Method for functionally improving environment modeling in a vehicle assistance system and

Abstract

A method (100) for functionally improving environment modeling in an assistance system (2) of a vehicle (1), comprising the steps of: providing (101) at least two computational paths (3) of the assistance system (2), the computational paths (3) comprising respective environment detections; ascertaining (102), from each calculation path, at least one behavior option for the vehicle (1) on the basis of the respective environment detection; exchanging (103) the respective environmental detection results between the respective computational paths (3); revising (104) the at least one behavior option of the at least one computational path (3) on the basis of the result of the context detection of the respective exchange; a target behavior option for the assistance system (2) is determined (105) based on a result of the revising (104). The invention further relates to a computer program, a device and a storage medium for this purpose.

Description

Technical Field

[0001] The present invention relates to a method for improving the functionality of environment modeling in an assistance system of a vehicle. The invention also relates to a computer program, a device and a storage medium for this purpose. Background Art

[0002] Accurately detecting the environment, including all relevant traffic participants, obstacles, and boundaries, is a crucial component of driver assistance systems. This requires various environmental sensors, such as cameras, radar sensors, lidar sensors, or ultrasonic sensors. These sensors have varying strengths and weaknesses due to their physical properties, installation locations, algorithms, or aging. Consequently, current assistance systems often utilize multiple, partially dissimilar sensors. A fusion algorithm, based on this, combines the sensor data, taking into account the various strengths and weaknesses of the sensors. Its goal is to create a mapping of the real environment that is as error-free as possible, encompassing all relevant object characteristics. Within the scope of this invention, this combination is referred to as environmental detection. The requirements for environmental detection increase with the increasing complexity of automated driving systems in terms of functional safety (FuSa) and safety of intended functions (SOTIF). Higher levels of automation often require diverse implementations to meet safety requirements, including for the aforementioned fusion algorithms. While various existing approaches provide multiple paths, these paths remain completely separate. This can lead to errors, for example, caused by the same source. Summary of the Invention

[0003] The present invention relates to a method, a computer program, a device, and a machine-readable storage medium for improving the functionality of environment modeling in an assistance system of a vehicle. Further features and details of the invention are apparent from the following description and the accompanying drawings. Features and details described in conjunction with the method according to the invention also apply in conjunction with the computer program according to the invention, the device according to the invention, and the computer-readable storage medium according to the invention, and vice versa, respectively, so that cross-references are always possible within the scope of the present invention.

[0004] The subject matter of the present invention is, in particular, a method for improving the functionality of environment modeling in an assistance system of a vehicle, the method comprising the following steps, wherein these steps can be performed repeatedly and / or sequentially. The assistance system is, in particular, a multipath assistance system, i.e., an assistance system having at least two redundant calculation paths. The assistance system can be designed for at least partially automated vehicle control or for providing at least one driving function, and for this purpose, for example, uses a processor for calculations and has various software modules.

[0005] In a first step, at least two calculation paths for the assistance system are preferably provided. These at least two calculation paths are, in particular, two redundant calculation paths and provide the same driving function for the vehicle, such as control or range control. The calculation paths particularly include respective environmental detections, wherein sensor data are analyzed and fused within the scope of these environmental detections. During the fusion of the sensor data, a representation of the vehicle's environment is generated, particularly based on the analyzed sensor data. The sensor data are preferably obtained from detections performed by at least one sensor of the vehicle. The at least one sensor can be, for example, a camera, a radar sensor, an ultrasonic sensor, and / or a lidar sensor.

[0006] In a next step, at least one behavior option for the vehicle is preferably determined from each of the at least two calculation paths based on the respective environmental detection. The behavior options include, in particular, specific vehicle control methods or ways of implementing driving functions, such as values ​​for parameters of the driving functions. For example, the environmental detection is taken into account or used as a basis by controlling the vehicle based on the behavior option to enter a free space determined based on the environmental detection.

[0007] In the next step, the respective environmental detection results are preferably exchanged between the various computation paths. In particular, a one-sided exchange can also be provided, whereby only one computation path provides the corresponding environmental detection result to at least one other computation path. It is also conceivable that each computation path provides its own environmental detection result to all other computation paths. This exchange can, for example, involve data exchange between corresponding software modules. The environmental detection result can, for example, be an environmental representation generated by fusing analyzed sensor data.

[0008] In a next step, at least one behavior option of at least one calculated path is preferably revised based on the corresponding exchanged environmental detection results, wherein at least one safety requirement, in particular all safety requirements, of the corresponding calculated path are taken into account in order to provide a functional improvement in the environmental modeling in the assistance system. A safety requirement may be, for example, the distance to a vehicle traveling ahead, or the distance to a guardrail, or compliance with a speed limit. The revision may, for example, be an adaptation of parameter values ​​of a driving function, in particular an adaptation of a trajectory when the behavior option is implemented as a trajectory. Furthermore, within the scope of the revision, the acceleration or braking process of the vehicle may change. At the end of the revision, a verification of the at least one behavior option may be performed, wherein an index is determined and established as to whether the at least one behavior option meets at least one safety requirement, in particular all safety requirements.

[0009] In a next step, a target behavior option for the assistance system is preferably determined based on the results of the revision. The results of the revision can, for example, index behavior options that have been validated with respect to at least one safety requirement. Within the scope of determining the target behavior option, a trade-off between at least two behavior options can also be performed, and one of the at least two behavior options can be selected while taking into account the at least one safety requirement.

[0010] It can further advantageously be provided that one of the at least two calculated paths is a main path and at least one other calculated path is a secondary path, wherein the main path is preferentially used for controlling the vehicle. The secondary path is therefore particularly a backup path, which can be employed, for example, when an error is detected in the main path. In this case, the at least one secondary path is preferably switched to the main path, in particular, only from the at least one secondary path. In an alternative embodiment, the main path can be switched to the at least one secondary path when an error is detected in the main path.

[0011] It is also advantageous if the respective environmental measurements of the calculation paths differ from one another in terms of the processing of the sensor data and / or the sensors used. Therefore, in particular, there is no homogeneous redundancy between at least two calculation paths. Thus, the data base serving as the basis for revising at least one action option and determining a target action option can be advantageously expanded, since at least two calculation paths or their respective environmental measurements are taken into account.

[0012] In another possibility, it can be provided that at least one behavior option describes a trajectory to be driven by the vehicle. The trajectory is in particular a planned movement of the vehicle through space and can also include planning of acceleration and / or braking processes of the vehicle.

[0013] Furthermore, within the scope of the present invention, it is also conceivable that at least two calculation paths each process at least two data inputs, wherein each data input is provided for a corresponding environmental detection. In this case, preferably only one environmental detection associated with the corresponding calculation path is used to determine at least one behavioral option for the vehicle. Thus, the independent determination of at least one behavioral option can advantageously be performed initially via the at least two calculation paths.

[0014] Furthermore, it is advantageous within the scope of the present invention if the revision comprises the following steps:

[0015] Based on the evaluation of the exchanged corresponding environmental detection results, at least one determined behavior option is checked while taking into account at least one safety requirement, in particular all safety requirements, of the corresponding calculation path.

[0016] This check can also be referred to as verification and includes, for example, an analysis of the determined trajectory, ie, the action options, with respect to the distance to the preceding vehicle, ie, with respect to the safety requirements.

[0017] It may be provided within the scope of the present invention that the determination comprises the following steps:

[0018] - selecting one of the at least two calculation paths based on the result of said checking, in order to use the selected calculation path to determine a target behavior option for the assistance system.

[0019] Thus, when one of the calculation paths is the primary path, provision can be made for this primary path to be selected by default. However, if, for example, an error is determined in the primary path based on the environmental monitoring, at least one further calculation path can be selected. Provision can be made for the at least one further calculation path to use the environmental monitoring of the primary path, taking into account the error, in order to determine a target behavior option for the assistance system, despite the presence of an error in the primary path.

[0020] The method according to the present invention can be used in a vehicle. The vehicle can be configured, for example, as a motor vehicle and / or a passenger vehicle and / or as an at least partially autonomous or partially automated vehicle. The vehicle can include vehicle facilities and / or assistance systems, for example, for providing at least partially autonomous or partially automated driving functions. The vehicle facilities can be implemented to at least partially automatically control and / or accelerate and / or brake and / or steer the vehicle.

[0021] The present invention also relates to a computer program, in particular a computer program product, comprising instructions which, when executed by a computer, cause the computer to carry out the method according to the invention. The computer program according to the invention thus offers the same advantages as those described in detail with respect to the method according to the invention.

[0022] The present invention also provides a data processing device configured to implement the method according to the present invention. For example, the device may include a computer that executes a computer program according to the present invention. The computer may include at least one processor for executing the computer program. A non-volatile data memory may also be provided, in which the computer program can be stored and from which the processor can read the computer program for execution.

[0023] Similarly, the subject matter of the present invention may also be a computer-readable storage medium having a computer program according to the present invention and / or including instructions that, when executed by a computer, cause the computer to carry out the method according to the present invention. The storage medium may be configured, for example, as a data storage device, such as a hard disk and / or a non-volatile memory and / or a memory card. The storage medium may be integrated into the computer, for example.

[0024] Furthermore, the method according to the present invention can be implemented as a computer-implemented method.

[0025] Further advantages, features and details of the present invention can be found in the following description, in which exemplary embodiments of the present invention are described in detail with reference to the accompanying drawings. The features mentioned in the present invention may be essential to the invention individually or in any combination. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings show:

[0027] Figure 1 Schematic visualization of a method, a vehicle with an assistance system and sensors, a device, a storage medium, and a computer program according to an embodiment of the invention,

[0028] Figure 2 Schematic diagram of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] exist Figure 1 Schematically depicts a method 100 according to an exemplary embodiment of the present invention, a vehicle 1 with an assistance system 2 and a sensor 3 , a device 10 , a storage medium 15 , and a computer program 20 .

[0030] Figure 1In particular, a method 100 for improving the functional modeling of an assistance system 2 of a vehicle 1 is shown. In a first step 101, at least two calculation paths 3 of the assistance system 2 are provided, wherein the calculation paths 3 include respective environmental detections, and within the scope of these environmental detections, an analysis and fusion 4 of sensor data 5a is performed. This provision may include accessing a data memory and executing at least one software module of the assistance system 2. The sensor data 5a is obtained from detections by at least one sensor 5 of the vehicle 1. In a second step 102, at least one behavior option for the vehicle 1 is determined from each of the at least two calculation paths 3 based on the respective environmental detections, in particular by executing at least one software module by a processor of the assistance system 2. In a third step 103, the respective environmental detection results are exchanged between the individual calculation paths 3, in particular by executing at least one software module by a processor of the assistance system 2. In a fourth step 104, at least one behavior option of at least one calculation path 3 is revised based on the corresponding exchanged environmental detection results, taking into account at least one safety requirement, in particular all safety requirements, of the corresponding calculation path 3, in order to provide a functional improvement to the environmental modeling of the assistance system 2. This step is performed in particular by executing at least one software module by the processor of the assistance system 2. In a fifth step 105, a target behavior option for the assistance system 2 is determined based on the result of the updating 104, in particular by executing at least one software module by the processor of the assistance system 2.

[0031] Figure 2 A method according to an embodiment of the present invention is schematically illustrated. In the two illustrated calculation paths 3, an environment detection is first performed based on sensor data 5a and a fusion step 4. The results of these environment detections are exchanged between the calculation paths 3. In a subsequent step 102, at least one behavior option is determined from each calculation path 3 by a planning module 6, taking both environment detections into account. In a subsequent step 105, a target behavior option is selected by the planning module 6. The planning module 6 can be a software module and is implemented by a processor of the assistance system 2.

[0032] According to an exemplary embodiment, the present invention describes a method for combining fusion concepts of calculation paths 3 or for jointly using alternative fusion concepts of calculation paths 3 in order to increase safety with respect to SOTIF, for example, while avoiding “common causes”.

[0033] A prerequisite for the method according to the exemplary embodiment is, in particular, a safety architecture having two functional computation paths 3, both of which include different environmental detections, including sensor systems 2 and fusion 4. Such multi-path systems are already known, in particular for L3 and L4 solutions with a large functional scope.

[0034] The differences in environmental detection can include different sensor configurations, different processing, or both. For SOTIF benefits, there must preferably be certain differences, i.e., there is no purely homogeneous redundancy. One aspect of the present invention according to an embodiment is, in particular, to utilize environmental detection results from one calculation path 3 in another calculation path to reduce the frequency or risk of erroneous reactions. At the same time, this can advantageously eliminate errors that occur simultaneously in both calculation paths 3 and eliminate the possibility that the entire system or auxiliary system 2 can no longer be used. This ensures that errors in one calculation path 3 do not negatively affect other calculation paths 3 at the same time.

[0035] The basic premise for using this method is two redundant calculation paths 3. For reasons of functional safety, this basic premise may often exist for L3 / L4 systems or vehicles 1. In the event of an error, it should preferably be ensured that operation continues until a safer state is reached. In order to exclude errors caused by common causes (English: "common causes"), a variety of calculation paths 3 can provide assistance. If there are a variety of calculation paths, the method according to the embodiment can be particularly advantageous. Another aspect of safety is in particular the safety of the intended function (SOTIF). In short, the system, i.e. in particular the vehicle 1, preferably must function well enough for this purpose so that critical scenarios can also be overcome (in the absence of errors from functional safety). In order to function well enough, it may be desirable to utilize the results of the diverse environmental detection of at least two calculation paths 3 when implementing normal functions (in normal operation).

[0036] Therefore, the present invention provides according to an embodiment that the environmental detection results of the calculation path 3 are provided to the other calculation paths respectively. In order to exclude "common cause" errors, the results are preferably not simply merged. Therefore, the calculation path 3 using these results must preferably be able to cope with two inputs. The input of the own calculation path 3 preferably has priority. Preferably, the input of the own calculation path 3 is used alone to obtain at least one behavior option, for example by generating a trajectory bundle (Trajektorienbündels). The possible behavior options are preferably checked with a second input. The behavior option that is considered safe by both inputs will be preferred. If in an extreme scenario it occurs that no calculated behavior option meets the boundary conditions of the second input, then this can be interpreted as the first sign of an error. Since the same method is executed in parallel in the second calculation path 3, in this case, if a trajectory that is considered safe according to the environmental detection has been found in the second calculation path 3, the calculation path 3 can be replaced.

[0037] The above explanations of the embodiments describe the present invention only within the scope of examples. Of course, individual features of the embodiments can be freely combined with one another without departing from the scope of the present invention, as long as they are technically meaningful.

Claims

1. A method (100) for improving the functionality of environment modeling in an assistance system (2) of a vehicle (1), comprising the following steps: - providing (101) at least two calculation paths (3) of the auxiliary system (2), wherein, The calculation path (3) includes a respective environmental detection, wherein within the scope of the environmental detection, an analysis and fusion (4) of sensor data (5a) is performed, wherein the sensor data (5a) is obtained from detection by at least one sensor (5) of the vehicle (1), - determining (102) at least one behavior option for the vehicle (1) from each of the at least two calculation paths (3) based on a respective environmental detection, - exchanging (103) respective environmental detection results between the various computation paths (3), - revising (104) at least one behavior option of at least one of the calculation paths (3) based on the respectively exchanged environmental detection results, wherein at least one safety requirement, in particular all safety requirements, of the corresponding calculation path (3) are taken into account in order to provide a functional improvement of the environmental modeling in the assistance system (2), - determining (105) a target behavior option for the assistance system (2) based on the result of the revision (104).

2. The method (100) according to claim 1, characterized in that: One of the at least two calculated paths (3) is a main path and at least one other calculated path (3) is an auxiliary path, wherein the main path is used preferentially for controlling the vehicle (1), wherein the switching (103) is performed by at least one of the auxiliary paths to the main path.

3. The method (100) according to any one of the preceding claims, characterized in that: The respective environmental detections of the calculation paths (3) differ from one another in terms of the processing of the sensor data (5a) and / or in terms of the sensors used.

4. The method (100) according to any one of the preceding claims, characterized in that: The at least one behavior option describes a trajectory to be driven by the vehicle (1).

5. The method (100) according to any one of the preceding claims, characterized in that: The at least two calculation paths (3) each process at least two data inputs, wherein each data input is respectively provided for a corresponding environmental detection, wherein only the environmental detection assigned to the corresponding calculation path (3) is used to determine (102) the at least one behavior option for the vehicle (1).

6. The method (100) according to any one of the preceding claims, characterized in that: Said revision (104) comprises the following steps: Based on the evaluation of the respective exchanged environmental detection results, at least one determined behavior option is checked taking into account at least one safety requirement, in particular all safety requirements, of the corresponding calculation path (3).

7. The method (100) according to claim 6, characterized in that: The determining (105) comprises the following steps: - selecting one of the at least two calculation paths (3) based on the result of the checking, in order to use the selected calculation path (3) to determine the target behavior option for the assistance system (2).

8. A computer program (20) comprising instructions which, when executed by a computer (10), cause the computer to carry out the method (100) according to any one of the preceding claims.

9. A device (10) for data processing, which is configured to carry out the method (100) according to any one of claims 1 to 7.

10. A computer-readable storage medium (15) comprising instructions which, when executed by a computer (10), cause the computer to perform the method (100) according to any one of claims 1 to 7.

Images