DE102021204660A1 - Method, computer program, data carrier, sensor device and device for checking the suitability of a sensor unit for autonomous driving - Google Patents

Abstract

Exemplary embodiments of the present invention create a method, a computer program, a sensor device, a data carrier and a device for checking the suitability of a sensor unit for autonomous driving. The method includes a stationary arrangement of the sensor unit for monitoring a traffic area. In addition, the method includes acquiring sensor data from the sensor unit relating to the traffic area. The method also includes determining a distance of the sensor unit from at least one road user located in the traffic area using the sensor data. In addition, the method includes comparing the distance determined using the sensor data with a distance threshold value to check the suitability of the sensor unit for autonomous driving.

Description

The present invention relates to a method, a data carrier, a computer program, a sensor device and a device for checking the suitability of a sensor unit for autonomous driving.

Autonomous driving plays an important role in concepts of automotive mobility. Various sensors can be selected and used to recognize the environment in autonomous driving. In order to avoid using unsuitable sensors for autonomous driving or semi-autonomous driving, the sensors can be checked for their suitability for autonomous driving.

Existing concepts for checking the suitability of sensors for autonomous driving envisage moving the sensors in road traffic using vehicles equipped with them in a test operation and recording several traffic situations in order to test the sensors for their suitability for use in an automatically controlled vehicle based on recordings of the traffic situations check vehicle.

The greater the distance covered by the vehicle in test mode, the greater the number of recorded traffic situations that can be used to check the suitability of the sensors. In the case of such validation concepts, the reliability of the validation therefore increases in particular with the distance covered by the vehicle in test operation. Such checking concepts can be time-consuming and costly due to the distance covered by the vehicle.

It can therefore be seen as an object of the present invention to create an improved concept for checking the suitability of a sensor for autonomous driving.

This object can be solved by means of the independent and dependent claims of the present disclosure.

According to a first aspect, the present invention relates to a method for checking the suitability of a sensor unit for autonomous driving. The method includes a stationary arrangement of the sensor unit for monitoring a traffic area. In addition, the method includes acquiring sensor data from the sensor unit relating to the traffic area. The method also includes determining a distance of the sensor unit from at least one road user located in the traffic area using the sensor data. In addition, the method includes comparing the distance determined using the sensor data with a distance threshold value to check the suitability of the sensor unit for autonomous driving.

Autonomous driving (sometimes also referred to as automatic driving, automated driving or piloted driving) is understood to be the control/locomotion of vehicles, mobile robots and driverless transport systems that are at least partially (e.g. in the case of "partially autonomous/partially automated driving") or behave largely autonomously.

The suitability of the sensor unit relates in particular to suitability for one or more applications in the field of autonomous driving. For example, the suitability relates to the suitability of the sensor unit for detecting road users over a sufficiently wide area in order to avoid dangerous situations or accidents involving these road users. In particular, the suitability can depend on whether the sensor unit is suitable for detecting road users in certain traffic situations, such as when entering a lane or on a freeway from a standstill or when driving slowly (for example at walking speed or below a prescribed maximum or recommended speed) at a sufficient distance. Other traffic situations relevant to autonomous driving can arise for an autonomous/automated vehicle when entering an intersection, driving over an intersection, driving past a stationary vehicle, turning right or left with oncoming traffic. The suitability of the sensor unit can therefore relate in particular to detecting oncoming traffic or cross traffic.

Checking the suitability of the sensor unit can also be understood as "validation" of the sensor unit.

The sensor unit comprises, for example, a lidar sensor, a radar sensor, an ultrasonic sensor, a mono and/or camera system suitable for stereo-optical distance measurement, or a combination of these.

A “stationary” arrangement of the sensor unit during the monitoring can be understood in particular as a stationary arrangement with respect to the traffic area. The sensor unit is attached, for example, to a movable or immovable support structure and/or outside of the so-called flowing traffic. Under the flowing traffic understands the expert, in particular, a set of vehicles participating in traffic (moving and temporarily waiting).

The support structure is designed, for example, as a vehicle or vehicle trailer, which is located in stationary traffic while the traffic area is being monitored. Such a vehicle or such a vehicle trailer is parked, for example, while the traffic area with the sensor unit is being monitored. Alternatively, the support structure includes an object or infrastructure object, such as a traffic sign, a building, a bridge or the like, to which the sensor unit can be attached in a stationary manner.

The sensor data can contain image data and/or a three-dimensional image of the traffic area. An example of such a three-dimensional image is a three-dimensional, so-called “point cloud”, which in this context contains, for example, a three-dimensional image of a number of measurement points of the lidar sensor, the ultrasonic sensor and/or the radar sensor. Alternatively, the three-dimensional image can be understood to mean a digital three-dimensional reconstruction of the traffic area, which is generated using such a point cloud, for example.

The traffic area can be parts of a road traffic system, such as sections of a freeway, a country road, a local road, an intersection. Include a freeway entrance, roadway rejuvenation, or the like.

The at least one road user is, for example, a vehicle (for example a passenger car, a truck, a bus, a motorcycle or the like), a bicycle (rider) or a pedestrian. To increase the reliability of the check, several such road users can be detected when monitoring the traffic area. Road users who are approaching the sensor unit can be of particular interest for checking the suitability of the sensor unit, since they potentially interact with a vehicle that is equipped with the sensor unit for autonomous driving during autonomous driving.

The distance can relate to a distance between a sensor or a reference point of the sensor unit and a (measurement) point closest to the sensor unit of an individual road user. For the person skilled in the art it is clear that several distances can be determined when several road users are detected.

The distance indicates, for example, a distance between a photosensitive/light-sensitive sensor element or a reference point of the sensor unit and a measurement or reference point of the road user.

The distance can be determined from the sensor data, for example by means of a processor circuit designed for this purpose, which is coupled to the sensor unit. Furthermore, the processor circuit can compare the distance determined using the sensor data with the threshold value.

The distance threshold value can be understood as a minimum distance, which can be predetermined and, as explained in more detail later, can be dependent on an intended application for the sensor unit, to which the verification of the suitability of the sensor unit is aimed. By comparing the determined distance with the distance threshold value, it can be checked whether the sensor unit can detect the road user at a sufficient distance for the application. The suitability can be confirmed, for example, if the determined distance of the road user is greater than the distance threshold. Conversely, the suitability can at least be doubted or refuted if the determined distance of the road user is less than the distance threshold. The distance threshold is, for example, a predefined, fixed and/or a variable value.

By checking the suitability of the sensor unit according to the method, it can be found out whether the sensor unit is suitable, for example, for use in automated control of a vehicle designed for autonomous driving when entering a freeway, a country road or a local road. Compared to existing concepts for checking the suitability of the sensor unit for autonomous driving, the sensor unit in the method proposed here is stationary, for example arranged in stationary traffic, which can result in cost savings compared to the existing concepts.

In some exemplary embodiments, the distance determined using the sensor data corresponds to a maximum distance of the sensor unit from the at least one road user, at which the at least one road user can be detected using the sensor data.

The maximum distance of the sensor unit from the road user is, for example, a distance of the road user when the road user is detected “for the first time” by the sensor unit when the road user approaches the sensor unit. the Detection of the road user can be understood as recognition of the road user based on the sensor data.

For some applications of the sensor unit, such as for example for controlling autonomously controlled vehicles when starting from the roadside, it is particularly relevant at what maximum distance the road user can be recognized by the sensor unit. When starting from the side of the road, it is relevant, for example, at what maximum distance road users in the traffic behind can be detected.

Taking into account the maximum distance, the suitability of the sensor unit for controlling autonomously controlled vehicles in traffic situations with high relative speeds between autonomously controlled vehicles and other road users, such as when starting from the side of the road, can be checked.

In some exemplary embodiments, the method includes determining a state of motion of the at least one road user, the distance threshold being dependent on the state of motion of the at least one road user.

In some cases, a variable distance threshold can be preferred to a predefined constant distance threshold for checking the sensor unit. The distance threshold value depends, for example, on the state of movement of the road user.

The state of motion denotes, for example, in particular a speed, an acceleration and/or a direction of movement of the road user. The state of motion is determined, for example, using runtime-based or satellite-supported measurement methods. In some exemplary embodiments, the sensor unit is used to determine the state of movement.

The distance threshold behaves, for example, proportionally or exponentially to the determined speed of the road user and increases, for example, proportionally or exponentially with the determined speed of the road user.

When determining the distance threshold, a ratio of the distance threshold to the speed can be chosen such that the distance threshold corresponds to at least one expected reaction distance, braking distance or stopping distance of the road user, which results from the speed of the road user.

In addition, the distance threshold value can be dependent on the acceleration and/or the direction of movement of the road user, ie, for example, dependent on the direction from which the road user is approaching the sensor unit.

A distance threshold value dependent on the state of motion makes it possible in particular to check the suitability of the sensor unit as a function of the state of motion.

In some exemplary embodiments, the distance threshold value is dependent on legal regulations that apply within the traffic area.

The distance threshold value depends, for example, on a speed limit within the monitored traffic area and/or depends on which road users are allowed to drive in the traffic area.

The distance threshold value depends, for example, on legal provisions such as access restrictions for some vehicle types (trucks, passenger cars). Since lorries can have a longer braking distance than passenger cars, the distance threshold can be greater if, for example, the monitored traffic area can be driven by lorries according to the applicable legal provisions, than if a driving ban for lorries applies to the monitored traffic area.

Optionally, the legislation includes speed limits. Speed limits can indicate the speed at which the road user is expected to move within the traffic area and what reaction distance, braking distance and/or stopping distance the road user is therefore expected to have. The distance threshold value can thus be adjusted on the basis of the applicable speed limits within the monitored traffic area and thus, for example, to a stopping distance that the road user can usually expect.

In this way, for example, it is possible to check the suitability of the sensor unit with regard to the legal regulations that apply in the traffic sector.

In some exemplary embodiments, comparing the distance determined using the sensor data with the distance threshold value includes a Comparing a difference between the distance determined using the sensor data and a measurement uncertainty of the sensor unit with the distance threshold value.

The measurement uncertainty is predetermined, for example, by a technical design/design of the sensor unit and indicates, for example, a maximum deviation or a standard deviation of the determined distance from an actual distance of the sensor unit from the road user. The measurement uncertainty results, for example, from a spatial resolution of the sensor unit and/or errors when determining the distance based on the sensor data.

By taking into account the measurement uncertainty when checking the suitability of the sensor unit, the suitability of the sensor unit can be prevented from being incorrectly confirmed due to a measurement deviation or an influence of a measurement error on the determined distance.

In some exemplary embodiments, the method also includes determining a distance of the sensor unit from at least one other road user located in the traffic area using the sensor data. In addition, the method can detect position data from a position sensor which is attached to the at least one other road user. The method also includes, for example, determining the measurement uncertainty of the sensor unit based on the position data and the distance from the at least one other road user determined based on the sensor data.

The other road user can be a test vehicle, which is equipped with the position sensor for detecting the position data. The position sensor is designed, for example, to determine the position data of the test vehicle using satellite-supported position determination.

The position data can include a geographic position of the position sensor, the test vehicle and/or a relative position of the test vehicle with respect to the sensor unit.

Knowing a geographic position of the sensor unit, a distance of the test vehicle from the sensor unit can be determined using the position data of the test vehicle. The measurement uncertainty can result in particular from a difference in the distance of the test vehicle, which was determined using the position data, and the distance determined using the sensor data.

As those skilled in the art will understand, taking measurement uncertainty into account allows for a more reliable assessment of the suitability of the sensor unit.

In some embodiments, the sensor data includes image data.

The verification of the suitability of the sensor unit based on the image data can also be reconstructed/verified visually by one or more persons.

In some exemplary embodiments, the method also includes anonymizing at least part of the image data.

In the event that the sensor unit includes a camera system that is used to monitor the traffic area, in order to comply with data protection regulations, in particular that part of the image data that contains personal information (e.g. a license plate number or faces of people) can be anonymized will. Anonymization provides, for example, for a manipulation of this part of the image data, which makes it unrecognizable. Anonymization provides, for example, for "pixeling" this part of the image data. Alternatively, this part of the image data can be blurred, for example.

In some exemplary embodiments, the method includes acquiring sensor data from a first traffic area and at least one second traffic area.

The suitability of the sensor unit for autonomous driving within different traffic areas can be checked by means of sensor data from different traffic areas. In particular, the suitability of the sensor unit can be checked taking into account different circumstances such as different climatic conditions or different traffic routes within the different traffic areas (and periods of time). When checking the suitability of the sensor unit, different weather conditions (e.g. rain, sunrise, snowfall, thunderstorms and/or hail) and other site-specific conditions such as driving behavior of road users depending on the traffic area can be taken into account.

In some embodiments, the method further includes moving the sensor unit from a first position for capturing the sensor data of the first traffic area to a second Position for acquiring the sensor data of the at least one second traffic area.

Moving the sensor unit provides for example moving the support structure on which the sensor unit is installed. The movement can optionally be automated, semi-automated or manual. The support structure is designed, for example, as a vehicle, which can be moved automatically, partially automatically, or manually by a driver to move the sensor unit.

In some exemplary embodiments, the sensor unit includes a first and at least one second sensor.

The first and the at least one second sensor are, for example, sensors of the same type or of different types. Accordingly, the sensor unit comprises, for example, a number of lidar sensors, radar sensors, ultrasonic sensors, a number of mono and/or camera systems suitable for stereo-optical distance measurement and/or a combination of these. Optionally, the sensor unit can include sensors of the same type, each of which has different specifications such as a different spatial resolution, measurement uncertainty or field of view, or whose field of view is oriented differently.

Thereby the suitability of the sensor unit for different combinations of the sensors mentioned herein can be determined. Alternatively, the suitability of individual sensors can be checked at the same time. As a result, the time required to check several options for implementing the sensor unit can be reduced.

According to a further aspect, the present invention relates to a computer program with a program code which, when it is executed on a computer, a processor, a processor circuit, a control module or a programmable hardware component, carries out a method with the following steps: acquiring sensor data for monitoring a traffic area permanently installed sensor unit; determining a distance of the sensor unit from at least one road user located in the traffic area based on the sensor data; and comparing the distance determined using the sensor data with a distance threshold value to check the suitability of the sensor unit for autonomous driving.

The acquisition of the sensor data includes, for example, acquiring the sensor data from a data carrier for storing the sensor data or acquiring the sensor data from the sensor unit.

According to a further aspect, the present invention relates to a device for checking the suitability of a sensor unit for autonomous driving. The device comprises at least one interface for communication and a processor circuit for executing the computer program disclosed herein.

The processor circuit is available, for example, as a computer processor (CPU = Central Processing Unit), a graphics processor (GPU = Graphics Processing Unit), a computer, a computer system, an application-specific integrated circuit (ASIC = Application-Specific Integrated Circuit), an integrated circuit ( IC = Integrated Circuit), a system on chip (SOC = System on Chip), a programmable logic element, a field programmable gate array with a microprocessor (FPGA = Field Programmable Gate Array) or as a combination of these.

The processor circuit is coupled to the at least one interface, for example, in order to use the interface to acquire sensor data from a sensor unit installed in a stationary manner for monitoring a traffic area. The interface is designed, for example, as an interface to a source of the sensor data. In particular, the interface can be embodied as an interface to the sensor unit or as an interface to a memory/data carrier for storing the sensor data in order to acquire the sensor data from the sensor unit or the memory/data carrier.

According to a further aspect, the present invention relates to a sensor device for checking the suitability of a sensor unit for autonomous driving. The sensor device comprises a carrier structure with which the sensor unit is arranged in a stationary manner while a traffic area is being monitored by means of the sensor unit for generating sensor data of the traffic area. In addition, the sensor device includes a processor circuit. The processor circuit is designed to determine a distance from the sensor unit to a road user in the traffic area using the sensor data and to compare the distance determined using the sensor data with a distance threshold value to check the suitability of the sensor unit for autonomous driving.

The support structure can be an immovable structure of an infrastructure surrounding the traffic area. The support structure is, for example, a bridge, an underpass, a building or a traffic light system. Optionally, the support structure can be a mobile structure to which the sensor unit is attached.

In order to operate the sensor unit, the sensor device (for example the vehicle to which the sensor unit is attached) can also be equipped with a power generator and an air conditioning or cooling system. To protect against theft of the sensor device, it can also be equipped with an alarm system and a direction finder or "GPS tracker".

The processor circuit can optionally be attached to the support structure together with the sensor unit or be present separately from the sensor unit of the support structure. In order to transmit the sensor data to the processor, the sensor unit can be connected or at least connectable to the processor circuit either by cable or wirelessly, by radio.

In some embodiments, the support structure is embodied as a vehicle or vehicle trailer.

The support structure is designed, for example, as a passenger car or truck, which is equipped with a fastening option for attaching the sensor unit. Alternatively, the support structure is designed as a vehicle trailer. In order to arrange the sensor unit in a stationary manner while the traffic area is being monitored, the vehicle or the vehicle trailer can be brought to stationary traffic and, for example, parked.

Compared to immovable exemplary embodiments of the carrier structure, the vehicle or the vehicle trailer can be moved in order to monitor, for example, several different traffic areas in order to check the suitability of the sensor unit for autonomous driving. In order to monitor several different traffic areas, the vehicle can be parked at different locations, for example, in order to check the suitability of the sensor unit for autonomous driving within different traffic areas.

When using the sensor unit in an autonomously or semi-autonomously driving vehicle, in particular an orientation and arrangement of the sensor unit can influence the suitability of the sensor unit for autonomous driving. In order to be able to check the sensor unit under conditions of use, the sensor unit can be attached to the support structure, for example to the vehicle, in such a way as is provided for the use of the sensor unit. The arrangement and alignment of the sensor unit on the support structure can also be changed in order to check the suitability of the sensor unit depending on its alignment and arrangement.

Further exemplary embodiments of the present invention create a data carrier which includes data determined in the method proposed here, in particular sensor data.

The data determined using the method proposed here can in particular also include the distance values generated in the method and information about the suitability of the sensor unit.

The data carrier comprises, for example, a storage medium that can be read mechanically, in particular mechanically, electronically, magnetically, optically and/or magneto-optically, for (non-volatile or volatile) storage of the aforementioned data. In particular, the data carrier can be designed as a mobile data carrier, for example as a hard disk drive, as a digital versatile/video disc (DVD) or as a compact disc (CD).

Using the computer program proposed here, the data ascertained in the method can be read out by machine in order to ascertain or establish the suitability of the sensor unit for autonomous driving.

• 1 ein Verfahren zur Überprüfung einer Eignung einer Sensoreinheit für autonomes Fahren;
• 2 ein Anwendungsbeispiel einer Sensorvorrichtung zur Überprüfung der Eignung der Sensoreinheit für autonomes Fahren;
• 3 ein Prinzip zur Ermittlung einer Messunsicherheit der Sensoreinheit; und
• 4 eine Vorrichtung zur Überprüfung einer Eignung einer Sensoreinheit für autonomes Fahren.

A few examples of exemplary embodiments of the invention are explained in more detail below with reference to the accompanying figures, merely by way of example. Show it:

• 1 a method for checking suitability of a sensor unit for autonomous driving;
• 2 an application example of a sensor device for checking the suitability of the sensor unit for autonomous driving;
• 3 a principle for determining a measurement uncertainty of the sensor unit; and
• 4 a device for checking the suitability of a sensor unit for autonomous driving.

Various embodiments will now be described in more detail and with reference to the accompanying drawings, in which some embodiments are illustrated.

Although example embodiments can be modified and altered in various ways, example embodiments are illustrated in the figures and are described in detail herein. However, it should be made clear that it is not intended to limit exemplary embodiments to the forms disclosed in each case, but rather that On the contrary, example embodiments are intended to cover all functional and/or structural modifications, equivalents, and alternatives falling within the scope of the invention.

The aim of driver assistance systems is in particular to keep a moving vehicle “on the road” without accidents. In addition, the controllers of autonomously driving vehicles are expected to be able, for example, to merge the autonomously driving vehicle into flowing traffic or into a freeway lane when starting from the roadside.

Other possible driving maneuvers include, for example, driving past a stationary vehicle, turning left or right with oncoming traffic and/or traffic coming from behind, or driving straight ahead through an intersection.

The driving maneuvers described above have in common that the autonomously driving vehicle is initially stationary and the driving maneuver requires clearance to set off. Other road users can move at high relative speeds (e.g. 50 km/h in city traffic) in relation to the autonomously driving vehicle, so that for safe control of such driving maneuvers it may be necessary to recognize the road users at a sufficient distance. Before using a sensor unit for controlling an autonomously driving vehicle, it may therefore be necessary to check the suitability of the sensor unit.

In order to check sensors for their suitability for autonomous driving, existing concepts envisage attaching the sensors to be checked to a vehicle and moving the sensors through different traffic areas to record several traffic situations. The costs for such checks of the sensors increase with the distance traveled by the vehicle.

Therefore, an object of the present invention can be seen as creating an improved and in particular more cost-efficient concept for checking the suitability of a sensor unit for autonomous driving.

1 FIG. 1 schematically shows a method 100 for checking the suitability of a sensor unit for autonomous driving.

The method includes a stationary arrangement 110 of the sensor unit for monitoring a traffic area. In addition, the method 100 includes a detection 120 of sensor data of the sensor unit relating to the traffic area. Optionally, the method 100 can include storing the sensor data, in particular to enable further processing of the sensor data independently of the sensor unit and/or outside of the traffic area. The method also includes determining 130 a distance of the sensor unit from at least one road user located in the traffic area based on the sensor data. In addition, the method includes a comparison 140 of the distance determined using the sensor data with a distance threshold value for checking the suitability of the sensor unit for autonomous driving.

For explanation, the method 100 is described in relation to an in 2 illustrated application example of the method 100 for checking a sensor unit 219 is described. However, it should be made clear that exemplary embodiments of the method 100 are not limited to the 2 shown application example are limited.

As in 2 can be seen, the sensor unit 219 comprises a plurality of sensors 220. For the stationary arrangement 110 of the sensor unit 210, it can be attached to a vehicle 210 as a support structure for the sensor unit 219 and the vehicle 210 can be used to monitor a traffic area 250 in stationary traffic, for example in Environment of the monitored traffic area 250 are turned off / parked. In order to simulate the application in an autonomously driving vehicle, the sensors 220 can be attached at the same height as in the application of the sensor unit 210 .

In alternative exemplary embodiments, the support structure can be in the form of a vehicle trailer or an infrastructure object (e.g. a building or traffic sign).

In the application example shown, traffic area 250 includes one lane. Alternatively or additionally, the traffic area 250 can include a parking lot, a cycle path and/or other areas that can be driven over. The vehicle 210 is parked at the edge of the road next to the lane 250 .

The sensors 220 are, for example, lidar sensors, each with a different “field of view” and/or different spatial resolution. In an alternative embodiment, at least some of the sensors 220 could also be in the form of radar sensors, cameras or the like. The fields of view of the lidar sensors 220 are arranged, for example, in such a way that they enclose a predetermined section of the lane 250 . In monitoring lane 250, lidar sensors 220 may generate sensor data that indicates distances contain information about multiple measurement points within their fields of view. For example, a processor circuit 230 coupled to the sensor unit 219 or a storage medium for the sensor data serves to acquire 120 the sensor data. The processor circuit 230 is installed, for example, on the vehicle 210 or alternatively remote from the vehicle 210 and is wired or wirelessly connected to the sensor unit to acquire 120 the sensor data 219, or coupled to the storage medium.

The distance information of the sensor data allows a determination 130 of a distance 260 of a road user 240 located in the lane 250 to the sensor unit 219. The distance 260 corresponds, for example, to a distance to a measuring point closest to the sensor unit 219 or to a predetermined reference point of the road user 240, which is determined by the Lidar sensors 220 was detected.

Road user 240 approaches sensor unit 219 along lane 250, for example. Road user 240 “simulates” in this way, for example, rearward or oncoming traffic of a stationary autonomously driving vehicle, which is equipped with sensor unit 219 in the application. In the present application example, checking the suitability of sensor unit 219 for autonomous driving can therefore relate, for example, in particular to the suitability of sensor unit 219 for recognizing rearward or oncoming traffic, which is used, for example, to control autonomously driving vehicles when turning off or entering a lane from the roadside is relevant/required.

The processor circuitry 230 may determine the distance 260 using the sensor data. Optionally, the processor circuit 230 can determine the distance 260 from the sensor data of a combination or individual sensors in order to check the suitability of the combination or individual sensors.

The suitability check provides, for example, for the processor circuit 230 to compare 140 the distance 260 determined using the sensor data with a distance threshold value. The distance threshold is, for example, predetermined and stored in a memory (not shown) coupled to processor circuitry 230 .

The distance 260 can be a maximum distance at which the road user 240 can be detected (for the first time) by means of the sensor unit 219 while he is approaching the sensor unit 219 . In the context of autonomous driving, the road user 240 can be understood as “detected” if the road user 240 is recognized as an approaching object based on the sensor data, is recognized as such, can be distinguished from an environment of the sensor unit 219 based on the sensor data and/ or for example a (vehicle) category (for example bicycle, car, truck, motorcycle or pedestrian) of the road user 240 can be determined.

By comparing 140 the (maximum) distance 260 with the distance threshold value, it can thus be determined whether the sensor unit 219 is able to detect rear traffic sufficiently “early”. If the determined distance 260 exceeds the distance threshold value, then the suitability of the sensor unit 219 can be confirmed. In the opposite case, the suitability can be refuted or at least doubted.

Depending on the application of the sensor unit 219 in autonomous driving, for which the suitability of the sensor unit 219 is checked, the distance threshold can be different. To check the suitability of sensor unit 219 for controlling autonomously driving vehicles when merging onto a freeway, the distance threshold value can be greater than when merging onto a local road, for example, since relative speeds, reaction distances, braking distances and/or stopping distances of other road users compared to autonomously driving vehicles when merging onto the motorway are larger than when merging into the local road.

The distance threshold value can be dependent on applicable legal regulations, ie for example dependent on a speed limit within the traffic area.

Alternatively, the distance threshold value can be dependent on a movement state of road user 240 . The distance threshold is dependent, for example, on a speed and a direction of movement of road user 240.

The distance threshold behaves, for example, proportionally or exponentially to the determined speed of road user 240 and increases, for example, exponentially with the determined speed of road user 240 . A typical braking distance, which can represent the distance threshold value, for example, can in particular be quadratically proportional to the determined speed of road user 240 plus a constant reaction time and a movement of one's own.

When determining the distance threshold, a ratio of the distance threshold to the speed can be selected for safety reasons such that the distance threshold includes at least one reaction distance, braking distance or stopping distance of the road user, which results from the speed of the road user.

For this purpose, the state of movement of the road user 240 can be determined either by means of the sensor unit 219 or a separate device designed for this purpose. The state of motion results, for example, from a measurement based on the Doppler shift or distance information/distances of the road user 240 to the sensor unit 219 measured with a time delay.

In order to take into account a measurement uncertainty of the sensor unit 219, the processor circuit 230 can optionally compare a difference between the determined distance 260 and the measurement uncertainty of the sensor unit 219 with the distance threshold value.

The measurement uncertainty results, for example, from a spatial resolution of the sensor unit 219 and/or errors when determining the distance 260 using the sensor data.

The measurement uncertainty can also be stored in the data memory. The measurement uncertainty can correspond to a manufacturer's specification or can be determined experimentally.

3 shows a schematic of a principle for (experimental) determination of the measurement uncertainty of the sensor unit 219.

To determine the measurement uncertainty, another road user 240' approaching sensor unit 219 in lane 250 can be equipped with a position sensor 242 which is designed, for example, to determine a geographic position of road user 240'. A distance 270 of the road user 240' to the sensor unit 219 can be determined on the basis of the geographical position of the road user 240' and if a geographical position of the sensor unit 219 is known. The measurement uncertainty can thus be determined based on a deviation 262 of the distance 260 determined by means of the sensor unit 219 and the distance 270 determined by means of the position sensor. In addition, it is conceivable that different other road users 240' are used, for example to determine the effects of different paint finishes and shapes on the measurement uncertainty.

The procedure described above for determining the measurement uncertainty can be carried out repeatedly in order to determine a statistical/average measurement uncertainty of the sensor unit 219 .

To account for the measurement uncertainty when checking the suitability of the sensor unit 219, the comparison 140 of the distance 260 with the distance threshold value can provide a comparison of a difference between the distance 260 determined using the sensor data and a measurement uncertainty of the sensor unit 219 with the distance threshold value. As a result, false statements about the suitability of the sensor unit 219, which stem from its measurement uncertainty, can be avoided.

For greater reliability of a result of the suitability check, the method 100 described above can be carried out using a number of road users driving in the traffic area. In the application example shown, the suitability of sensor unit 219 can be checked, for example, on the basis of a number of road users who pass sensor unit 219 in lane 250 within a specific period of time.

In addition, the vehicle 210 can be repositioned/“reparked” after a specific time, so that different traffic areas can be monitored using the sensor unit 219 . This makes it possible to check the suitability of the sensor unit 219 for autonomous driving in different traffic areas and traffic situations. For example, the vehicle 210 is “reparked” after several days.

The lidar sensors 220 can also be cleaned in the process. For an automatic cleaning of the lidar sensors 220, the vehicle can be equipped with a cleaning system suitable for this purpose, which is designed to clean the lidar sensors 220 with water, compressed air and/or windshield wipers. In an alternative embodiment, the sensor unit 219 could also have cameras, radar sensors or the like, which can also be cleaned. Safety glass can also be fitted in front of the lidar sensors 220 to protect against vandalism.

In addition, an orientation, an arrangement of the lidar sensors 220 or a changed combination of the lidar sensors 220 in the sensor unit 219 can be used to check the suitability of different combinations, orientations and arrangements. For example, lidar sensors with different fields of view Dern and / or different spatial resolutions are combined with each other.

The method 100 described herein can be implemented at least in part in a computer program.

• Erfassen von Sensordaten einer zur Überwachung eines Verkehrsbereichs ortsfest installierten Sensoreinheit;
• Ermitteln eines Abstands der Sensoreinheit zu zumindest einem in dem Verkehrsbereich befindlichen Verkehrsteilnehmer anhand der Sensordaten; und
• Vergleichen des anhand der Sensordaten ermittelten Abstands mit einem Abstandsschwellenwert zur Überprüfung der Eignung der Sensoreinheit für autonomes Fahren.

The computer program comprises, for example, a program code which, when executed on a computer, a processor, a control module or a programmable hardware component, carries out a method with the following steps:

• Acquisition of sensor data from a sensor unit installed in a stationary manner for monitoring a traffic area;
• determining a distance of the sensor unit from at least one road user located in the traffic area based on the sensor data; and
• Comparing the distance determined using the sensor data with a distance threshold value to check the suitability of the sensor unit for autonomous driving.

The computer program can be implemented in a device for checking the suitability of a sensor unit for autonomous driving.

4 shows a device 400 for checking the suitability of a sensor unit for autonomous driving. The device 400 includes an interface 410 for communication and a processor circuit 420 for executing the computer program.

In exemplary embodiments, the interface 410 of the device 400 can be embodied as contacts of the aforementioned module. In exemplary embodiments, they can also be embodied as separate hardware. They can include memories that at least temporarily store the signals to be sent or received. The interface 410 can be designed to receive electrical signals, for example as a bus interface, or as an optical interface. In addition, it can be designed for radio transmission in exemplary embodiments and can include a radio front end and associated antennas.

The interface 410 is designed, for example, as an interface to the sensor unit. The interface 410 is optionally designed as an interface to a storage medium on which the sensor data are stored. The storage medium is, for example, a data memory or data carrier. The device 400 can be arranged separately from the sensor unit and/or outside the traffic area. The sensor data can be transmitted from the sensor unit to the storage medium and/or stored, in particular at least partially anonymized, and read out of the storage medium by device 400 for further processing with a time and/or location offset. In the device 400, different versions of software, for example perception software, can be checked using the sensor data. The device 400 can be embodied as a computer server, for example.

Furthermore, the interface 410 is coupled to the processor circuit 420 . This allows the processor circuit 420 to capture the sensor data of the sensor unit via the interface 410 and to determine a distance of the sensor unit from at least one road user located in a traffic area based on the sensor data. Furthermore, the processor circuit 420 can be designed to compare the distance with a distance threshold value for checking the suitability of the sensor unit for autonomous driving.

Processor circuitry 420 may include other elements of the aforementioned device in exemplary embodiments. Any processor cores, such as digital signal processor cores (DSPs), can be used. Exemplary embodiments are not limited to a specific type of processor core. Any processor cores or even a plurality of processor cores or microcontrollers are conceivable for implementing the processor circuit 420 . Implementations in integrated form with other devices are also conceivable, for example in a control unit for a vehicle, which additionally includes one or more other functions. In exemplary embodiments, the processor circuit 420 can be represented by a processor core, a computer processor core (CPU = central processing unit), a graphics processor core (GPU = graphics processing unit), an application-specific integrated circuit core (ASIC = application-specific integrated circuit), an integrated circuit (IC = Integrated circuit), a single-chip system core (SOC = System on Chip), a programmable logic element or a field-programmable gate array with a microprocessor (FPGA = Field Programmable Gate Array) can be realized as the core of the above-mentioned component or components.

The processor circuit 420 is implemented on the vehicle 210 as the processor circuit 230 or remote from the vehicle 210, for example.

For further details on the features and application of the device 400, refer to FIG the 1 , 2 and 3 related explanations.

The aspects and features described together with one or more of the previously detailed examples and figures can also be combined with one or more of the other examples to replace a same feature of the other example or to add the feature to the other example to introduce

Furthermore, the following claims are hereby incorporated into the Detailed Description, where each claim may stand on its own as a separate example. While each claim may stand on its own as a separate example, it should be noted that although a dependent claim in the claims may relate to a particular combination with one or more other claims, other examples also include a combination of the dependent claim and the subject-matter of each other dependent or independent claim. Such combinations are explicitly suggested herein unless it is indicated that a particular combination is not intended. Furthermore, features of a claim are also intended to be included for any other independent claim, even if that claim is not made directly dependent on the independent claim.

Reference List

100

procedure

110

arrange

120

Monitor

130

Determine

140

Compare

210

vehicle

219

sensor unit

220

lidar sensors

240

road users

240`

road users

242

position sensor

250

lane

260

determined distance

262

deviation

270

Distance

400

contraption

410

interface

420

processor circuit

Claims (16)

A method (100) for checking the suitability of a sensor unit (219) for autonomous driving, the method (100) comprising: stationary arrangement (110) of the sensor unit (219) for monitoring a traffic area (250); Acquisition (120) of sensor data from the sensor unit (219) relating to the traffic area (250); Determining (130) a distance (260) of the sensor unit (219) from at least one road user (240) located in the traffic area (250) using the sensor data; and Comparing (140) the distance (260) determined using the sensor data with a distance threshold value to check the suitability of the sensor unit (219) for autonomous driving. Method (100) according to claim 1 , wherein the distance (260) determined using the sensor data corresponds to a maximum distance of the sensor unit (219) from the at least one road user (240), at which the at least one road user (240) can be detected using the sensor data. Method (100) according to one of the preceding claims, further comprising determining a state of movement of the at least one road user (240), wherein the distance threshold value is dependent on the state of movement of the at least one road user (240). Method (100) according to claim 1 or 2 , where the distance threshold is dependent on legislation in force within the traffic area (250). Method (100) according to one of the preceding claims, wherein the comparison (130) of the distance (260) determined using the sensor data with the distance threshold value involves comparing a difference between the distance (260) determined using the sensor data and a measurement uncertainty of the sensor unit (219). the distance threshold. Method (100) according to claim 5 , further comprising: determining a distance (260) of the sensor unit to at least one other road user (240') located in the traffic area (250) based on the sensor data; Detection of position data from a position sensor (242) which is attached to the at least one other road user (240'); and determining the measurement uncertainty of the sensor unit (219) based on the position data and on the basis the distance (260) determined from the sensor data to the at least one other road user. Method (100) according to one of the preceding claims, wherein the sensor data comprises image data. Method (100) according to claim 7 , further comprising anonymizing at least part of the image data. Method (100) according to one of the preceding claims, comprising acquiring sensor data of a first traffic area and at least one second traffic area. Method (100) according to claim 9 , further comprising moving the sensor unit (219) from a first position for acquiring the sensor data of the first traffic area to a second position for acquiring the sensor data of the at least one second traffic area. Method (100) according to one of the preceding claims, wherein the sensor unit (219) comprises a first and at least one second sensor. Computer program with a program code which, when executed on a computer, a processor, a control module or a programmable hardware component, carries out a method with the following steps: Acquisition of sensor data from a sensor unit installed in a stationary manner for monitoring a traffic area; determining a distance of the sensor unit from at least one road user located in the traffic area based on the sensor data; and Comparing the distance determined using the sensor data with a distance threshold value to check the suitability of the sensor unit for autonomous driving. Device (400) for checking a suitability of a sensor unit for autonomous driving, the device (400) comprising: at least one interface (410) for communication; and a processor circuit (420) for executing the computer program claim 12 . Sensor device for checking the suitability of a sensor unit (219) for autonomous driving, comprising: a carrier structure (210) with which the sensor unit is stationarily arranged during a monitoring of a traffic area (250) by means of the sensor unit (219) for generating sensor data of the traffic area (250); and a processor circuit (230) configured to to determine a distance (260) of the sensor unit (219) from a road user (240) located in the traffic area (250) on the basis of the sensor data; and to compare the distance (260) determined using the sensor data with a distance threshold value for checking the suitability of the sensor unit (219) for autonomous driving. Sensor device according to Claim 14 , wherein the support structure (210) is designed as a vehicle. Data carrier, comprising in a method according to one of Claims 1 until 11 determined data, in particular sensor data.

Images