Method for generating data set

Abstract

The invention relates to a method for generating a data set, comprising the following steps: receiving environment data generated by a motor vehicle, which represents the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle, receiving environment data generated by an infrastructure, which represents the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle, the environment data generated by the infrastructure are representative of the surroundings of the motor vehicle as viewed from the infrastructure perspective, processing the environment data generated by the motor vehicle on the basis of the environment data generated by the infrastructure in order to determine processed environment data generated by the motor vehicle, and generating a data set, the data set includes processed environmental data generated by the motor vehicle. The invention relates to a method for testing at least one component of a process chain for a partially automatic driving function of a motor vehicle, to a method for generating at least one part of an algorithm for an at least partially automatic driving function of a motor vehicle, to a device, to a computer program, and to a machine-readable storage medium.

Description

Technical Field

[0001] The present invention relates to a method for generating a data set, a method for testing at least one component of a process chain for at least partially automated driving functions of a motor vehicle, a method for generating at least a part of an algorithm for at least partially automated driving functions of a motor vehicle, a device, a computer program and a machine-readable storage medium. Background Art

[0002] The published patent application DE 10 2013 001 326 A1 discloses a motor vehicle which is designed to exchange operating data with traffic objects in the surroundings of the motor vehicle and thereby to coordinate driving maneuvers of the motor vehicle with the traffic objects.

[0003] Environmental data generated by the motor vehicle are used to test an algorithm for implementing at least partially automated driving functions of the motor vehicle. This means that the motor vehicle detects its surroundings during its travel by means of an environmental sensor system and provides environmental data corresponding to the detection. The environmental data can be used to test an algorithm for implementing at least partially automated driving functions of the motor vehicle. Summary of the invention

[0004] The object of the present invention is to provide a method for generating a data set.

[0005] Object of the Invention It is also an object of the present invention to provide a method for testing at least one component of a process chain for an at least partially automated driving function of a motor vehicle.

[0006] Object of the Invention It is also an object of the present invention to specify a method for generating at least a part of an algorithm for an at least partially automated driving function of a motor vehicle.

[0007] Object of the Invention The object of the present invention is also to provide a provision device.

[0008] OBJECT OF THE INVENTION The object of the present invention is also to provide a computer program.

[0009] Object of the Invention The object of the present invention is also to provide a machine-readable storage medium.

[0010] The objects are each achieved by the respective subject matter of the independent claims. Advantageous embodiments of the invention are the subject matter of the respective dependent claims.

[0011] According to a first aspect, there is provided a method for generating a data set, comprising the following steps:

[0012] receiving environment data generated by a motor vehicle, the environment data generated by the motor vehicle representing the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle,

[0013] receiving environment data generated by the infrastructure, the environment data generated by the infrastructure representing the surroundings of the motor vehicle as viewed from the perspective of the infrastructure,

[0014] processing the surroundings data generated by the motor vehicle based on the surroundings data generated by the infrastructure in order to determine processed surroundings data generated by the motor vehicle, and

[0015] A data set is generated that includes processed environmental data generated by the motor vehicle.

[0016] According to a second aspect, a method for testing at least one component of a process chain for an at least partially automated driving function of a motor vehicle is provided, comprising the following steps:

[0017] receiving a data set comprising vehicle-generated environmental data processed by using infrastructure-generated environmental data representing the surroundings of the vehicle as viewed from the perspective of the infrastructure, the vehicle-generated environmental data representing the surroundings of the vehicle as viewed from the perspective of the vehicle,

[0018] At least one component of the process chain is tested based on the data set.

[0019] According to a third aspect, a method for generating at least a part of an algorithm for an at least partially automated driving function of a motor vehicle is provided, comprising the following steps:

[0020] receiving a data set comprising vehicle-generated environmental data processed by using infrastructure-generated environmental data representing the surroundings of the vehicle as viewed from the perspective of the infrastructure, the vehicle-generated environmental data representing the surroundings of the vehicle as viewed from the perspective of the vehicle,

[0021] At least a portion of an algorithm for an at least partially automated driving function of a motor vehicle is generated based on the data set.

[0022] According to a fourth aspect, a device is provided, which is configured to implement all steps of the method according to the first aspect and / or according to the second aspect and / or according to the third aspect.

[0023] According to a fifth aspect, a computer program is provided, the computer program comprising the following instructions, when the computer program is implemented by a computer, for example, a device according to the fourth aspect, the instructions cause it to implement the method according to the first aspect and / or according to the second aspect and / or according to the third aspect.

[0024] According to a sixth aspect there is provided a machine-readable storage medium on which the computer program according to the fifth aspect is stored.

[0025] The invention is based on and includes the knowledge of generating a data set comprising processed environment data generated by a motor vehicle, wherein the environment data describe or represent the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle. In other words, the environment data generated by the motor vehicle are processed, wherein the processing is carried out based on environment data generated by an infrastructure, wherein the environment data generated by the infrastructure represent the surroundings of the motor vehicle as viewed from the perspective of the infrastructure.

[0026] This provides information that the motor vehicle itself does not have for processing the environmental data generated by the motor vehicle. The environmental sensor system of the motor vehicle detects the surroundings of the motor vehicle to generate the environmental data generated by the motor vehicle, and the environmental sensor system of the motor vehicle has a limited effective range for detecting the surroundings. Objects in the surroundings of the motor vehicle may also conceal, i.e. block, other objects in the surroundings of the motor vehicle in front of the environmental sensor system, so that the environmental sensor system of the motor vehicle cannot detect other objects.

[0027] The infrastructure surroundings sensor system can detect the other objects better than this. The infrastructure surroundings sensor system can also have a greater range than the surroundings sensor system of the motor vehicle, so that the surroundings of the motor vehicle can be detected more by the infrastructure surroundings sensor system than by the surroundings sensor system of the motor vehicle itself.

[0028] As a result, the motor vehicle-generated surroundings data can be processed efficiently, so that the processed motor vehicle-generated surroundings data can be provided in the form of a data set.

[0029] Based on the data set, at least one component of a process chain for an at least partially automated driving function of a motor vehicle can be tested in an advantageous and efficient manner and / or at least a part of an algorithm for an at least partially automated driving function of a motor vehicle can be generated.

[0030] The approach described here can be seen in comparison with the above-mentioned prior art that, according to the approach, only environmental data generated by the motor vehicle are used as unprocessed environmental data for testing an algorithm for at least partially automated driving functions of a motor vehicle. According to the prior art, only the surroundings observed from the perspective of the motor vehicle, but not from the perspective of the infrastructure, are used for testing the algorithm.

[0031] Thus, for example, even if a relatively minor hardware change to the surroundings sensor of the motor vehicle, for example an increase in the field of view of the surroundings sensor by 1° or an increase in the effective range of 20 m, is a better solution in the future, further development of the algorithm or a component of a process chain for at least partially automated driving functions of the motor vehicle based solely on unprocessed surroundings data generated by the motor vehicle is very time-consuming and very expensive. Since, according to the approach described here, the surroundings data generated by the motor vehicle are processed based on the surroundings data generated by the infrastructure, the question of whether a relatively minor hardware change is not a better solution, for example, can be answered effectively.

[0032] The approach described here includes the generation of a data set, which includes processed environmental data generated by the motor vehicle. This means that the environmental data generated by the motor vehicle are processed. This can be seen as being different from the fusion of environmental data generated by the motor vehicle with environmental data generated by the infrastructure. Because in the data fusion a completely new data set is generated: the fused data set.

[0033] However, the term "processed" makes it particularly clear that the original data set, ie the motor vehicle-generated surroundings data, is still recognizable or present. This means in particular that the original motor vehicle-generated surroundings data is still recognizable or present in the processed motor vehicle-generated surroundings data.

[0034] This is similar to the treatment of water or furniture parts. After the treatment, the original furniture part may still be recognizable or still exist. Corners on the furniture part are rounded if necessary, or the furniture part is polished within the scope of the treatment. However, a completely new furniture part is not produced.

[0035] The same applies to treated water. Here, for example, substances are filtered out so that the original water is still present and recognizable, and no new water is produced.

[0036] It is clear from this that data fusion does not involve the processing of surroundings data generated by the motor vehicle.

[0037] Therefore, processing in the sense of this description differs in particular from data fusion, ie the merging of data.

[0038] The processed surroundings data generated by the motor vehicle are therefore in particular not fused surroundings data, and accordingly are in particular non-fused surroundings data.

[0039] In one specific embodiment of the method according to the first aspect, it is provided that the processing includes a validation of the surroundings data generated by the motor vehicle in order to determine a validation result, so that the processed surroundings data generated by the motor vehicle include the validation result.

[0040] The technical advantage achieved thereby, for example, is that the environmental data generated by the motor vehicle can be processed efficiently. The verification can, for example, check the size of an object that is detected based on the environmental data generated by the motor vehicle in the surroundings of the motor vehicle. Motor vehicles sometimes move very dynamically (quickly) in an unknown world. The infrastructure is stationary and part of the known world (all stationary objects, their properties, positions, etc.) is known. In other words, the objects are defined as correct, i.e. as a reference. With the help of the "correct" data of the infrastructure, i.e. the reference ("Ground Truth" in English), the "correctness" of the detected and determined data of the motor vehicle can be checked, i.e. verified.

[0041] In one specific embodiment of the method according to the first aspect, it is provided that the processing includes a plausibility check of the surroundings data generated by the motor vehicle in order to determine a plausibility check result, so that the processed surroundings data generated by the motor vehicle include the plausibility check result.

[0042] The technical advantage achieved thereby, for example, is that the motor vehicle-generated environment data can be processed efficiently. The verification can, for example, include a test for plausibility. The verification can, for example, include the step of checking whether the information determined from the motor vehicle-generated environment data exactly matches the information determined from the infrastructure-generated environment data. The plausibility check can, for example, include the step of checking whether the information determined from the motor vehicle-generated environment data matches the information determined from the infrastructure-generated environment data within a tolerance range.

[0043] In one specific embodiment of the method according to the first aspect, it is provided that the processing includes cleaning erroneous data from the surroundings data generated by the motor vehicle, so that the processed surroundings data generated by the motor vehicle include cleaned surroundings data generated by the motor vehicle.

[0044] The technical advantage achieved thereby, for example, is that the environmental data generated by the motor vehicle can be processed efficiently. An error or problem can be, for example, that the environmental sensor of the motor vehicle is blocked and that correspondingly erroneous environmental sensor data is therefore present in the environmental data generated by the motor vehicle. The data error can be detected and removed by comparison with the environmental data of the infrastructure and replaced, for example, by correct environmental data of the infrastructure. One advantage is, for example, that the environmental data generated by the motor vehicle can also be used.

[0045] Cleaning thus includes, in particular, the removal of erroneous data and, in particular, the replacement of erroneous data with correct environmental data of the infrastructure (environmental data generated by the infrastructure).

[0046] In one embodiment of the method according to the first aspect, it is provided that the processing includes analyzing environmental data generated by the infrastructure in order to obtain environmental information about the surroundings of the motor vehicle, wherein the processing also includes enriching the environmental data generated by the motor vehicle with the environmental information, so that the processed environmental data generated by the motor vehicle includes the environmental information.

[0047] This achieves the technical advantage that, for example, the motor vehicle-generated environment data can be processed efficiently. The environment data of the infrastructure also include, for example, data about areas of the surroundings that the motor vehicle cannot detect with the aid of its environment sensor system, for example because these areas are not in the field of view of the environment sensor and / or there are obstructions in the scene, for example because a pedestrian is behind the bicycle (cyclist). These undetected areas can be added as environmental information to the motor vehicle-generated environment data in order to enrich the motor vehicle-generated environment data.

[0048] In one specific embodiment of the method according to the first aspect, it is provided that the environment data generated by the infrastructure include background data which represent the background of the surroundings, so that the environment data generated by the motor vehicle are processed based on the background data.

[0049] This achieves, for example, the technical advantage that the surroundings data generated by the motor vehicle can be processed efficiently.

[0050] Processing the surrounding data generated by the motor vehicle alone usually does not reveal the context in which the motor vehicle is located during its travel. For example, a certain traffic behavior in the surroundings of the motor vehicle only makes sense if the context is known. For example, if all motor vehicles in the surroundings of the motor vehicle are parked or driving on the left or right side of the road, this is a very unusual behavior without knowing the context. However, if the context of emergency rescue forces flashing blue lights on the road is known, this behavior makes sense, because the motor vehicle must make way for the emergency rescue forces flashing blue lights on the road.

[0051] In one implementation of the method according to the first aspect, it is provided that the background includes one or more of the following backgrounds: construction background, traffic background, event background, weather background, and road condition background.

[0052] This achieves, for example, the technical advantage that the surroundings data generated by the motor vehicle can be processed efficiently.

[0053] The construction context indicates, for example, whether a construction site is present in the surroundings of the motor vehicle.

[0054] The traffic context indicates, for example, how the light signal system is switched, ie, for example, which signal pattern is present, and / or whether there is a traffic diversion, for example, by traffic sign control, and / or whether the speed limit is changed.

[0055] The event context indicates, for example, whether there are any emergency vehicles operating in the vicinity of the motor vehicle.

[0056] The weather background indicates, for example, the weather in the surroundings of the motor vehicle.

[0057] The road condition context indicates, for example, whether there are one or more holes in the road, whether the road is icy and / or whether the road is wet.

[0058] In one embodiment of the method, it is provided that the environmental data generated by the motor vehicle are generated based on detecting the surroundings with the aid of an environmental sensor device of the motor vehicle, wherein at least one technical parameter of the environmental sensor device of the motor vehicle is received, in particular a range and / or a field of view, and wherein a data set is generated such that it includes at least one technical parameter of the environmental sensor device of the motor vehicle.

[0059] For example, the technical advantage achieved thereby is that the data set regarding processed environmental data generated by the motor vehicle contains additional information which can, for example, be used in an advantageous manner for testing at least one component of a process chain for an at least partially automated driving function of a motor vehicle and / or for generating at least a part of an algorithm for an at least partially automated driving function of a motor vehicle.

[0060] In one embodiment of the method according to the first aspect, it is provided that the environmental data generated by the infrastructure are generated based on detecting the surrounding environment with the help of an environmental sensor device of the infrastructure, wherein infrastructure environmental sensor device data are received, which infrastructure environmental sensor device data specify at least one technical parameter of the environmental sensor device of the infrastructure, wherein a data set is generated in this way, which data set includes at least one technical parameter of the environmental sensor device of the infrastructure.

[0061] For example, the technical advantage achieved thereby is that the data set regarding processed environmental data generated by the motor vehicle contains additional information which can, for example, be used in an advantageous manner for testing at least one component of a process chain for an at least partially automated driving function of a motor vehicle and / or for generating at least a part of an algorithm for an at least partially automated driving function of a motor vehicle.

[0062] In one embodiment of the method according to the first aspect, it is provided that the environmental data generated by the motor vehicle represent the respective surroundings observed by a plurality of motor vehicles from the perspective of the respective motor vehicles when they respectively travel through respective infrastructure areas of the infrastructure, wherein the environmental data generated by the infrastructure represent the respective surroundings of the motor vehicles observed from the perspective of the infrastructure.

[0063] The technical advantage achieved thereby, for example, is that vehicle-generated surroundings data of a plurality of motor vehicles are provided for generating a data set. Thus, a data set can be generated which is particularly suitable in an advantageous manner for the method according to the second aspect and / or according to the third aspect.

[0064] In one embodiment of the method according to the first aspect, it is provided that the corresponding infrastructure areas are respectively selected from one of the following groups of infrastructure areas: road nodes, construction sites, tunnels, highway entrances, highway exits, highways, bus stations, entrances, exits, narrow sections, narrow roads, narrow lanes, underpasses, locations with proximity to public transportation, in particular football stadiums or train stations.

[0065] For example, the technical advantage achieved thereby is that particularly important infrastructure areas are taken into account with regard to testing at least one component of a process chain for an at least partially automated driving function for a motor vehicle and / or with regard to generating at least a part of an algorithm for an at least partially automated driving function for a motor vehicle.

[0066] A road node is, for example, an intersection, a junction, a T-junction, a roundabout, or a highway intersection.

[0067] In one embodiment of the method according to the first aspect, it is provided that at least some of the respective infrastructure areas are identical to one another, wherein identical refers to the same infrastructure area type and the same infrastructure area, and / or wherein at least some of the infrastructure areas are different from one another, wherein different refers to the same infrastructure area type but different infrastructure areas, or different infrastructure area types.

[0068] This achieves the technical advantage, for example, that the surroundings data generated by the motor vehicle can be used by the same infrastructure area and / or by different infrastructure areas.

[0069] In one specific embodiment of the method according to the first aspect, it is provided that the data set is generated in such a way that it includes the surroundings data generated by the motor vehicle and / or the surroundings data generated by the infrastructure.

[0070] The technical advantage achieved thereby, for example, is that the data set is efficiently generated. According to this embodiment, it is provided that the data set does not only include processed environmental data generated by the motor vehicle, but also includes unprocessed environmental data generated by the motor vehicle and / or environmental data generated by the infrastructure, which are used to process the environmental data generated by the motor vehicle.

[0071] This advantageously provides a data set which is particularly suitable for the method according to the second aspect and / or the method according to the third aspect.

[0072] In one embodiment of the method according to the second aspect, it is provided that the at least one component comprises an algorithm for implementing at least a partial automated driving function, wherein the test comprises simulating a motor vehicle which is at least partially automated driving based on the at least partial automated driving function by implementing the algorithm.

[0073] This achieves the technical advantage, for example, that algorithms for implementing at least partial automated driving functions can be tested efficiently.

[0074] In one embodiment of the method, according to the second aspect, the environmental data generated by the motor vehicle are generated based on detecting the surrounding environment with the help of an environmental sensor device of the motor vehicle, wherein the data set includes at least one technical parameter of the environmental sensor device of the motor vehicle, in particular an effective range and / or a field of view, wherein the at least one component includes the environmental sensor device of the motor vehicle, wherein the test includes changing at least one technical parameter based on the data set in order to obtain at least one changed technical parameter, wherein the test includes simulating a motor vehicle with a simulated environmental sensor device with the help of the at least one changed technical parameter, the motor vehicle being at least partially autonomously driven based on at least a partial autonomous driving function by using the simulated environmental sensor device.

[0075] The technical advantage achieved thereby, for example, is that the surrounding sensor system of a motor vehicle can be tested efficiently. In particular, it can be tested in an advantageous manner whether the changed technical parameters allow further development. Thus, for example, a wider field of view and / or a larger effective range is sufficient so that the motor vehicle can be driven at least partially automatically more efficiently by using its surrounding sensor system. For example, if a small technical change to the surrounding sensor system of the motor vehicle can produce a similar effect, the algorithm for the at least partially automated driving function does not have to be further developed in a long and expensive manner. In one application case, the motor vehicle is not driven at least partially automatically well enough, so that the algorithm must be improved accordingly, wherein, however, in this specific application case, it is not the algorithm that is at fault, but the inadequate surrounding sensor system, which can be, for example, due to an insufficient effective range and / or an insufficient field of view, a pedestrian enters the visible range of the surrounding sensor system of the motor vehicle later, so that the motor vehicle brakes later. An increase in the effective range and / or field of view can alleviate this situation without having to further develop the algorithm in a long time.

[0076] In one embodiment of the method according to the third aspect, it is provided that at least part of the algorithm is generated by using an artificial neural network.

[0077] This achieves, for example, the technical advantage that at least a part of the algorithm can be generated efficiently.

[0078] According to one specific embodiment of the method according to the second aspect and / or according to the third aspect, it is provided that the data set includes the surroundings data generated by the motor vehicle and / or the surroundings data generated by the infrastructure.

[0079] This achieves the technical advantage, for example, that a particularly suitable data set can be used.

[0080] In one embodiment of the method according to the second aspect and / or according to the third aspect, it is provided that the data set is generated according to the method according to the first aspect.

[0081] This achieves, for example, the technical advantage that a particularly suitable data set is used.

[0082] Embodiments relating to the method according to the first aspect and / or according to the second aspect apply analogously to the method according to the third aspect, and vice versa.

[0083] The technical functions and features of the method according to the first aspect and / or according to the second aspect result analogously from the corresponding technical functions and features of the method according to the third aspect, and vice versa.

[0084] The embodiments described here can be combined with one another as desired, even if this is not specifically described.

[0085] This means, in particular, that features of embodiments of the method according to the first aspect can be used in embodiments of the method according to the second aspect and / or according to the third aspect, and vice versa.

[0086] Furthermore, device features result from corresponding method features and vice versa.

[0087] In one embodiment of the method according to the first aspect and / or according to the second aspect and / or according to the third aspect, the respective method is a computer-implemented method.

[0088] The at least partially automated driving function enables at least partially automated driving of a motor vehicle.

[0089] The expression "at least partially automated driving" includes one or more of the following: assisted driving, partially automated driving, highly automated driving, and fully automated driving. The expression "at least partially automated" includes one or more of the following: assisted, partially automated, highly automated, and fully automated.

[0090] Assisted driving means that the driver of the motor vehicle continuously performs lateral or longitudinal steering of the motor vehicle. The corresponding other driving tasks (i.e. lateral or longitudinal steering of the motor vehicle) are automatically performed. In other words, in assisted driving of the motor vehicle, lateral or longitudinal steering is automatically controlled.

[0091] Partially automated driving means that the longitudinal and lateral steering of the motor vehicle is automatically controlled in certain situations (e.g. driving on a motorway, driving in a parking lot, overtaking an object, driving in a lane defined by lane markings) and / or for a certain period of time. The driver of the motor vehicle does not have to manually control the longitudinal and lateral steering of the motor vehicle himself. However, the driver must constantly monitor the automatic control of the longitudinal and lateral steering so that he can intervene manually if necessary. The driver must be ready to fully take over the driving of the motor vehicle at any time.

[0092] Highly automated driving means that the longitudinal and lateral driving of a motor vehicle is automatically controlled in certain situations within a certain period of time (for example: driving on a highway, driving in a parking lot, overtaking an object, driving in a lane determined by lane markings). The driver of the motor vehicle does not have to manually control the longitudinal and lateral driving of the motor vehicle himself. The driver does not have to continuously monitor the automatically controlled longitudinal and lateral driving so that he can intervene manually when necessary. When necessary, a takeover request is automatically output to the driver to take over the control of the longitudinal and lateral driving, especially when sufficient time is reserved. In other words, the driver must be able to take over the control of the longitudinal and lateral driving. The automatic control limits of the lateral or longitudinal driving are automatically identified. It is not possible in highly automated driving to automatically achieve a state of risk minimization in any initial situation.

[0093] Fully automated driving means that in certain situations (e.g. driving on a highway, driving in a parking lot, overtaking an object, driving in a lane defined by lane markings), the longitudinal and lateral driving of a motor vehicle is automatically controlled. The driver of the motor vehicle does not have to manually control the longitudinal and lateral driving of the motor vehicle. The driver does not have to monitor the automatic control of the longitudinal and lateral driving so that he can intervene manually when necessary. Before the automatic control of the lateral or longitudinal driving is terminated, the driver is automatically requested to take over the driving task (control the lateral or longitudinal driving of the motor vehicle), especially if sufficient time is reserved. If the driver does not take over the driving task, the system automatically returns to the state with the lowest risk. The automatic control limits of the lateral or longitudinal driving are automatically identified. In all cases, it is possible to automatically return to the system state with the lowest risk.

[0094] The expression "at least one" includes the expression "one or more".

[0095] An environment sensor system within the meaning of this description, ie, in particular an environment sensor system of a motor vehicle and / or an environment sensor system of an infrastructure, comprises, for example, one or more environment sensors.

[0096] The surroundings sensors of the surroundings sensor system of the infrastructure are, for example, arranged in a spatially distributed manner within the infrastructure.

[0097] The surroundings sensors of the surroundings sensor system of the motor vehicle are, for example, arranged in a spatially distributed manner on the motor vehicle and / or are formed on the motor vehicle, ie are comprised by the motor vehicle.

[0098] An environmental sensor within the meaning of this description is, for example, one of the following environmental sensors: a radar sensor, a lidar sensor, an image sensor, in particular an image sensor of a camera, a magnetic field sensor, an infrared sensor and an ultrasonic sensor.

[0099] An infrastructure environment sensor device may be referred to as an infrastructure environment sensor device, for example.

[0100] An environment sensor system of a motor vehicle may be referred to, for example, as a motor vehicle environment sensor system.

[0101] Environmental data generated by a motor vehicle are therefore generated or produced, in particular, in the motor vehicle.

[0102] Infrastructure-generated environmental data are data that are generated or produced, in particular, within the infrastructure.

[0103] The terms "produce" and "generate" may be used synonymously.

[0104] A data set in the sense of the present description may, for example, also be referred to as a test data set, if the data set can be used, in particular, to test at least one component of a process chain for an at least partially automated driving function of a motor vehicle.

[0105] The test data set may in particular be used for the method according to the third aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Embodiments of the present invention are shown in the drawings and are explained in detail in the following description. In the drawings:

[0107] Figure 1 A flow chart showing a method according to the first aspect,

[0108] Figure 2 shows a flow chart of the method according to the second aspect,

[0109] Figure 3 shows a flow chart of a method according to the third aspect,

[0110] Figure 4 shows a motor vehicle driving through infrastructure,

[0111] Figure 5 A device according to a fourth aspect is shown, and

[0112] Figure 6 A computer-readable storage medium according to the sixth aspect is shown. DETAILED DESCRIPTION

[0113] Figure 1 A flow chart showing a method for generating a data set comprising the following steps:

[0114] receiving 101 environmental data generated by a motor vehicle, the environmental data generated by the motor vehicle representing the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle,

[0115] receiving 103 environment data generated by the infrastructure, the environment data generated by the infrastructure representing the surroundings of the motor vehicle as viewed from the perspective of the infrastructure,

[0116] Processing 105 the surroundings data generated by the motor vehicle based on the surroundings data generated by the infrastructure in order to determine processed surroundings data generated by the motor vehicle, and

[0117] A data set is generated 107 , which includes processed environmental data generated by the motor vehicle.

[0118] According to an embodiment of the method according to the first aspect, it is provided that the generated data set is output. According to an embodiment, the outputted data set is stored on a computer-readable storage medium. According to an embodiment, the outputting comprises sending the generated data set via a communication network.

[0119] Figure 2 A flow chart showing a method for testing at least one component of a process chain for an at least partially automated driving function of a motor vehicle comprises the following steps:

[0120] receiving 201 a data set comprising environmental data generated by a motor vehicle processed by using environmental data generated by an infrastructure, the environmental data generated by the infrastructure representing the surroundings of the motor vehicle as viewed from the perspective of the infrastructure, the environmental data generated by the motor vehicle representing the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle,

[0121] At least one component of the process chain is tested 203 based on the data set.

[0122] Figure 3 A flow chart showing a method for generating at least a part of an algorithm for an at least partially automated driving function of a motor vehicle, comprising the following steps:

[0123] receiving 301 a data set comprising environmental data generated by a motor vehicle processed by using environmental data generated by an infrastructure, the environmental data generated by the infrastructure representing the surroundings of the motor vehicle as viewed from the perspective of the infrastructure, the environmental data generated by the motor vehicle representing the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle,

[0124] At least a portion of an algorithm for an at least partially automated driving function of a motor vehicle is generated 303 based on the data set.

[0125] Figure 4 Motor vehicle 401 is shown driving, in particular during at least partially automated driving, on road 403 of infrastructure 405 .

[0126] The infrastructure may include, by way of example, one or more of the following infrastructure areas, which are represented by pictograms with corresponding reference numerals: intersection 407 , construction site 409 , tunnel 411 .

[0127] That is to say that in particular the road 403 can be driven through one or more of the aforementioned infrastructure areas 407 , 409 , 411 .

[0128] An environment sensor device 413 is arranged in the infrastructure 405, which is configured to monitor, for example, the road 403 and, for example, an infrastructure area, for example, one of the infrastructure areas 407, 409, 411, on the infrastructure side. The environment sensor device 413 exemplarily includes a camera 415 with an image sensor (not shown), a microphone 417, a temperature sensor 419, and a gas sensor 421, which is symbolically indicated by a nose symbol. In addition, three points are indicated by reference numeral 423, which means that the environment sensor device 413 can include other environment sensors.

[0129] Camera 415, microphone 417, temperature sensor 419 and gas sensor 421 are arranged in a spatially distributed manner within infrastructure 405. The environmental sensors detect their respective surroundings and output environmental sensor data corresponding to the detection. Environmental sensor data are environmental data generated by the infrastructure.

[0130] The environmental sensor data can be processed, for example, by means of first computer system 425, in order to detect objects in the corresponding surroundings of the environmental sensor. When an object is detected, information about the detected object, such as position, speed and direction of movement, can be sent to motor vehicle 401, for example, by means of first wireless communication interface 427, in order to assist the motor vehicle when it is driven through infrastructure 405 in an infrastructure-supported manner.

[0131] Motor vehicle 401 comprises an environment sensor system 429, which has a further camera 431, which has an image sensor, not shown, and a radar sensor 433. The environment sensor detects the surroundings 403 of the motor vehicle and provides environment sensor data corresponding to the respective detection. The environment sensor data are environment data generated by the motor vehicle.

[0132] The motor vehicle-generated surroundings data describe the surroundings of motor vehicle 401 as viewed from the perspective of motor vehicle 401 .

[0133] The infrastructure-generated environmental data of environmental sensors of infrastructure 405 , which detect the surroundings of motor vehicle 401 , describe the surroundings of motor vehicle 401 as viewed from the perspective of infrastructure 405 .

[0134] The motor vehicle 401 sends the motor vehicle-generated environment data to the computer center 435 via a wireless communication network, such as WLAN and / or mobile communication, and the computer center receives the environment data by means of a second wireless communication interface 437. As an alternative or in lieu of the wireless transmission of the motor vehicle-generated environment data from the motor vehicle 401 to the computer center 435, a wired or cable communication may be provided after the motor vehicle 401 has traveled, i.e. when the motor vehicle has arrived at the destination and stopped there, and / or the environment data may be stored on a machine-readable storage medium, such as a hard disk, in particular an SSD, wherein the machine-readable storage medium is sent to the computer center 435.

[0135] A second computer system 439 is arranged in the computer center 435, which processes the environmental data generated by the motor vehicle. The processing includes processing the environmental data generated by the motor vehicle. For the processing, the second computer system 439 uses the environmental data generated by the infrastructure, which are sent from the first computer system 425 to the computer center by means of the first wireless communication interface 427.

[0136] This processing is indicated by functional block 441. As a result, a data set 443 is generated which includes processed surroundings data generated by the motor vehicle.

[0137] Data set 443 can be used, for example, for simulation 445, according to which another motor vehicle 447 is at least partially autonomously driven through infrastructure 405. For the at least partially autonomous driving, an at least partially autonomous driving function is used, which is implemented by an algorithm. The algorithm can be generated once or can be further developed, so that the correspondingly generated or further developed algorithm can be tested within the framework of the simulation.

[0138] Furthermore, infrastructure-side environmental sensor system 413 and / or vehicle-side environmental sensor system 429, for example, can be tested and / or further developed based on data set 443. Infrastructure-side environmental sensor system 413, vehicle-side environmental sensor system 429, and algorithms are exemplary components of a process chain for at least partially automated driving functions of a motor vehicle.

[0139] Figure 5 A device 501 is shown which is configured to carry out all steps of the method according to the first aspect and / or according to the second aspect and / or according to the third aspect.

[0140] Figure 6A machine-readable storage medium 601 is shown, on which a computer program 603 is stored. The computer program 603 comprises instructions which, when executed by a computer, cause the computer to execute the method according to the first aspect and / or the second aspect and / or the third aspect.

[0141] In the known prior art, only information from the perspective of the vehicle is used for the testing or improvement of the algorithm.

[0142] Therefore, situations occurring outside the area detectable by the surroundings sensor system of the motor vehicle cannot be provided according to the prior art.

[0143] The approach described here provides that, in addition to the perspective of the motor vehicle, the perspective of the infrastructure is additionally used with respect to the respective traffic situation in order to process the surroundings data generated by the motor vehicle. The correspondingly generated data sets can be used, for example, for further development of the motor vehicle surroundings sensor system.

[0144] This makes it possible, for example, to avoid expensive and lengthy further development of algorithms or environmental sensors by making only minor hardware changes, such as a larger viewing range, for example an increase in the viewing range of 1 degree, and / or a larger effective range, for example an increase in the effective range of 20 m, which can only be determined by the tests based on the data set.

[0145] By using the surroundings data generated by the infrastructure, the surroundings data generated by the motor vehicle can be checked, for example, verified, within the scope of processing.

[0146] The context of the current traffic situation is usually not known from the surroundings data generated by the motor vehicle.

[0147] The background includes, for example:

[0148] Traffic control data (e.g. a light signal system changes its state (signal image), a lane is closed, the prescribed speed changes, a dynamic traffic sign changes its display, ...) and / or

[0149] Events that affect the surroundings, e.g. an ambulance is approaching from a distance and the traffic participants act accordingly and make room, which would be strange behavior if they were not aware of the approaching ambulance, and / or

[0150] Road condition data and / or construction site data and / or weather data

[0151] Environmental data within the meaning of this description include, for example, raw data, such as images, and / or evaluated data, such as object data from evaluated images).

[0152] As a result, a data set can be generated in an efficient manner, which data set is particularly suitable for generating and improving algorithms for at least partially automated driving functions of a motor vehicle.

[0153] The environmental data generated by the motor vehicle can be automatically verified, that is, is the environmental data correct?

[0154] This makes it possible to provide the algorithm with additional data (processed environmental data generated by the motor vehicle), ie the algorithm can be tested and / or verified based on these data.

[0155] This also makes it possible to verify / further develop triggering algorithms for recording surroundings data generated by the motor vehicle.

[0156] As a result, components of a process chain for at least partially automated driving functions of a motor vehicle can be optimally and efficiently developed and improved. For example, an optimal range of effectiveness and / or an optimal field of view of an ambient sensor can be determined.

Claims

1. A method for generating a data set, comprising the following steps: receiving (101) environmental data generated by a motor vehicle, the environmental data generated by the motor vehicle representing the surrounding environment of the motor vehicle as viewed from the perspective of the motor vehicle, receiving (103) environmental data generated by an infrastructure, the environmental data generated by the infrastructure representing the surroundings of the motor vehicle as viewed from the perspective of the infrastructure, processing (105) the vehicle-generated environmental data based on the infrastructure-generated environmental data in order to determine processed vehicle-generated environmental data, and A data set is generated (107) comprising the processed environment data generated by the motor vehicle.

2. The method according to claim 1, wherein: The processing includes a validation of the motor vehicle-generated surroundings data in order to determine a validation result, so that the processed motor vehicle-generated surroundings data include the validation result.

3. The method according to claim 1 or 2, wherein: The processing includes a plausibility check of the surroundings data generated by the motor vehicle in order to determine a plausibility check result, so that the processed surroundings data generated by the motor vehicle include the plausibility check result.

4. A method according to any one of the preceding claims, wherein: The processing includes cleaning erroneous data from the motor vehicle-generated surroundings data, so that the processed motor vehicle-generated surroundings data include cleaned motor vehicle-generated surroundings data.

5. A method according to any one of the preceding claims, wherein: The processing includes analyzing the environmental data generated by the infrastructure in order to obtain environmental information about the surroundings of the motor vehicle, wherein the processing also includes enriching the environmental data generated by the motor vehicle with the environmental information so that the processed environmental data generated by the motor vehicle include the environmental information.

6. A method according to any one of the preceding claims, wherein: The environment data generated by the infrastructure include background data, which represent the background of the surrounding environment, so that the environment data generated by the motor vehicle are processed based on the background data.

7. The method according to claim 6, wherein: The background includes one or more of the following backgrounds: construction background, traffic background, event background, weather background, and road condition background.

8. A method according to any one of the preceding claims, wherein: The environmental data generated by the motor vehicle are generated based on detecting the surroundings with the help of an environmental sensor system of the motor vehicle, wherein at least one technical parameter of the environmental sensor system of the motor vehicle is received, in particular a range and / or a field of view, and a data set is generated so that the data set includes at least one technical parameter of the environmental sensor system of the motor vehicle.

9. A method according to any one of the preceding claims, wherein: The environmental data generated by the infrastructure are generated based on detecting the surrounding environment with the help of an environmental sensor device of the infrastructure, wherein infrastructure environmental sensor device data are received, the infrastructure environmental sensor device data specifying at least one technical parameter of the environmental sensor device of the infrastructure, and wherein a data set is generated so that the data set includes at least one technical parameter of the environmental sensor device of the infrastructure.

10. A method according to any one of the preceding claims, wherein: The motor vehicle-generated environment data represent respective surroundings viewed from the perspective of respective motor vehicles when the plurality of motor vehicles respectively travel through respective infrastructure areas of the infrastructure, wherein the infrastructure-generated environment data represent respective surroundings of the motor vehicles viewed from the perspective of the infrastructure.

11. The method according to claim 10, wherein: The respective infrastructure areas are each selected from one of the following infrastructure areas: road nodes, construction sites, tunnels, motorway entrances, motorway exits, motorways, bus stops, entrances, exits, narrow sections, narrow roads, narrow lanes, underpasses, locations with proximity to public transport, in particular football stadiums or train stations.

12. The method according to claim 10 or 11, wherein: At least some of the respective infrastructure areas are identical to one another, wherein identical means identical infrastructure area type and identical infrastructure area, and / or wherein at least some of the respective infrastructure areas are different from one another, wherein different means that they may be of the same infrastructure area type but may be different infrastructure areas or may be different infrastructure area types.

13. A method according to any one of the preceding claims, wherein: A data set is generated such that it includes the surroundings data generated by the motor vehicle and / or the surroundings data generated by the infrastructure.

14. A method for testing at least one component of a process chain for an at least partially automated driving function of a motor vehicle, comprising the following steps: Receiving (201) a data set, the data set comprising environmental data generated by a motor vehicle processed by using environmental data generated by an infrastructure, the environmental data generated by the infrastructure representing the surroundings of the motor vehicle as viewed from the perspective of the infrastructure, the environmental data generated by the motor vehicle representing the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle; At least one component of the process chain is tested (203) based on the data set.

15. The method according to claim 14, wherein: The at least one component comprises an algorithm for implementing the at least partially automated driving functionality, wherein the test comprises simulating a motor vehicle that is at least partially automated driving based on the at least partially automated driving functionality by implementing the algorithm.

16. The method according to claim 14 or 15, wherein: The environmental data generated by the motor vehicle are generated based on detecting the surroundings with the help of an environmental sensor device of the motor vehicle, wherein the data set includes at least one technical parameter of the environmental sensor device of the motor vehicle, in particular an effective range and / or a field of view, wherein the at least one component includes the environmental sensor device of the motor vehicle, wherein the test includes changing the at least one technical parameter based on the data set in order to obtain at least one changed technical parameter, wherein the test includes simulating a motor vehicle with a simulated environmental sensor device with the help of the at least one changed technical parameter, the motor vehicle being at least partially autonomously driven based on the at least partially automated driving function by using the simulated environmental sensor device.

17. A method for generating at least a part of an algorithm for at least partially automated driving functions of a motor vehicle, comprising the following steps: receiving (301) a data set comprising environmental data generated by a motor vehicle processed by using environmental data generated by an infrastructure, the environmental data generated by the infrastructure representing the surroundings of the motor vehicle as viewed from the perspective of the infrastructure, the environmental data generated by the motor vehicle representing the surroundings of the motor vehicle as viewed from the perspective of the motor vehicle, At least a portion of an algorithm for an at least partially automated driving function of a motor vehicle is generated ( 303 ) based on the data set.

18. The method according to claim 17, wherein: At least a portion of the algorithm is generated using an artificial neural network.

19. The method according to any one of claims 14 to 18, wherein: The data set includes the surroundings data generated by the motor vehicle and / or the surroundings data generated by the infrastructure.

20. The method according to any one of claims 14 to 18, wherein: The dataset is generated according to the method of any one of claims 1 to 13.

21. An apparatus (501) arranged to carry out all the steps of the method according to any one of the preceding claims.

22. A computer program (603) comprising instructions which, when executed by a computer, cause the computer to execute the method according to any one of claims 1 to 20.

23. A machine-readable storage medium (601) on which a computer program (603) according to claim 22 is stored.

Images