Method for providing a modelable model for adjusting functions for

Abstract

The invention relates to a method for providing a transferable model of a control function for a technical system, comprising the following steps: providing the control function, the control function determining at least one output variable at least on the basis of input data and a frequency specified by the technical system, wherein the predetermined frequency is specific to a clock frequency of a hardware module of the technical system that performs the adjustment function, and a random distribution of the input data and the predetermined frequency is defined on the basis of a characteristic of at least one hardware that can perform the adjustment function, determining a transferable model of the adjustment function on the basis of the determined input data and the random distribution of the predetermined frequency using an uncertainty quantification method in order to model the random distribution of the at least one output variable, checking whether the random distribution of the at least one output variable meets at least one defined adjustment requirement, wherein a transferable model of the adjustment function is provided based on a result of the checking. The invention also relates to a computer program, a device and a storage medium for this purpose.

Description

Technical Field

[0001] The invention relates to a method for providing a transferable model of regulatory functions for a technical system. The invention also relates to a computer program, a device and a storage medium for this purpose. Background Art

[0002] Regulation functions, for example for vehicle or manufacturing applications, are mostly designed for a specific hardware environment in accordance with the specific frequency of the hardware and the intended functionality.

[0003] However, it is particularly disadvantageous here that the computing power of the hardware is predetermined and limited, and in the event of a hardware failure or malfunction, the control functions are no longer available or are available only to a limited extent. Summary of the invention

[0004] The subject matter of the invention is a method having the features of claim 1, a computer program having the features of claim 8, a device having the features of claim 9 and a computer-readable storage medium having the features of claim 10. Further features and details of the invention are derived from the respective dependent claims, the description and the drawings. Features and details described in conjunction with the method according to the invention are of course also suitable 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, so that reference is always made or can be made to each other for the disclosure of the various aspects of the invention.

[0005] The subject of the invention is in particular a method for providing a technical system with a transferable model of a control function, comprising the following steps, wherein these steps can be performed repeatedly and / or sequentially. In the context of the invention, transferable is understood in particular to mean that the model of the control function can be transferred to other technical systems or other data processing devices and / or can be implemented on multiple technical systems or multiple data processing devices. The technical system can be, for example, a vehicle or a manufacturing system in production. In the example of a vehicle as a technical system, the control function can be, for example, lane keeping control or distance control.

[0006] In a first step, a regulating function is preferably provided, wherein the regulating function determines at least one output variable based on at least input data and a frequency predetermined by the technical system. The regulating function can be provided as a mathematical model. The predetermined frequency is in particular specific to the clock frequency of a hardware module of the technical system that performs the regulating function. The hardware module can be, for example, an embedded system. The input data can be, for example, sensor data or include sensor data. The sensor data can be generated due to the detection of at least one sensor of the technical system or, for example, due to the detection of a sensor of another system related to the technical system. In the case of a vehicle as a technical system, another system related to the technical system can be, for example, a traffic light. The predetermined frequency can also be specific to the periodicity of the regulating function or can be predetermined or influence the periodicity. At least one output variable can be, for example, the trajectory of the vehicle or the horizontal distance from the driving lane or the distance from the vehicle in front.

[0007] In a further step, a random distribution of input data and a predetermined frequency is preferably defined based on the characteristics of at least one hardware that can perform the control function. In short, the random distribution of input data and predetermined frequencies is used to describe the fluctuation ranges that may occur due to implementation on different possible hardware environments. The characteristics are, for example, the corresponding clock frequency or the corresponding computing power of the hardware. Depending on the characteristics, these fluctuation ranges also describe fluctuations caused, for example, by different types of connection lines (such as wireless or wired) and can be predetermined by protocol and / or empirical values ​​or can be determined by corresponding trials.

[0008] In a further step, a transferable model of the control function is preferably determined based on the determined input data and the random distribution of the predetermined frequency using an uncertainty quantification method in order to model the random distribution of at least one output variable. The uncertainty quantification method can be, for example, a Monte Carlo simulation or Latin hypercube sampling or a chaotic polynomial-based scheme. The transferable model is therefore preferably constructed to determine the corresponding value of at least one output variable based on specific values ​​from the random distribution of the input data and the predetermined frequency, in particular with an uncertainty specification.

[0009] In a further step, it is preferably checked whether the random distribution of at least one output variable meets at least one defined control requirement. A transferable model of the control function can then be provided based on the result of this check. The at least one defined control requirement can be based on empirical values ​​or on a pre-determination from a standard or a law. One example is that at least one output variable is the distance between the vehicle and the vehicle traveling ahead, and it is checked whether the random distribution of this distance is above a threshold value pre-determined by a standard or a law. Another example is that in the case of lane keeping control as a control function, the distance between the vehicle and at least one lane predetermines the control requirement.

[0010] Thus, the method can advantageously provide the control function in the form of a portable model for different hardware environments, thereby enabling, for example, the control function to be outsourced or further executed in a backup hardware environment.

[0011] Furthermore, the method advantageously further comprises the following steps:

[0012] - A portable model for providing the regulation functionality based on an implementation on a technical system and / or on an external data processing device.

[0013] The migratable model can also be provided on another technical system. In other words, the migratable model of the regulating function is implemented on the technical system and / or on an external data processing device according to this step. The external data processing device can be, for example, a cloud server. Advantageously, the additional provision of the migratable model of the regulating function on the external data processing device can realize the backup of the regulating function, that is, the redundant second determination of at least one output variable. Here, the external data processing device can, for example, access additional input data. In the case where the technical system is a vehicle, the external data processing device can, for example, advantageously access traffic data such as traffic light data and use the traffic data as additional input data for the migratable model of the regulating function. Alternatively, the migratable model of the regulating function can also be implemented and executed only on the external data processing device, so as to, for example, cyclically transmit only the current value of at least one output variable to the technical system. Thereby, the computational workload in the technical system can be advantageously reduced.

[0014] Furthermore, optionally within the scope of the present invention, the method may further comprise the following steps:

[0015] - executing the implemented transferable model on a technical system and / or on an external data processing device,

[0016] - Checking whether the output of the executed transferable model meets at least one defined reconciliation requirement.

[0017] The output is in particular a corresponding value of at least one output variable. This advantageously makes it possible to check whether the transferable model is suitable for use on a technical system and / or on an external data processing device.

[0018] According to a further possibility, provision can be made for a control and / or regulation of the technical system to be carried out based on the output of the executed transferable model.

[0019] Furthermore, it is conceivable that the migratable model is determined, within a certain range, in particular as a function of changes in the frequency of the underlying hardware during implementation. This dependence on changes in frequency can be referred to as sensitivity and in this context expresses in particular how much the respective migratable model reacts to changes in the frequency of the hardware implementing the migratable model.

[0020] It can further be advantageously provided that the determination comprises the following steps:

[0021] - determine a defined number of samples from a defined random distribution of input data and pre-defined frequencies using uncertainty quantification methods,

[0022] - calculating a corresponding value of at least one output variable based on the determined input data and a defined number of samples of a predefined frequency.

[0023] Uncertainty quantification methods can be based on chaotic polynomial schemes. In this case, in particular, the output variable y is represented as a polynomial, such as

[0024]

[0025] ψ k For example, a known statistical distribution is used. The determination of a transferable model for the output variable y then comprises in particular the determination of the coefficients y as a function of the calculated values ​​of the output variable k .

[0026] For example, it can be provided that the technical system is a vehicle and that control and / or regulation of vehicle functions of the vehicle are provided based on transferable regulation functions. The control of vehicle functions can be, for example, automatic steering or braking. Thus, the method according to the 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 an autonomous vehicle. The vehicle can have a vehicle device, for example, for providing at least partially automatic driving functions and / or driver assistance systems. The vehicle device can be designed to control and / or accelerate and / or brake and / or steer the vehicle at least partially automatically.

[0027] The subject of the 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 perform the method according to the invention. The computer program according to the invention thus brings the same advantages as described in detail with reference to the method according to the invention.

[0028] The subject matter of the invention also relates to a device for data processing, which is configured to carry out the method according to the invention. As such a device, for example, a computer can be provided which executes a computer program according to the invention. The computer can have at least one processor for executing the computer program. A non-volatile data memory can also be provided, in which the computer program is stored and from which the processor can read the computer program for execution.

[0029] The subject matter of the present invention may also relate to a computer-readable storage medium having a computer program according to the present invention and / or comprising instructions which, when executed by a computer, cause the computer to perform the method according to the present invention. The storage medium is, for example, constructed 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, for example, be integrated into a computer.

[0030] In addition, the method according to the present invention can also be designed as a computer-implemented method. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Further advantages, features and details of the invention are apparent from the following description, in which exemplary embodiments of the invention are described in detail with reference to the accompanying drawings. The features mentioned in the claims and the description may be essential to the invention either individually or in any combination.

[0032] Figure 1 Schematic visualizations of methods, technical systems, hardware modules, external data processing devices, devices, storage media and computer programs according to embodiments of the present invention are shown,

[0033] Figure 2 A schematic visualization of a method according to an embodiment of the invention is shown. DETAILED DESCRIPTION

[0034] Figure 1 A method 100 , a technical system 1 , a hardware module 2 , an external data processing device 3 , a device 10 , a storage medium 15 and a computer program 20 according to an embodiment of the present invention are shown.

[0035] Figure 1 In particular, an exemplary embodiment of a method 100 for providing a technical system 1 with a transferable model 8 of a control function 4 is shown. In a first step 101, a control function 4 is provided, wherein the control function 4 determines at least one output variable 7 based on at least input data 5 and a frequency 6 predefined by the technical system 1, wherein the predefined frequency 6 is specific to a clock frequency of a hardware module 2 of the technical system 1 that executes the control function 4. In a second step 102, a random distribution of the input data 5 and the predefined frequency 6 is defined based on characteristics of at least one hardware that can execute the control function 4. In a third step 103, a transferable model 8 of the control function 4 is determined based on the determined random distribution of the input data 5 and the predefined frequency 6 using an uncertainty quantification method 9 in order to model the random distribution of the at least one output variable 7. In a fourth step 104, it is checked whether the random distribution of the at least one output variable 7 meets at least one defined control requirement, wherein a transferable model 8 of the control function 4 is provided based on the result of the check.

[0036] Figure 2A schematic visualization of a method according to an embodiment of the invention is shown. Here, first a control function 4 is provided, which determines an output variable 7 based on input data 5 and a predetermined frequency 6. Using an uncertainty quantification method 9 and based on the definition of a random distribution of the input data 5 and the predetermined frequency 6, a transferable model 8 can be determined in order to model the random distribution of the output variable 7.

[0037] The regulation functions 4 in the distributed system should preferably be able to run on different hardware environments and have a small impact on the changing frequency 6 of the respective hardware environments. For example, for distributed regulation functions 4 such as cross-section control or edge-controlled manufacturing, it would be advantageous to provide the controller in the cloud or to provide a similar backup solution.

[0038] According to an embodiment of the present invention, uncertainty quantification methods 9 can be used to determine a transferable model 8 of a control function 4 from a fixed control function 4 using a random distribution of input data and a predetermined frequency 6. The transferable model 8 can advantageously be implemented and executed on different hardware with different frequencies 6. Therefore, the transferable model 8 of the control function 4 should preferably not be set to a frequency 6, but should preferably be able to run on a plurality of different frequencies 6, in particular with low sensitivity to frequency changes. The transferable model 8 of the control function 4 can then be checked with respect to the random distribution of at least one output variable 7 compared to a defined control requirement (i.e., in particular an allowed range), and preferably can also be checked with respect to sensitivity (e.g., sensitivity of the frequency value distribution).

[0039] Almost every complex function requires, in particular, a closed control loop in order to react to and / or control different actuator data or sensor data. If the control function 4 is not used on a fixed system but on a distributed system, a flexible system structure (e.g. cloud / edge) or a regional architecture, the resources in terms of the hardware environment of the control function 4 may vary and alternative solutions with similar values ​​for at least one output variable 7 may be required.

[0040] An existing mathematical model of a control function 4 can form the basis for a transformation into a transferable model 8 of the control function 4. First, the input data and the predetermined frequencies 6 are preferably defined as random distributions. These areas are, for example, corresponding environmental changes, for example depending on hardware changes that the control function 4 should be able to cope with. The existing model of the control function 4 is then preferably combined with these random distributions and used to generate the transferable model 8 using an uncertainty quantification method 9 (for example, based on polynomial chaos). The transferable model 8 preferably calculates at least one output variable 7 mainly as a random distribution, but it is also possible to calculate the value of this output variable 7 from specific frequency values. The resulting random distribution of the at least one output variable 7 can then be checked against the defined control requirements, in particular the permissible range of the at least one output variable 7, in order to determine the suitability.

[0041] One possible application is, for example, the use of the same control function 4 on different computing platforms or data processing devices which provide different frequencies 6 and possibly additional information. In one possible configuration, the vehicle can be controlled as a technical system 1 via the cloud as an external data processing device 3 and can call up additional traffic light information even if the vehicle's direct view of the corresponding traffic light is blocked.

[0042] According to an embodiment, a further optional step may be to use the knowledge about the inherent sensitivity of the transferable model 8 to select a transferable model 8 with low sensitivity to frequency changes, so that the transferable model 8 is less affected by changes in the underlying hardware. Another possibility may be to determine the transferable model 8 from the noisy input data 5 and / or the frequency 6 in order to take into account real interferences and make the functional behavior of the transferable model 8 closer to reality.

[0043] The above explanation of the embodiment describes the present invention only within the scope of the example. Of course, if technically reasonable, the individual features of the embodiment can be freely combined with each other without departing from the scope of the present invention.

Claims

1. A method (100) for providing a transferable model (8) of a regulation function (4) for a technical system (1), comprising the following steps: - providing (101) a control function (4), wherein the control function (4) determines at least one output variable (7) based at least on input data (5) and a frequency (6) predetermined by the technical system (1), wherein the predetermined frequency (6) is specific to a clock frequency of a hardware module (2) of the technical system (1) that executes the control function (4), - defining (102) a random distribution of the input data (5) and the predetermined frequency (6) based on properties of at least one hardware capable of performing the regulating function (4), - determining (103) a transferable model (8) of the control function (4) based on the determined random distribution of the input data (5) and the predetermined frequency (6) using an uncertainty quantification method (9) in order to model the random distribution of the at least one output variable (7), - checking (104) whether the random distribution of the at least one output variable (7) meets at least one defined control requirement, wherein a transferable model (8) of the control function (4) is provided based on the result of the checking (104).

2. The method (100) according to claim 1, It is characterized in that The method (100) further comprises the following steps: - providing a transferable model (8) of the regulation function (4) based on an implementation on the technical system (1) and / or on an external data processing device (3).

3. The method (100) according to claim 2, It is characterized in that The method (100) further comprises the following steps: - executing the implemented transferable model (8) on the technical system (1) and / or on the external data processing device (3), - Checking whether the output of the executed transferable model (8) meets the at least one defined regulation requirement.

4. The method (100) according to claim 3, It is characterized in that Control and / or regulation of the technical system (1) is performed based on the output of the executed transferable model (8).

5. The method (100) according to any one of the preceding claims, It is characterized in that In the context of the determination (103), the transferable model (8) is determined, in particular, as a function of a change in the frequency (6) of the underlying hardware during implementation.

6. The method (100) according to any one of the preceding claims, It is characterized in that The determining (103) comprises the following steps: - using the uncertainty quantification method (9) to determine a defined number of samples from a defined random distribution of the input data (5) and the predetermined frequency (6), - calculating a corresponding value of the at least one output variable (7) based on the determined input data (5) and a defined number of samples of the predetermined frequency (6).

7. The method (100) according to any one of the preceding claims, It is characterized in that The technical system (1) is a vehicle and provides control and / or regulation of vehicle functions of the vehicle based on the transferable model (8).

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

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

Images