by the environmental sensors are matched with those elements of the digital map during the localization step can.

In this case, individual method steps of the method can preferably be carried out fully automatically by a control unit of the motor vehicle, which can be provided either as an independent control unit, as part of a complete control unit network or integrated into other control units. All environmental sensors preferably provide the control unit with its current measurements, ie the current sensor data. In this case, the provision can take place in different ways, for example directly via private bus connections, indirectly via the central vehicle bus, wherein a communication via wireless connection, in particular a radio connection, other discrete connections or via internal ECU bus systems is conceivable.

With the method for monitoring a plurality of environmental sensors of the motor vehicle, a decalibration of the environmental sensor can be detected in a particularly simple and reliable manner, and thus early or timely responds. This increases the safety and robustness of driver assistance systems and thereby increases safety in driving the motor vehicle, in particular in an autonomous driving of the motor vehicle, in traffic.

The invention also includes the combinations of the described embodiments.

In the method, a plurality of objects are particularly preferably detected by the respective environmental sensors of the motor vehicle. Each of the detected objects is assigned at least one environmental sensor of the plurality of environmental sensors, in particular based on a detected by means of the environmental sensor to the respective object feature. Additionally or alternatively, a current driving situation of the motor vehicle, in particular on the basis of the detected objects, can be determined by the control device. Additionally or alternatively, a driving trajectory, in particular based on the detected objects, determined and controlled the motor vehicle along this Fahrtrajektorie or out.

An advantageous embodiment provides that when the decalibration criterion is evaluated, a deviation of the at least one object from the reference object assigned to the detected object is determined. In other words, a deviation of the object sensed by the environmental sensor from the corresponding object of the digital map - the reference object - can be assigned. In particular, each detected object is assigned a world-fixed coordinate, as a result of which the reference object of the digital map can be assigned to this object. By "deviation" is meant in particular a geographical deviation or a distance between the detected object and the reference object. For example, a deviation between 1 cm and 1 m can be detected. In particular, by Dekalbirierung is meant that the environmental sensor detects the object at a position deviating from an actual position. By determining the deviation can be concluded in a particularly simple and reliable manner to a decalibration of the environmental sensor.

An advantageous development provides that the decalibration of the environmental sensor is detected when the deviation exceeds a predetermined threshold value. In other words, the decalibration criterion is met when the determined or sensed value of the deviation exceeds the predetermined threshold. In particular, if this deviation exceeds a certain value, this is an indication of an environmental sensor error caused by decalibration of the environmental sensor.

Advantageously, further objects are detected by the environmental sensor, wherein, when a number of predetermined deviations which exceed the predetermined threshold are exceeded in the detection of the further objects to the other objects associated reference objects, a decalibration of the environmental sensor is detected. In this case, each of the detected objects is preferably assigned to the environmental sensor, in particular based on the features of the objects detected by the environmental sensor. If, for example, for the environmental sensor, in particular over a predetermined period of time, several indications of an exceeding of the specific value of the deviation occur, then a decalibration of the environmental sensor can be concluded.

In an advantageous manner, the additional objects are detected by the environmental sensor during operation of the motor vehicle, that is to say during a driving operation. On the basis of each of the other detected objects, the motor vehicle is localized in the digital map, each detected further object each having a further reference object of the digital map is assigned. When evaluating the decalibration criterion, a deviation of the currently detected object from the reference object assigned to this object is determined in each case. Depending on how often the individual specific deviations exceed the predetermined threshold, a decalibration of the environmental sensor is present or not. For example, the crossing of the threshold value can be counted. In other words, an accumulation of the deviations of the environmental sensor can be determined. If the counted number exceeds the predetermined number, then the decalibration criterion is met. For example, the controller may be configured to count the number. The detection of the number of deviations or the counting of the number can also be referred to as accumulation.

In this case, in particular, the specific environmental sensor can be detected via the assignment of the object to its perceptible environmental sensor. Thus, if the motor vehicle, for example, moves along a predetermined driving trajectory or predetermined driving route and if the environment sensor detects successive objects during the travel trajectory or the predetermined trajectory, a deviation from its respective reference object is detected, in particular for each of these detected objects. If the deviations exceed the predetermined threshold value and the number of deviations exceeding this predetermined threshold exceeds a predetermined number, that is to say a predetermined amount of deviations, the decalibration criterion is met.

Additionally or alternatively, it can be provided that the determination of a number of deviations by a Bayesian estimation method or another estimation method for accumulation is used.

The described method can be carried out for each of the further environmental sensors of the plurality of environmental sensors of the motor vehicle, in particular simultaneously.

The invention is based on the finding that for the detection of a decalibration relatively many perceived objects in the vehicle environment, ie the surroundings of the motor vehicle, are required. Otherwise, erroneous objects could influence the position estimate, so that deviations would no longer be detectable. In particular, since this condition can not always be met, a further advantageous embodiment provides that data detected for determining the decalibration of the environmental sensor, which comprise the object detected by the environmental sensor and / or the position of the motor vehicle detected in the localization in the digital map , to a server device, wherein the server device is adapted to compare the acquired data with the server device provided data. The server device is in particular a vehicle-external server device, which can also be referred to as a back-end server. So that the data of the motor vehicle can be transmitted or transmitted to the server device, the motor vehicle can have, for example, a communication unit. In other words, information regarding the vehicle condition and / or the vehicle environment can be communicated to the server device. The communication unit is configured to communicate with the server device via a wireless connection, such as a WLAN connection, or to exchange data. Particularly preferably, the communication unit and the server device can communicate bidirectionally.

After the positioning of the motor vehicle, in particular relative to the digital map, the calculated position of the motor vehicle including the perceived object, ie together with the detected object, and the positioning error can be transmitted to the server device. The server device may then be configured to re-locate the motor vehicle in the digital map, taking into account data provided to the server device.

Advantageously, the data provided can be provided by other motor vehicles. In other words, as the data provided perceptions of a vehicle fleet, so the other motor vehicles, provided or the server device are deposited. Thus, the server device can be set up to carry out the localization of the motor vehicle in the map by taking into account the perception of a vehicle fleet.

An advantageous development provides that the provided data, with which the detected data are compared, are assigned to a situation which corresponds to a situation of the situation detected by the acquired data. In other words, the server device can be set up to compare the situation detected by the surroundings sensor of the motor vehicle with an identical or corresponding situation which is stored by the server device, in particular a memory of the server device. For example, if the environmental sensor detects three objects in a predetermined arrangement relative to the motor vehicle, the server device may be configured to compare a corresponding situation in which such a constellation or arrangement of three objects to a motor vehicle exists with the situation detected by the environmental sensor , The detected situation comprises the at least one object detected by the environmental sensor and / or the position of the motor vehicle in the digital map. Depending on whether a positioning error is present, this is transmitted by the detected position of the motor vehicle with the server device. The server device looks at this situation then comparable situations. These situations are provided in particular by other motor vehicles. The data provided by the situations and / or the server device are stored in a memory, in particular a database of the server device.

Advantageously, in evaluating the decalibration criterion, the server device compares the provided data with the acquired data and determines therefrom a deviation between the data, wherein for detecting the decalibration the deviation is compared with a predetermined threshold. On the one hand, a deviation is determined for the recorded data as well as for the provided data. If the deviation is greater than a deviation of the data provided, then there is a decalibration. Particularly preferably, the decalibration of the environmental sensor is detected when the deviation exceeds a predetermined threshold. In other words, if a predetermined deviation from the value calculated by the environmental sensor is detected, there is an indication of an error caused by a decalibration. Particularly preferably, a decalibration of the environmental sensor is not detected until a number of detected deviations exceeds a predetermined number.

When evaluating the decalibration criterion, in each case a deviation of the currently detected object from the reference object of the data provided for this object and / or a deviation of the position of the motor vehicle detected during the localization with a reference position of the provided data is determined. Depending on how often the individual specific deviations exceed the predetermined threshold, a decalibration of the environmental sensor is present or not. If the counted number exceeds a predetermined number of deviations, then the decalibration criterion is met.

A further advantageous embodiment provides that, if the deviation is below the predetermined threshold value, or if no provided data of the server device are stored, the detected data of the server device are stored. In other words, if the deviation is below the predetermined threshold, the data provided to the server device may be updated by the acquired data.

The invention also includes a system for monitoring a plurality of environmental sensors of a motor vehicle. The system includes a plurality of environmental sensors, wherein at least one environmental sensor of the plurality of environmental sensors is configured to detect at least one object in an environment of the motor vehicle. Furthermore, the system comprises a control device which is set up to associate the environmental sensor with the aid of the environmental sensor by means of a feature detected with respect to the at least one object to the environmental sensor from the plurality of environmental sensors that detects the object. Furthermore, the control unit is configured to locate the motor vehicle on the basis of the at least one detected object in a digital map, wherein the at least one object is assigned a reference object on the digital map. Finally, the control unit is configured to determine a decalibration of the environmental sensor of the plurality of environmental sensors by evaluating a decalibration criterion taking into account the detected at least one object by the environmental sensor associated with the detected object and the reference object.

Particularly preferably, the digital card is designed as a high-precision card, which can also be referred to as an HD card.

The maps used for navigation tasks contain in particular information such as a course of roads and / or the presence of intersections and / or speed limits and / or overtaking bans and / or a number of lanes and / or points of interest (short POls) and / or transitions between urban and / or rural regions and / or bridges and / or tunnels and / or road classes and / or curvature and / or rise of roads and / or heights. The position of the respective information is reproduced in particular with an accuracy, in particular relative accuracy, of 1 m to 3 m or with an accuracy, in particular absolute accuracy, of 5 m by 25 m. Maps map the environment with such accuracy are also referred to as SD cards in particular.

The high-precision card differs from the SD card in that the positions of the imaged map elements have a higher accuracy, in particular a higher relative accuracy and / or absolute accuracy, and on the other hand the map contains more information in particular. The information includes, for example, an illustration of the course of lanes and / or their course in intersections and / or the type of separation between lanes. Additionally or alternatively, information may be included which may be of importance for certain aspects of autonomous driving, in particular the geometric description of post-like objects, such as posts of traffic lights or lanterns or traffic signs or logs or guide posts, and / or road signs of traffic signs and / or road markings, such as traffic signs or arrows or texts or stop lines and / or stop lines and / or dashed and / or solid lines between lanes and / or restricted areas, and / or walls and / or barriers along the road, such as House walls and / or tunnel entrances and / or crash barriers and / or concrete protection walls, and / or traffic lights and / or curbs. Additionally or alternatively, highly accurate maps may include a description of predetermined elements or objects in the traffic and their meaning, such as the construction of traffic lights and / or the meaning of traffic signs.

The invention also includes further developments of the system according to the invention, which have features as they have already been described in connection with the developments of the method according to the invention. For this reason, the corresponding developments of the system according to the invention are not described here again.

• 1 eine schematische Darstellung eines Ablaufdiagramm eines Verfahrens zum Überwachen mehrerer Umgebungssensoren eines Kraftfahrzeugs;
• 2 eine schematische Darstellung einer erfassten Dekalibrierung eines Umgebungssensors; und
• 3 eine schematische Darstellung einer weiteren erfassten Dekalibrierung eines Umgebungssensors.

In the following, embodiments of the invention are described. This shows:

• 1 a schematic representation of a flowchart of a method for monitoring a plurality of environmental sensors of a motor vehicle;
• 2 a schematic representation of a detected decalibration of an environmental sensor; and
• 3 a schematic representation of another detected decalibration of an environmental sensor.

The exemplary embodiments explained below are preferred embodiments of the invention. In the exemplary embodiments, the described components of the embodiments each represent individual features of the invention, which are to be considered independently of one another, which each further develop the invention independently of one another and thus also individually or in a different combination than the one shown as part of the invention. Furthermore, the described embodiments can also be supplemented by further features of the invention already described.

In the figures, functionally identical elements are each provided with the same reference numerals.

In 1 is the procedure 10 to monitor multiple environmental sensors US1 . US2 . USn shown schematically in a flow chart. In the process 10 is through the multiple environmental sensors US1 . US2 ... USn an environment or environment of a motor vehicle 12 recorded or perceived. The multiple environmental sensors US1 to USn are on the motor vehicle 12 arranged and on the environment of the motor vehicle 12 directed. The environmental sensors US1 For example, USN have one or more cameras and / or one or more radars and / or one or more lidars and / or one or more laser scanners and / or one or more ultrasound sensors.

When sensing or sensing the environment, an environment sensor senses the multiple environmental sensors US1 to USn one or more objects in the environment of the motor vehicle 12 detected. Preferably, the motor vehicle moves 12 during a journey of the motor vehicle 12 along a predetermined travel trajectory or route along which the at least one environmental sensor detects one or more objects. Particularly preferably, the motor vehicle moves 12 autonomous or at least partially automated. Each of the environmental sensors US1 Up USn may be one or more objects in the environment of the motor vehicle 12 to capture. In other words, the environment may be different from the environmental sensors US1 to USn and their data in corresponding procedural steps V1 . V2 . Vn are processed.

As part of this processing, so the processing steps V1 to Vn , there is an extraction of features of the environment, which are important for further processing. Here, an assignment of the environmental sensor on the basis of a feature detected to the at least one object as the environmental sensor detecting the object from the plurality of environmental sensors US1 to USn respectively. The detected object is thus an environmental sensor of the plurality of environmental sensors US1 to USn assigned or assigned. In the processing steps V1 to Vn can be filtered out, which environmental sensor has detected which object. In other words, certain objects are perceptible and / or classifiable only by an environmental sensor. By virtue of this feature, the object thus contains information about which environmental sensor of the plurality of environmental sensors US1 to USn it was perceived.

After the processing steps V1 to Vn can the data of the environmental sensors US1 until USn be merged. In other words, any sensor data or sensor information can be converted by a sensor data fusion SF to a common environment model.

Subsequently, the motor vehicle 12 in a localization step L based on the respective detected object in a digital map 14 localized or positioned, wherein the respective object is a reference object on the digital map 14 assigned. In other words, the sensor data of the environmental sensors US1 to USn or the sensor data fusion SF another set of features - the reference objects - from the digital map 14 for locating the motor vehicle 12 relative to the digital map 14 assigned.

In a further method step, a decalibration of the environmental sensors is determined by evaluating a decalibration criterion taking into account the detected objects by the environmental sensor or associated environmental sensors assigned to the detected objects and the reference objects. When the decalibration criterion is evaluated, a deviation of the respective detected object from the detected object associated with the detected object is determined for each detected object. In other words, in the context of localization of the motor vehicle 12 Deviations detected. With "deviation" is a geographical deviation or a distance between the respective detected object 16 and the respective object associated reference object 18 meant.

If the determined deviation exceeds a predetermined threshold value, then there is a decalibration of the corresponding environmental sensor. In other words, the decalibration criterion is met when the determined deviation exceeds the predetermined threshold. Additionally or alternatively, it may be provided that each deviation is detected, in particular counted. If a number of detected deviations exceed a predetermined number, then there is a decalibration of one or more environmental sensors US1 to USn in front.

The counting of deviations can be done in single accumulation steps A1 . A2 , To be done. For example, the accumulation by an accumulator or an accumulation unit or the control unit of the motor vehicle 12 respectively. This is for each of the environmental sensors US1 to USn every deviation recorded and counted.

2 shows schematically determined deviations of the detected objects 16 from the objects 16 assigned reference objects 18 , In 2 became each of the objects 16 by an environmental sensor during operation of the motor vehicle 12 detected. So the environment sensor has a total of six objects 16 detected. Each of the detected objects 16 became a reference object 18 assigned and determined a deviation or calculated or determined. Overall, here 4 of the 6 detected objects yield 16 from their respective reference object 18 from. Exceeds the number of deviations of an environmental sensor, for example 3 So this may be an indication of a decalibration of the environmental sensor.

If the decalibration criterion for at least one of the environmental sensors is fulfilled, then a decalibration of an environmental sensor of the plurality of environmental sensors is present US1 to USn before, so takes place in a further process step 20 a recalibration of the environmental sensor and / or information is output to a driver. For example, the driver may be informed to visit a workshop because there is a decalibration of an environmental sensor. For example, for this purpose, a warning light flashing or an indication of a display device in an interior of the motor vehicle 12 to be displayed.

Until then, the method needs to monitor the multiple environmental sensors US1 to USn several objects perceived in the vehicle environment. To the method of monitoring the multiple environmental sensors US1 to USn can further improve a decalibration using a server device 22 determines or decalibration criterion using the server device 22 being checked.

This is done via a connection, in particular a radio link 24 , after the localization of the motor vehicle 12 relative to the digital map 14 the calculated position including the perceived objects 16 and the positioning error to the server device 22 transmitted. In other words, the server device becomes 22 by the environmental sensor or the environmental sensors US1 determined until USn recorded situation. The detected situation includes the detected object or objects as well as the location of the motor vehicle determined by the location 12 to the detected object or objects in the digital map 14 , Correspondingly, the deviation from the respectively detected object to the corresponding reference object, that is to say a positioning error, can also be transmitted, if present. Alternatively, an accumulated localization error can be transmitted from all objects.

The server device then leads 22 the location of the motor vehicle 12 in the digital map again. In this case, the server device 22 the location of the motor vehicle 12 by taking into account other motor vehicles, ie the server device 22 provided data.

Alternatively, in the server device 22 for locating the motor vehicle 12 in the digital map also a variant of the geometric hashing be performed. In this case, from the perception of a single further motor vehicle, a description, in particular characteristic and unambiguous description is calculated, which by means of a memory, in particular associative memory, the positioning error of the motor vehicle 12 is assigned. This assignment is made for a complete vehicle fleet and many different positions in the environment. This can ensure a robust association and an average error across all vehicles in the fleet for a particular constellation of objects - the situation - be determined. On the basis of the perceived objects transmitted by a motor vehicle, this association is used to determine the average or average error ØF of the vehicle fleet and the positioning error of the motor vehicle 12 compared.

In the process step 26 it checks if there is a localization error. To do this, the server device compares 22 , in particular a further control device which, by the situation detected on the basis of the acquired data of the environmental sensor, has a situation associated with the data provided.

In relation to 3 The detected situation represents five of those detected by the environmental sensor objects 16 This recorded situation becomes associated with this situation situations S1 . S2 , Sn compared to other motor vehicles. The server device 22 has a database on which the provided data is stored. As already explained, the data provided includes the situations perceived by the other motor vehicles or fleet of motor vehicles S1 . S2 , Sn. Each of these recorded situations S1 to sn is an average mistake ØF assigned. By associating the detected situation with a situation associated with the detected situation, the detected situation may be assigned an error F or the deviation. In a further process step 28 is then checked for an error F or a deviation exists. The resulting error F is compared with the localization error reported by the motor vehicle.

If the deviation or the error exceeds a predetermined value, then a decalibration can be derived therefrom. Will now be a deviation from that of the motor vehicle 12 determined (j), there is an indication of a caused by a decalibration error of the environmental sensor. In the server setup 22 or in the vehicle 12 can then do another analysis of the perceived object 16 be performed to detect the error a particular environmental sensor of the multiple environmental sensors US1 to allocate and / or cumulate to USn and / or upon reaching a threshold or number of deviations (accumulation steps A1 to On ) to conclude a decalibration.

If no error or deviation is detected, the database may be the server device 22 in a further process step 30 be updated and for further checks of localization errors of the database in one step 32 to provide.

Overall, the examples show how the invention provides card and backend-based detection of de-calibrated sensors.

The environment is perceived by sensors and their data processed by appropriate procedures. In the context of this processing often takes place the extraction of special characteristics of the environment, which are important for the further processing (road markings, posts etc.). Any sensor information can be converted in an optional step by a sensor data fusion to a common environment model. The features of the sensors / fusion, along with another set of features from the non-blanked digital maps, form the entrance to a location to position the vehicle relative to the map. Accordingly, as part of the localization deviations are detected and accumulated based on an assignment of features to the sensors in a sensor-specific register. If the value of the accumulator exceeds a predefined threshold, then there is a decalibration and the sensors are notified, this message additionally containing reference objects from the map and the environmental model that can be used for recalibration or recalibration. Additionally or alternatively, the result of the localization including the error can be reported to a backend, where the error is compared against a database. If the error deviates from the database, then a potential decalibration is reported to the vehicle and the procedure is carried out beginning with the accumulation. If there is no deviation of the error from the database, or if there are no reference data in the database, the database is updated with the new data. If necessary, entries in the database are refined to reflect an average error in the database.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102015113911 A1 [0004]
• DE 102010018088 A1 [0005]
• US 5394333 A [0007]

Claims (12)

Method (10) for monitoring a plurality of environmental sensors (US1, US2, USn) of a motor vehicle (12) comprising the steps: - detecting at least one object in an environment of the motor vehicle (12) with an environmental sensor of the plurality of environmental sensors (US1, US2, USn) characterized by - assigning the environmental sensor based on a detected to the at least one object feature as the at least one object detecting environmental sensor from the plurality of environmental sensors (US1, US2, USn); - locating the motor vehicle (12) on the basis of the at least one detected object in a digital map (14), wherein the at least one object is assigned a reference object on the digital map (14), and - determining a decalibration of the environmental sensor of the plurality of environmental sensors (US1 , US2, USn) by evaluating a decalibration criterion taking into account the detected at least one object by the environmental sensor associated with the detected object and the reference object. Method (10) according to Claim 1 , characterized in that a deviation of the at least one object from the detected object associated reference object is determined when evaluating the Dekalibrierungskritieriums. Method (10) according to Claim 2 characterized in that the decalibration criterion of the environmental sensor is met when the deviation exceeds a predetermined threshold. Method (10) according to Claim 3 , characterized in that further objects are detected by the environmental sensor, wherein when in the detection of further objects additionally a number of predetermined deviations which exceed the predetermined threshold is exceeded to the other objects associated reference objects, the Dekalibrierungskriterium the environment sensor is met. Method (10) according to any one of the preceding claims, characterized in that for determining the decalibration of the environmental sensor detected data which detects the detected by the environmental sensor object and / or detected in the localization position of the motor vehicle (12) in the digital map (14 ) are transmitted to a server device (22), wherein the captured data is compared with data provided to the server device (22). Method (10) according to Claim 5 , characterized in that the data provided by other motor vehicles are provided. Method (10) according to Claim 5 or 6 characterized in that the provided data with which the acquired data is compared is associated with a situation corresponding to a situation of the situation detected by the detected data. Method (10) according to one of Claims 5 to 7 , characterized in that for detecting the decalibration a deviation between the data is determined and the determined deviation is compared with a predetermined threshold value. Method (10) according to Claim 8 characterized in that the decalibration criterion of the environmental sensor is met when the deviation exceeds a predetermined threshold and / or a number of the determined deviations exceeds a predetermined number. Method (10) according to Claim 8 or 9 , characterized in that when the deviation is below the predetermined threshold, the detected data of the server device (22) are deposited. A system for monitoring a plurality of environmental sensors (US1, US2, USn) of a motor vehicle (12) comprising: - an environment sensor which is adapted to detect at least one object in an environment of the motor vehicle (12); characterized by , - a control device, which is adapted to associate the environmental sensor based on a detected to the at least one feature to the object detecting as the environment sensor from the plurality of environmental sensors, wherein the control device is further adapted to the motor vehicle (12 ) on the basis of the at least one detected object in a digital map (14), wherein the at least one object is assigned a reference object on the digital map, and wherein - the control device is adapted to decalibrate the environmental sensor of the plurality of environmental sensors (US1, US Pat. US2, USn) by evaluating a decalibration criterion taking into account the detected at least one object by the environmental sensor associated with the detected object and the reference object. System after Claim 11 , characterized in that the digital card (14) is formed as an HD card.

Images