Comparison of map data and sensor data

Abstract

Device (14) for fusing sensor data with map data, comprising: an input interface (24) for receiving sensor data from an environment sensor (18) with information about objects in the environment of a vehicle (12) and for receiving map data with information about the environment of the vehicle; an evaluation unit (26) for recognizing visible traffic signs (20) in the vicinity of the vehicle based on the sensor data and for reading out previously known traffic signs in the vicinity of the vehicle from the map data; and an assignment unit (28) for assigning the visible traffic signs to the previously known traffic signs and for determining a safety parameter that indicates a probability of a correct assignment, wherein the assignment unit (28) for assigning the recognized visible traffic signs (20) to the read-out pre-known traffic signs is designed based on an Iterative Closest Point algorithm, so that a current position of the vehicle in relation to the map data can be determined.

Description

[0001] The present invention relates to a device for fusing sensor data with map data. The present invention further relates to a control device for controlling an autonomous or semi-autonomous vehicle, as well as a system for controlling an autonomous or semi-autonomous vehicle. The invention also relates to a method and a computer program product.

[0002] Modern vehicles (cars, trucks, motorcycles, etc.) are equipped with a variety of sensors (radar, lidar, cameras, ultrasound, etc.) that provide information to the driver or the vehicle's control system. These environmental sensors detect the vehicle's surroundings and objects within them (other vehicles, infrastructure, people, moving objects, etc.). Based on the collected data, a model of the vehicle's environment can be created, and the system can react to changes in this environment. In particular, this allows the vehicle's driving functions to be performed semi- or fully autonomously.

[0003] In addition to sensor data, the evaluation and control algorithms are increasingly based on map data. Map data refers in particular to previously known information about roads, lanes on those roads, and objects in the surroundings. This map data is then used as the basis for short- and medium-term route planning, and increasingly also as the starting point for controlling driving or driver assistance functions. The sensor data from the environmental sensors are considered and evaluated in conjunction with the map data.

[0004] One challenge lies in fusing the information obtained from environmental sensors with map data. Map data, especially high-resolution map data, contains detailed information about the vehicle's surroundings and can improve decision-making in autonomous or semi-autonomous vehicles. In some cases, it may even be possible to execute functions of a semi-autonomous or autonomous vehicle, at least partially and / or temporarily, solely based on map data, for example, to compensate for sensor failures. This requires a reliable and robust fusion of map and sensor data, particularly regarding the vehicle's position relative to the map data.

[0005] In this context, previous approaches to fusing sensor and map data often place high demands on computing power and processing time. Especially when the behavior of an autonomous or semi-autonomous vehicle is to be defined based on sensor and map data, fast data processing and, in many cases, even real-time analysis are necessary.

[0006] Peker et al., “Fusion of Map Matching and Traffic Sign Recognition”, 2014, describes an approach for the high-performance recognition of traffic signs and the fusion of the recognized traffic signs with digital maps. A traffic sign is detected using a monochrome camera. Standard navigation maps are used.

[0007] German patent application DE 10 2019 101 405 A1 discloses a method for evaluating the positional information of a landmark in the vicinity of a motor vehicle. It also discloses an evaluation system, a driver assistance system, and a motor vehicle. The method proposes providing a map containing landmarks that correspond to real-world landmarks in the surroundings. The result of capturing a portion of the environment is provided as sensor data, comprising initial landmark data relating to a real-world landmark. Depending on the vehicle's position, secondary landmark data is derived from the map, relating to a landmark on the map that corresponds to the real-world landmark. Based on the comparison of the initial and secondary landmark data, the positional information of the real-world landmark and / or the landmark on the map is evaluated.

[0008] Based on this, the present invention aims to provide an approach for the efficient and reliable fusion of sensor data with map data. In particular, it aims to provide an approach that enables the efficient real-time mapping of map data and environmental sensor data. To achieve this objective, the present invention relates, in a first aspect, to a device for fusion of sensor data with map data, comprising: an input interface for receiving sensor data from an environmental sensor with information about objects in the environment of a vehicle and for receiving map data with information about the environment of the vehicle; an evaluation unit for recognizing visible traffic signs in the vicinity of the vehicle based on sensor data and for reading previously known traffic signs in the vicinity of the vehicle from map data; and an allocation unit for assigning the visible traffic signs to the previously known traffic signs and for determining a safety parameter that indicates a probability of a correct allocation, wherein the assignment unit (28) for assigning the recognized visible traffic signs (20) to the read-out pre-known traffic signs is designed based on an Iterative Closest Point algorithm, so that a current position of the vehicle in relation to the map data can be determined.

[0009] In another aspect, the present invention relates to a control device for controlling an autonomous or semi-autonomous vehicle, comprising: a receiving interface for receiving a safety parameter that indicates the reliability of an assignment of visible traffic signs detected based on sensor data from an environmental sensor to previously known traffic signs read from map data by a device as defined above; and a pre-planning unit for planning the behavior of the vehicle based on map data and a determined position of the vehicle in relation to this map data, wherein The advance planning unit is configured to extend the planning time horizon if the received safety parameter indicates a higher reliability of the assignment than in a previous time step. Furthermore, an aspect of the present invention relates to a system for controlling an autonomous or semi-autonomous vehicle, comprising: a device as previously described and a control device as previously described; and an environmental sensor for detecting objects in the vicinity of the vehicle.

[0010] Further aspects of the invention relate to a method designed according to the device and a control method designed according to the control device, as well as a computer program product with program code for carrying out the steps of the methods when the program code is executed on a computer. In addition, one aspect of the invention relates to a storage medium on which a computer program is stored which, when executed on a computer, causes the execution of the methods described herein.

[0011] Preferred embodiments of the invention are described in the dependent claims. It is understood that the features mentioned above and those to be explained below can be used not only in the combinations specified, but also in other combinations or individually, without departing from the scope of the present invention. In particular, the device, the control device, the system, the methods, and the computer program products can be implemented according to the embodiments defined for the device in the dependent claims.

[0012] According to the invention, data from an environmental sensor and map data are received. Both types of data include information about objects in the vehicle's surroundings. The sensor data is received from at least one environmental sensor, for example, a radar, lidar, camera, or ultrasonic sensor. The map data is received from a map database, which can be located locally or connected via a data connection, in particular a mobile data connection. Traffic signs are identified in the environmental sensor data. Traffic signs are read from the map data. Based on the identified and read traffic signs, an assignment is then made, and a safety parameter is determined that indicates the reliability of the assignment, in particular a probability of a correct assignment or a confidence level.

[0013] In contrast to previous approaches, which relied on a multitude of different detected objects for classification, the invention considers only traffic signs. Classification is based solely on traffic signs. This enables a significantly simplified and more efficient classification process. Furthermore, traffic signs can be detected with high accuracy based on environmental sensor data and, at the same time, made available in map data with high reliability. Therefore, reliable classification can be provided with reduced computational effort. By determining a safety parameter, additional information is provided for further data processing or for planning the behavior of an autonomous or semi-autonomous vehicle, enabling efficient further processing.In particular, the safety parameter allows the time horizon for predicting the behavior of the autonomous or semi-autonomous vehicle to be adjusted. If the safety parameter indicates a high probability of a correct mapping, a longer predictive period can be used. The safety parameter enables the adaptation of predictive planning of the behavior of an autonomous or semi-autonomous vehicle based on the reliability of mapping current sensor data to map data. This can improve the operational safety of the autonomous or semi-autonomous vehicle.

[0014] In a preferred embodiment, the mapping unit for assigning visible traffic signs to previously known traffic signs is based on an iterative closest-point algorithm. The iterative closest-point algorithm (ICP) is most commonly used for localizing autonomous systems based on lidar or radar point clouds or semantic points in camera data. This mapping is usually computationally intensive. In contrast, the mapping of visible traffic signs to previously known traffic signs proposed according to the invention can be calculated much more efficiently, since only a comparatively small number of data points need to be included. This results in fast and efficient computation of the mapping.

[0015] In a preferred embodiment, the assignment unit for determining a one-dimensional value on a predefined scale is configured as a reliability value. In particular, the use of an easily computable one-dimensional value enables high efficiency in further processing when planning the behavior of the autonomous or semi-autonomous vehicle. For example, a percentage or decimal value can be used. Efficient computability is achieved.

[0016] In a preferred embodiment, the evaluation unit is configured to determine the positions of the detected visible traffic signs and the read-out pre-existing traffic signs. The assignment unit is configured to assign the detected visible traffic signs to the read-out pre-existing traffic signs based on the determined positions. Preferably, the positions of the traffic signs are read and determined. The assignment refers in particular to the positions of the signs. A vehicle position relative to its surroundings or to map data can then be determined. For most applications, position is a crucial prerequisite for planning the behavior of an autonomous or semi-autonomous vehicle. Precise positional knowledge extends the planning horizon.

[0017] In a preferred embodiment, the assignment unit is designed to minimize a mean square distance between the positions of the recognized visible traffic signs and the read-out pre-known traffic signs.

[0018] A mean squared distance is an efficiently calculable error measure and thus offers a simple way of finding an assignment.

[0019] In a preferred embodiment, the evaluation unit is designed to determine the classes of the detected visible traffic signs and the previously read traffic signs. The assignment unit is designed to assign the detected visible traffic signs to the previously read traffic signs based on the determined classes. A traffic sign class refers to the type of traffic sign, particularly with regard to its message about traffic rules or other regulations. For example, stop signs, no-stopping signs, etc., can represent a class of traffic sign. This type or class of traffic sign is taken into account during the assignment process. This further improves the reliability of the assignment and achieves efficient assignment.

[0020] In a preferred embodiment, the evaluation unit is designed to determine the orientations of the detected visible traffic signs and the read-out pre-existing traffic signs. The assignment unit is designed to assign the detected visible traffic signs to the read-out pre-existing traffic signs based on the determined orientations. Orientation refers to the orientation of the traffic signs, specifically their position relative to a two-dimensional road surface. The direction in which the traffic sign points is taken into account. The orientation is considered as additional information. This results in further improved reliability in the assignment. The orientation can usually be efficiently derived from both the environmental sensor data and the map data.

[0021] An object in the vicinity of a vehicle can be, in particular, a vehicle, a cyclist, a pedestrian, an animal, or a static object such as a car tire lying on the road or a traffic sign, etc. An area can be, in particular, the surroundings of a vehicle or an area visible from the vehicle. An area can also be defined by a radius or other distance measurement. Map data, in this context, refers in particular to a representation of an area or region with regard to roads, cycle paths, footpaths, traffic signs, and other objects, etc. Map data can be in any format. An autonomous or semi-autonomous vehicle is a vehicle in which a computer unit provides at least part of a driving function.A traffic sign is a sign placed in the area of ​​a roadway with information on regulating traffic, objects and sights in the vicinity, destinations and distances, directions, etc.

[0022] The invention is described and explained in more detail below with reference to some selected embodiments in conjunction with the accompanying drawings. These show: Fig. 1 a schematic representation of a system according to the invention for controlling an autonomous or semi-autonomous vehicle; Fig. 2 a schematic representation of a device according to the invention for fusing sensor data with map data; Fig. 3 a schematic representation of a control device according to the invention; Fig. 4 a schematic representation of the inventive approach for fusing sensor data with map data; Fig. 5 a schematic representation of an assignment that is highly likely to be correct; Fig. 6 a schematic representation of an assignment that is correct with reduced probability; Fig. 7 a schematic representation of a situation in which no reliable assignment can be achieved; and Fig. 8 a schematic representation of a method according to the invention for fusing sensor data with map data.

[0023] In the Fig. Figure 1 schematically depicts a system 10 according to the invention for controlling an autonomous or semi-autonomous vehicle 12. The system 10 comprises a device 14 for fusing sensor data with map data, a control device 16 for controlling the vehicle 12, and an environmental sensor 18 for detecting objects in the vicinity of the vehicle 12. In the illustrated embodiment, the system 10 is integrated into the vehicle 12. The illustration is to be understood as a side sectional view of the vehicle 12 on a roadway. The vicinity of the vehicle 12 includes, in particular, traffic signs 20, which are detected as objects by the environmental sensor 18. The device 14 and the control device 16 can, for example, be integrated into a control unit or a central computer of the vehicle 12. It is also possible that the device 14 or the control device 16 is integrated into the environmental sensor 18.The environmental sensor 18 can be mounted on the vehicle 12. However, it is also possible that the device 14, the control device 16 and / or the environmental sensor 18 are designed separately, for example integrated into a smartphone.

[0024] According to the invention, the environment surrounding the vehicle 12 is detected by means of the environmental sensor 18. Traffic signs 20 are recognized from the sensor data. The device 14 is also configured to receive map data. In the illustrated embodiment, the map data is received from a central server 22 via a mobile data connection. However, it is also possible for the map data to be received from a database located within the vehicle 12, within the device 14 itself, or elsewhere. The sensor data from the environmental sensor 18 is fused with the map data in the device. In particular, a mapping between sensor data and map data is performed based on the traffic signs 20. A safety parameter is calculated that expresses a probability of a correct mapping.

[0025] In the Fig. Figure 2 schematically depicts a device 14 according to the invention for fusing sensor data with map data. The device comprises an input interface 24, an evaluation unit 26, and an allocation unit 28. The units and interfaces can be implemented partially or completely in software and / or hardware. In particular, the units can be configured as a processor, processor modules, or as software for a processor. The device 14 can, in particular, be configured as a control unit or a central computer of an autonomous or semi-autonomous vehicle, or as software for a control unit or a central computer of an autonomous or semi-autonomous vehicle.

[0026] Input interface 24 receives both sensor data from the environmental sensor and map data. Input interface 24 is connected to an environmental sensor, such as a radar, lidar, camera, or ultrasonic sensor. It is understood that the input interface can be connected to multiple sensors and can receive pre-processed sensor data. In particular, it can receive, for example, a point cloud or a camera image. The map data can be received from either a local or a remote database.

[0027] In evaluation unit 26, visible traffic signs are detected by analyzing the received sensor data. Visible traffic signs are defined here as those located within the field of view of the environmental sensor. This detection of visible traffic signs can be based on sensor data processing algorithms, particularly image analysis. Specifically, pattern recognition can be performed using image data.

[0028] Furthermore, in evaluation unit 26, previously identified traffic signs are read from the map data. Depending on the map data format, a corresponding query or evaluation is performed. Either a query is performed for all traffic signs present in the map data, or only traffic signs from a specific area corresponding to the vehicle's current position or estimated position are read. Both the recognition and the reading of traffic signs relate specifically to determining or reading a position within a corresponding coordinate system. For example, a vehicle-specific coordinate system can be used to determine the positions of visible traffic signs from the sensor data. Similarly, a coordinate system from the map data can be used to read the positions of the traffic signs contained within it.Furthermore, the recognition of traffic signs may also include determining their orientation in relation to the relevant coordinate system. It is also possible to determine the class, i.e., the type of traffic sign, based on its meaning, using sensor data and map data. This means determining, for example, whether it is a stop sign, a yield sign, etc.

[0029] Based on the detected visible traffic signs and the previously identified traffic signs, an assignment is then made in assignment unit 28, and the probability of a correct assignment is determined. For this purpose, a corresponding assignment algorithm creates the most suitable mapping possible between the traffic signs detected based on the sensor data and the traffic signs read from the map data. In particular, the positions of the signs can be used for the assignment. Furthermore, the orientation and / or the determined class of the traffic signs can also be taken into account.

[0030] For the assignment, an iterative closest-point algorithm can be used. To determine the quality of the assignment, i.e., the confidence parameter, a mean squared error can be calculated for each position. The confidence parameter then indicates how reliable the assignment is. In particular, the confidence parameter provides a measure of whether the assignment is reliable or whether no assignment could be made. The confidence parameter is determined on a predefined scale. For example, a percentage can be used. The scale can also be open-ended.

[0031] The output of the Iterative Closest Point algorithm typically includes information on the translational and rotational relationship between the sensor data and the map data, or between the positions of the detected traffic signs in the sensor data and the positions of the traffic signs read from the maps. From this, the current position of the environmental sensor or the vehicle relative to the map data can be derived. Two criteria can be used to calculate the safety parameter. Firstly, an assignment error of the Iterative Closest Point algorithm can be considered. Secondly, an uncertainty in the localization, for example in the form of covariance or another measure, can be used as a safety parameter or as the basis for calculating the safety parameter.

[0032] In the Fig. Figure 3 schematically depicts a control device 16 according to the invention for controlling an autonomous or semi-autonomous vehicle. The control device 16 comprises a receiver interface 30 and a planning unit 32. As previously described, the units and interfaces can be implemented partially or completely in software and / or hardware. The control device 16 can be implemented together with the device 14.

[0033] The safety parameter is received via the receiving interface 30. For this purpose, the receiving interface 30 can be connected in particular to a device 14 for positioning sensor data and map data.

[0034] In the pre-planning unit 32, the behavior of an autonomous or semi-autonomous vehicle is planned. Planning the behavior of an autonomous or semi-autonomous vehicle includes, for example, determining a short-term route or making a decision regarding braking, acceleration, or evasive maneuvers. Pre-planning unit 32 is designed to extend the time horizon of this planning if the safety parameter indicates a high degree of reliability in the mapping. In this respect, the pre-planning horizon is dependent on the previously determined safety parameter. The more reliable the mapping between sensor data and map data, the longer the time horizon of the pre-planning.

[0035] It is therefore proposed to use the safety parameter as the basis for fusing sensor data with map data in subsequent time steps. The higher the safety parameter, the more the map data can be used to plan the behavior of the autonomous or semi-autonomous vehicle. Compared to previous approaches using point cloud data or semantic data, which also employ an iterative closest-point algorithm for localization and for determining the certainty of the localization or assignment, the invention's exclusive use of traffic signs enables significantly more efficient computation. This can, for example, improve the real-time decision-making capability of the autonomous or semi-autonomous vehicle.

[0036] In the Fig. 4, Fig. 5, Fig. 6 to Fig. Figure 7 schematically illustrates the inventive approach of fusing sensor data with map data. The left side represents the evaluation / processing of the map data. The right side relates to the evaluation / processing of the sensor data.

[0037] In the Fig. Figure 4 on the left schematically illustrates that, according to map data, two traffic signs 20 (priority and pedestrian crossing) are located in the vicinity of vehicle 12 and are read from the map. The right side shows that the two traffic signs 20 in the vicinity of vehicle 12 are detected by evaluating the sensor data from the environmental sensor in the same way. After the traffic signs have been detected or read from the map, a mapping between the visible traffic signs and the previously known traffic signs is performed in the mapping unit 28. For this purpose, an iterative closest-point algorithm can be used, which generates corresponding rotation and translation matrices. Based on this mapping, a safety parameter can then be determined.

[0038] In one embodiment, this safety parameter explicitly specifies a probability of a correct assignment. If the probability of a correct assignment is high, a control system of an autonomous or semi-autonomous vehicle can use the map data to plan the vehicle's behavior. For example, the safety parameter can provide binary information indicating a correct / incorrect assignment.

[0039] In the Fig. Figure 5 schematically depicts a traffic situation. On the left (top), a vehicle 12 and the traffic signs 20, readable from map data, are shown in the vehicle's vicinity. On the right (top), the vehicle 12's environmental sensor's perception within the sensor's field of view 34 is shown. Below, it is shown that the vehicle 12 is clearly located at position 1, as there is a complete match between the traffic signs 20 detected based on the sensor data and the traffic signs 20 read from the map data. The safety parameter thus indicates a high probability of a correct assignment.

[0040] In the Fig. Figure 6 schematically depicts a case where, as shown on the right in the Fig. As indicated in Figure 6, the environmental sensor on the vehicle 12 only detects three traffic signs 20 within the sensor's field of view 34. The corresponding map data for the same area is shown on the left, according to which a total of six traffic signs 20 should be present. However, since a correct assignment can still be made based on the orientation, position, and class of the traffic signs, the result is again position 1 as the vehicle's position. Here too, the safety parameter indicates a high probability of a correct assignment.

[0041] In the Fig. Figure 7 depicts a situation in which only a single traffic sign 20 is detected by the environmental sensors on vehicle 12 within the sensor's field of view 34. Comparison with the map data shown on the left, which contains six traffic signs 20 in the corresponding area, reveals an ambiguity between positions 1, 2, and 3. At all three positions, there is a round traffic sign 20 to the right of vehicle 12. Therefore, a reliable and accurate assignment cannot be made. The safety parameter indicates a low probability of an accurate assignment.

[0042] In the Fig.Figure 8 schematically illustrates a method according to the invention for fusing sensor data with map data. The method comprises steps S10 of receiving sensor data and map data, S12 of recognizing visible traffic signs, S14 of reading previously known traffic signs, S16 of assigning the visible traffic signs to the previously known traffic signs, and S18 of determining a safety parameter. The method can be implemented, in particular, as software that runs on a processor of a vehicle or a vehicle control unit. It is understood that the vehicle can also be implemented as a smartphone app.

[0043] The invention has been comprehensively described and explained with reference to the drawings and the description. The description and explanation are to be understood as examples and not as limiting. The invention is not limited to the disclosed embodiments. Other embodiments or variations will become apparent to a person skilled in the art when using the present invention and upon a detailed analysis of the drawings, the disclosure, and the subsequent claims.

[0044] In the patent claims, the words "comprise" and "with" do not preclude the presence of further elements or steps. The undefined article "a" or "an" does not preclude the presence of multiple elements. A single element or unit can perform the functions of several of the units mentioned in the patent claims. An element, unit, interface, device, and system can be implemented partially or completely in hardware and / or software. The mere mention of some measures in several different dependent patent claims is not to be understood as precluding the advantageous use of a combination of these measures. A computer program can be stored / distributed on a non-volatile data carrier, for example, on optical storage media or on a solid-state drive (SSD).A computer program can be distributed together with hardware and / or as part of hardware, for example via the internet or via wired or wireless communication systems. Reference punctuation in the patent claims is not to be understood as limiting. Reference sign 10 System 12 vehicles 14 Device 16 Control device 18 Environmental sensor 20 traffic signs 22 central servers 24 input interfaces 26 evaluation units 28 Assignment unit 30 Receiving interface 32 Advance planning unit 34 field of vision

Claims

[1] Device (14) for fusing sensor data with map data, comprising: an input interface (24) for receiving sensor data from an environment sensor (18) with information about objects in the environment of a vehicle (12) and for receiving map data with information about the environment of the vehicle; an evaluation unit (26) for recognizing visible traffic signs (20) in the vicinity of the vehicle based on the sensor data and for reading out previously known traffic signs in the vicinity of the vehicle from the map data; and an assignment unit (28) for assigning the visible traffic signs to the previously known traffic signs and for determining a safety parameter that indicates a probability of a correct assignment, wherein the assignment unit (28) for assigning the recognized visible traffic signs (20) to the read-out pre-known traffic signs is designed based on an Iterative Closest Point algorithm, so that a current position of the vehicle in relation to the map data can be determined. [2] Device (14) according to one of the preceding claims, wherein the allocation unit (28) is designed to determine a one-dimensional value on a predefined scale as a reliability value. [3] Device (14) according to one of the preceding claims, wherein the evaluation unit (26) is designed to determine the positions of the detected visible traffic signs (20) and the read-out previously known traffic signs; and the assignment unit (28) is designed to assign the recognized visible traffic signs to the previously known traffic signs read out based on the determined positions. [4] Device (14) according to claim 4, wherein the allocation unit (28) is designed to minimize a mean square distance between the positions of the detected visible traffic signs (20) and the read-out pre-known traffic signs. [5] Device (14) according to any one of the preceding claims, wherein the evaluation unit (26) is designed to determine classes of the recognized visible traffic signs (20) and the read-out previously known traffic signs; and the assignment unit (28) is designed to assign the recognized visible traffic signs to the previously known traffic signs read out based on the determined classes. [6] Device (14) according to one of the preceding claims, wherein the evaluation unit (26) is designed to determine the orientations of the detected visible traffic signs (20) and the read-out previously known traffic signs; and the assignment unit (28) is designed to assign the detected visible traffic signs to the previously known traffic signs read out based on the determined orientations. [7] Control device (16) for controlling an autonomous or semi-autonomous vehicle (12), comprising: a receiving interface (30) for receiving a safety parameter that indicates the reliability of an assignment of visible traffic signs (20) detected on the basis of sensor data from an environment sensor (18) to previously known traffic signs read from map data by a device according to one of the preceding claims; a preplanning unit (32) for planning the behavior of the vehicle based on map data and a determined position of the vehicle in relation to this map data, wherein The advance planning unit is designed to extend a planning time horizon if the received safety parameter indicates a higher reliability of the assignment than in a previous time step. [8] System (10) for controlling an autonomous or semi-autonomous vehicle (12), comprising: a device (14) according to any one of claims 1 to 6 and a control device (16) according to claim 7; and an environmental sensor (18) for detecting objects in the vicinity of the vehicle (12). [9] Method for fusing sensor data with map data, comprising the steps: Receiving (S10) sensor data from an environment sensor (18) with information about objects in the environment of a vehicle (12) and receiving map data with information about the environment of the vehicle; Recognition (S12) of visible traffic signs (20) in the vicinity of the vehicle based on sensor data and retrieval (S14) of previously known traffic signs in the vicinity of the vehicle from map data; and Assigning (S16) the visible traffic signs to the previously known traffic signs and determining (S18) a safety parameter that indicates a probability of a correct assignment, The matching of the recognized visible traffic signs to the read-out previously known traffic signs is based on an iterative closest point algorithm, so that a current position of the vehicle in relation to the map data can be determined. [10] Computer program product comprising program code for performing the steps of the method according to claim 9 when the program code is executed on a computer.

Images