DE102023203910A1 - METHOD, APPARATUS AND COMPUTER-READABLE STORAGE MEDIUM FOR SUPPORTING VEHICLE TESTING - Google Patents

Abstract

The present disclosure provides a method for supporting vehicle testing, comprising: inputting one or more parameters from the vehicle into a regression algorithm; Calculating a predicted output value using the regression algorithm based on the one or more parameters, the predicted output value serving to support a test of overall vehicle control of the vehicle.

Description

FIELD OF THE INVENTION

The present disclosure relates to the field of vehicle testing, more specifically to a method, an apparatus and a computer-readable storage medium for supporting vehicle testing.

STATE OF THE ART

During the research and development stage of a vehicle, various characteristics of the vehicle must be tested. For example, after developing a total vehicle control (VCU) of the vehicle to check the performance of the developed overall vehicle control, it is conceivable to transmit current and/or torque and other parameters as input to the overall vehicle control through hardware-in-the-loop (HiL) in order to monitor the performance of the vehicle To test the entire vehicle control under such conditions. The entire vehicle control to be tested receives the current and/or the torque and other parameters primarily by connecting to the vehicle in order to obtain the current and/or the torque and other parameters directly from the vehicle, or by connecting to a computer in order to to obtain the current and/or torque and other parameters resulting from a simulation based on a mathematical model.

However, when testing the overall vehicle control of a vehicle using the above method, there is a further improvement in terms of, among other things, flexibility and efficiency are necessary.

DISCLOSURE OF INVENTION

In view of the need for improvement in the prior art, embodiments of the present disclosure provide a method, an apparatus, and a computer-readable storage medium for supporting vehicle testing.

According to a first aspect, an embodiment of the present disclosure provides a method for supporting vehicle testing, comprising: inputting one or more parameters from the vehicle into a regression algorithm; Calculating a predicted output value using the regression algorithm based on the one or more parameters, the predicted output value serving to support a test of overall vehicle control of the vehicle.

According to a second aspect, an embodiment of the present disclosure provides an apparatus for supporting vehicle testing, comprising: at least one processor; and a memory that communicates with the at least one processor and in which executable code is stored, the executable code, when executed by the at least one processor, causing the at least one processor to execute the method according to the first aspect.

According to a third aspect, an embodiment of the present disclosure provides a computer-readable storage medium that stores executable code that, when executed, causes a computer to execute the method of the first aspect.

According to a fourth aspect, an embodiment of the present disclosure provides a system for vehicle testing, comprising: an overall vehicle control to be tested; and a device for executing the method according to the first aspect or a device for supporting a vehicle test according to the second aspect, wherein the predicted output value is used to support a test of an overall vehicle control of the vehicle.

DESCRIPTION OF THE FIGURES

• 1 zeigt ein Ablaufdiagramm eines Verfahrens zum Unterstützen einer Fahrzeugprüfung nach der vorliegenden Offenbarung.
• 2 zeigt ein Ablaufdiagramm eines anderen Verfahrens zum Unterstützen einer Fahrzeugprüfung nach der vorliegenden Offenbarung.
• 3 zeigt eine schematische Darstellung einer Vorrichtung zum Unterstützen einer Fahrzeugprüfung nach der vorliegenden Offenbarung.
• 4 zeigt eine schematische Darstellung eines Systems zum Unterstützen einer Fahrzeugprüfung nach der vorliegenden Offenbarung.

The above and other objects, features, and advantages of the embodiments of the present disclosure will become more apparent through a more detailed description of the embodiments of the present disclosure in conjunction with the figures, wherein the same reference numerals generally represent the same elements in individual figures.

• 1 shows a flowchart of a method for supporting vehicle testing according to the present disclosure.
• 2 shows a flowchart of another method for supporting vehicle testing according to the present disclosure.
• 3 shows a schematic representation of a device for supporting vehicle testing according to the present disclosure.
• 4 shows a schematic representation of a system for supporting vehicle testing according to the present disclosure.

CONCRETE EMBODIMENTS

The following is based on the individual exemplary embodiments on the subject of the present The following revelation is discussed in more detail. It is to be understood that the explanation of such embodiments is merely for the better understanding of those skilled in the art and for realizing the subject matter of the present disclosure, without limiting the scope, applicability or examples in the claims. The functions and arrangement of the described elements can be changed without departing from the scope of protection of the claims. For individual exemplary embodiments, it is conceivable to omit, replace and add various processes or components as required.

The term “comprehensive” and its variants used herein may be an open term meaning “including, but not limited to”. The term “based on” can mean “based at least in part on”. The terms “one embodiment,” “some embodiments,” etc. may mean “at least one embodiment.” The terms “first”, “second”, etc. can stand for different or the same objects.

In the automobile manufacturing industry, after research and development of individual components of a vehicle is completed, it is usually necessary to test individual components to ensure that the performance of each component of the vehicle meets the requirements. In the present disclosure, the inventive basic idea of the disclosure is described using the example of an entire vehicle control as a component to be tested. It is understood that the method for supporting a vehicle test according to the present disclosure can be used not only to support a test of an overall vehicle control, but also to support a test of another component.

When testing a total vehicle controller, current and/or torque and other parameters must be transmitted as input to the overall vehicle controller in order to test the performance of the overall vehicle controller under such conditions. The overall vehicle control to be tested can receive current and/or torque and other parameters in the following two ways.

In a first method, the overall vehicle control under test is connected to an actual vehicle and the current and/or torque and other parameters are transmitted directly to the overall vehicle control under test, so that the overall vehicle control under test receives the current and/or torque and other parameters directly can be obtained from the vehicle.

Since the whole vehicle control to be tested can obtain the accurate current and/or torque and other parameters in this way, the first method can ensure the accuracy of the test of the whole vehicle control. However, since the actual vehicle must be present at the test location when testing the entire vehicle control system, the first method places relatively high demands on the test location. If the test location does not meet the requirement, no test of the entire vehicle control is possible. Therefore, the first method has insufficient flexibility when testing the overall vehicle control.

In a second method, a mathematical model is created via a computer and current and/or torque and other parameters to be entered into the vehicle control unit under test are used as the target output value of the mathematical model. The target output value may be a corresponding setpoint of the current and/or torque and the other parameters; Adjustment based on a value of a parameter entered into the mathematical model and/or a calculation formula in the mathematical model is possible to calculate a simulated output value that is as close as possible to the target output value. Thereafter, the calculated simulated output value is input to the overall vehicle control under test so that the overall vehicle control under test can perform a test based on the value of the current and/or torque and the other parameters resulting from a computer simulation.

Compared to the first method, the second method can lower the requirements for the examination location. However, in the second method, repeated adjustments to the value of the parameter entered into the mathematical model and/or the calculation formula in the mathematical model are necessary in order to obtain a simulated output value that is as close as possible to the target output value, resulting in high mathematical calculation complexity. Accordingly, the high computational complexity leads to high time costs, which reduces the efficiency of testing the overall vehicle control.

If a vehicle of the same type is tested using the second method and it is necessary to check the performance of the entire vehicle control system at different currents and/or torques and other parameters, Then, based on the mathematical model, repeated adjustments of the value of the parameter entered into the mathematical model and/or the calculation formula in the mathematical model are necessary in order to be able to obtain corresponding simulated output values that are as close as possible to the respective target output values. Due to the high mathematical calculation complexity in such an application scenario, the second method can potentially significantly affect the efficiency of testing the overall vehicle control system.

Furthermore, when testing vehicles of different types using the second method, it is necessary to test overall vehicle controls of vehicles of different types at different currents and/or torques and other parameters with regard to work performance, which further significantly increases the calculation complexity and accordingly increases the efficiency of the testing Overall vehicle control is further significantly reduced. Therefore, the second method is not suitable for the application scenario in which testing is to be carried out on vehicles of different types.

To increase flexibility and efficiency in testing a vehicle, the present disclosure proposes a method for supporting vehicle testing.

1 shows a flowchart of a method for supporting vehicle testing according to the present disclosure. The procedure according to 1 can be carried out by an auxiliary device (e.g. a computer or other similar computing device) that can communicate with the overall vehicle control under test.

In block 102, one or more parameters from the vehicle are entered into the regression algorithm. Specifically, the auxiliary device can enter one or more parameters from the vehicle into the regression algorithm. The one or more parameters are one or more parameters read from the actual vehicle, and their values may reflect the current operating state of the vehicle.

In one implementation, the one or more parameters include at least one of the following parameters: target speed of the vehicle, engine speed, state of charge (SOC) of a 48V battery, accelerator pedal ratio, power of an electric motor, operating mode of the electric motor, etc. It is understood that the listed parameters of the present disclosure are merely examples and the one or more parameters are not limited thereto. Rather, other parameters that can be used to implement the basic inventive idea of the disclosure may serve as the parameter or one or more of the parameters.

In one possible implementation, the one or more parameters include a parameter matrix that is extracted from the actual data of the vehicle. Specifically, actual data from each individual vehicle can serve as a parameter set that can be consolidated into a parameter matrix. The parameter matrix can contain the value of the parameter or values of the individual parameters at different points in time. The extraction of a parameter matrix from the actual data of the vehicle can be done using the auxiliary device or by another calculation device. Extracting a parameter matrix from the vehicle's actual data can be accomplished through machine learning or other means that enable data extraction.

In one embodiment, the regression algorithm includes the Gaussian regression algorithm. It is understood that the Gaussian regression algorithm merely represents an exemplary algorithm with which the basic inventive idea of the disclosure can be achieved, and other machine learning algorithms or other algorithms with which the basic inventive idea of the disclosure can be achieved, also in the method to support a vehicle inspection in accordance with 1 can be used.

In block 104, a predicted output value is calculated using the regression algorithm based on the one or more parameters, the predicted output value being used to support a test of the vehicle's overall vehicle control. Specifically, the auxiliary device may calculate a predicted output value using the regression algorithm based on the one or more parameters.

After the assistive device inputs the one or more parameters from the actual vehicle into the regression algorithm, the regression algorithm calculates a predicted output value based on the one or more input parameters. The calculated predicted output value can be entered into the overall vehicle control to be tested and used as an input parameter of the overall vehicle control to be tested. This means that when an examination is carried out, the person to be examined can Overall vehicle control, the calculated predicted output value is used to support the testing of the overall vehicle control to be tested.

In one implementation, calculating a predicted output value using the regression algorithm based on the one or more parameters includes: calculating a predicted output value for a first time point and one or more time points before the first time point using the regression algorithm based on the current value of the individual parameters in the parameter matrix at the first time and a historical value at one or more times before the first time.

Specifically, a description is given below using the example of entering a parameter into the Gaussian regression algorithm. In this

In this case, a simplified expression of the Gaussian regression algorithm can be expressed as follows: $$p\left(x|\mu ,\sigma \right)=\frac{1}{\sqrt{2\pi }\sigma }\text{exp}\left[-\frac{{\left(x-\mu \right)}^2}{2{\sigma }^2}\right]$$

Here x stands for a first parameter, where the values of the first parameter x can be marked as (x _t , x _t-1 , x _t-2 , ..., x _m ) at several (m) times, µ for the Average of the values (x _t , x _t-1 , x _t-2 , ..., x _m ) of the first parameter x at the several points in time, σ for the second power of the deviation and ρ(x|µ,σ) for the predicted initial value.

Below is a more detailed description using the example of entering two parameters into the Gaussian regression algorithm. In this

In this case, the expression of the kernel function of the Gaussian regression algorithm is: $$k\left(x,x\text{'}\right)={\sigma }_{ƒ}^2\text{exp}\left[\frac{-{\left(x-x\text{'}\right)}^2}{2l^2}\right]$$

Here x stands for a first parameter, x' for a second parameter, σ _f for a standard deviation function, for the range of a feature length, ϑ[σ _f ,l] for a hyperparameter for adapting the Gaussian regression algorithm and k(x,x' ) for the predicted baseline value.

oder in einer anderen Form ausgedrückt werden.Specifically, the values of the first parameter x at several points in time can be expressed as (x _t , x _t-1 , x _t-2 , ..., x _m ). The values of the second parameter x' at several points in time can be expressed as (x' _t , x' _t-1 , x' _t-2 , ...,x' _m ). x _t represents the value of the first parameter x at the first time t, x _t-1 represents the value of the first parameter x at time t-1 (ie a time before the first time t), and so on. The parameter matrix can be as $$\left[\begin{array}{ccccc}{x}_t& x_{t-1}& x_{t-2}& \cdots & x_m\\ x{\text{'}}_t& x{\text{'}}_{t-1}& x{\text{'}}_{t-2}& \cdots & x{\text{'}}_m\end{array}\right]$$

or expressed in another form.

In one embodiment, the vehicle according to the present disclosure may include an electric vehicle. In this case, the predicted output value may include at least one of the predicted torque value and current value. The target output value includes at least one of the corresponding expected values of torque value and current value.

In one embodiment, the vehicle according to the present disclosure may include a gasoline/electric hybrid vehicle. In this case, the predicted output value may include at least one of the predicted torque value and current value. The target output value includes at least one of the corresponding expected values of torque value and current value.

In one embodiment, the vehicle according to the present disclosure may include a fuel-powered vehicle. In this case, the predicted output value may include a predicted torque value and the target output value may include an expected torque value.

It is understood that the method for supporting a vehicle test according to the present disclosure is not only for supporting a test of an overall vehicle control to be tested in an electric vehicle or a gasoline/electric hybrid vehicle, but also for supporting a test of an entire vehicle control to be tested in an any type of vehicle being tested can be used. Furthermore, the torque value and the current value represent only exemplary predicted output values, and other parameter values with which the inventive basic idea can be realized can also be used as the predicted output value.

The method of supporting vehicle testing according to the present disclosure is not limited by the testing location and solely by adjusting the hyperparameter The condition of the regression algorithm can be realized, whereby the mathematical complexity is significantly reduced and thus the efficiency of the test of the overall vehicle control of the vehicle to be tested can be significantly increased. Furthermore, the method for supporting vehicle testing according to the present disclosure reduces mathematical complexity and therefore can be used to support testing on vehicles of various types and models. In this way, the efficiency of the testing of the entire vehicle control system to be tested can also be ensured for different vehicles.

2 shows a flowchart of another method for supporting vehicle testing according to the present disclosure.

In block 202, one or more parameters from the vehicle are entered into the regression algorithm. Specifically, the auxiliary device can enter one or more parameters from the vehicle into the regression algorithm. The one or more parameters are one or more actual parameters of the actual vehicle. For the specific implementation possibility of the process in block 202, reference can be made to the implementation possibility explained in block 102.

In block 204, a predicted output value is calculated using the regression algorithm based on the one or more parameters, the predicted output value being used to support a test of the vehicle's overall vehicle control. Specifically, the auxiliary device may calculate a predicted output value using the regression algorithm based on the one or more parameters. For the specific implementation possibility of the process in block 204, reference can be made to the implementation possibility explained in block 104.

In block 206, a determination is made based on an assessment of the deviation of the predicted output value from a target output value from the vehicle as to whether the regression algorithm should be adjusted based on a hyperparameter. Specifically, the auxiliary device may determine whether the regression algorithm should be adjusted based on the hyperparameter based on the assessment of the deviation of the predicted output value from the target output value from the vehicle.

The method of supporting vehicle testing according to the present disclosure aims to obtain a predicted output value that is as close as possible to the desired target output value from the actual vehicle by adjusting based on the regression algorithm. A smaller deviation of the predicted baseline value from the target baseline value indicates a higher degree of adaptation, i.e. H. a stronger approximation. In contrast, a larger deviation of the predicted baseline value from the target baseline value means a lower degree of adaptation of the two values, i.e. H. a weaker approximation.

If it is determined based on the assessment of the deviation of the predicted output value from the target output value that the approximation of the predicted output value to the target output value already meets the requirement, the predicted output value can be entered into the overall vehicle control to be tested to support the vehicle testing .

If it is determined based on the assessment of the deviation of the predicted output value from the target output value that the approximation of the predicted output value to the target output value does not yet meet the requirement, the regression algorithm should be adjusted by adjusting the hyperparameter. Then, using the customized regression algorithm, the next calculation is performed until the approximation of the predicted output value to the target output value meets the requirement. Then the predicted output value that meets the requirement can be input into the overall vehicle control under test to support the vehicle testing.

By assessing the predicted output value and adjusting the regression algorithm as necessary, the accuracy of the predicted output value delivered to the overall vehicle control under test can be ensured, which helps ensure the accuracy of the result of the test of the overall vehicle control under test.

In one implementation possibility, the approximation of the predicted output value from the target output value can be manifested by the root mean square error (RMSE), as given by the following formula: $$RMSE=\sqrt{\frac{1}{m}{\displaystyle {\sum }_{i=1}^m{\left(X_{i,pred}-X_{i,mean}\right)}^2}}$$

X _i.pred stands for the predicted target value of a parameter from _the vehicle at the i-th time of the m times, the target output value of the parameter from the vehicle at the i-th time and RMSE for assessing the deviation of the predicted output value from the target output value. RMSE can manifest the stability of the learning result of the regression algorithm with respect to one or more parameters from the vehicle.

A lower value of RMSE indicates a smaller deviation of the predicted baseline from the target baseline. In other words, a higher degree of adaptation of the two values, i.e. H. assumed a stronger approximation. Accordingly, this means that the predicted output value can be entered into the overall vehicle control to be tested to support the vehicle testing.

A larger value of RMSE indicates a larger deviation of the predicted baseline from the target baseline. In other words, there is a lower degree of adaptation of the two values, i.e. H. assumed a weaker approximation. Accordingly, the regression algorithm should be adjusted by adjusting the hyperparameter. Then, using the customized regression algorithm, the next calculation is performed until the approximation of the predicted output value to the target output value meets the requirement. Then the predicted output value that meets the requirement can be input into the overall vehicle control under test to support the vehicle testing.

$$SST={{\displaystyle {\sum }_{i=1}^m\left(X_{i,pred}-{\overline{X}}_{i,mean}\right)}}^2$$

$$R^2=1-\frac{SSR}{SST}$$

In one implementation possibility, the approximation of the predicted output value from the target output value can be manifested by a determination coefficient (R ² ), as given by the following formulas: $$SSR={{\displaystyle {\sum }_{i=1}^m\left(X_{i,pred}-X_{i,mean}\right)}}^2$$

$$SST={{\displaystyle {\sum }_{i=1}^m\left(X_{i,pred}-{\overline{X}}_{i,mean}\right)}}^2$$

$$R^2=1-\frac{SSR}{SST}$$

X _i,pred stands for the predicted target value of a parameter from the vehicle at the i-th time of the m times, X _i.mean for the target initial value of the parameter from the vehicle at the i-th time, SSR for the regression sum the squares of the predicted output value and the target output value, SST for the sum of squares of the deviations of the predicted output value from the target output value and R ² for assessing the deviation of the predicted output value from the target output value.

R ² may manifest a relative measurement of appraisal model error.

For example, if 0.9 < R ² <1, it is considered that a smaller deviation of the predicted output value from the target output value indicates a higher degree of adaptation, ie a stronger approximation. Accordingly, the predicted output value can be entered into the overall vehicle control to be tested to support the vehicle testing. If 0.6 < R ² <0.8, it is considered that despite a certain deviation of the predicted output value from the target output value, the predicted output value is still within the range of possible application in supporting vehicle testing. Therefore, the predicted output value can be input into the overall vehicle control under test to support the vehicle testing. For example, if 0 < R ² <0.5, it is assumed that a higher deviation of the predicted output value from the target output value indicates a lower degree of adaptation, ie a weaker approximation. Now the regression algorithm should be adjusted by adjusting the hyperparameter. Then, using the customized regression algorithm, the next calculation is performed until the approximation of the predicted output value to the target output value meets the requirement. Then the predicted output value that meets the requirement can be input into the overall vehicle control under test to support the vehicle testing.

It is understood that the evaluation criteria of the above value ranges of R ² are only exemplary and the evaluation criteria of the values of R ² are not limited thereto. Rather, a suitable value range can be selected as required.

In the above implementation options, the test of the overall vehicle control of the vehicle can include a test of the software logic of the overall vehicle control of the vehicle. By checking the software logic of the whole vehicle control and optimizing the software logic of the whole vehicle control according to the test result, the work performance of the whole vehicle control of the vehicle can be improved.

3 shows a schematic representation of a device for supporting vehicle testing according to the present disclosure. The before Device 300 for supporting vehicle testing includes at least one processor 302 and a memory 304 that communicates with the at least one processor 302. An executable code is stored in the memory 304, which, when executed by the at least one processor 302, causes the at least one processor 302 to carry out one of the methods 1 and 2 executes.

An embodiment of the present disclosure further provides a computer-readable storage medium. The computer-readable storage medium may store executable code that, when executed by a computer, causes the computer to perform one of the methods 1 and 2 executes.

4 shows a schematic representation of a system for supporting vehicle testing according to the present disclosure. The system 400 for supporting vehicle testing includes a total vehicle control under test 404 and a device 402 for supporting a vehicle testing, wherein the predicted output value of the device 402 for supporting vehicle testing can be used to support a test of the total vehicle control under test 404. The device 402 for supporting vehicle testing may be a device for performing one of the methods according to 1 and 2 or a device 300 for supporting a vehicle inspection 3 include.

For example, the computer-readable storage medium may include a random access memory (RAM), a read-only memory (ROM), an electrically-erasable programmable read-only memory (ROM), EEPROM), a static random access memory (SRAM), a hard drive, a flash memory, and so on.

Specific embodiments of the present disclosure have been described so far. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims may be performed in a different order than that in the embodiments and may still achieve the desired result. Furthermore, the process illustrated in the figures does not necessarily require a particular sequence shown or a continuous sequence to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Not all steps and units in the above processes and system structure diagrams are necessary, and certain steps or units can be omitted according to actual needs. The device structure described in the above embodiments may be a physical structure or a logical structure, that is, some units may be implemented by one and the same physical object, and some units may be implemented by multiple physical objects, or they may be implemented together by some components in several independent devices.

The term “exemplary” used throughout the specification means “serving as an example” and does not mean “preferred” or “advantageous” over other embodiments. To provide an understanding of the technologies described, the specific embodiments include specific details. However, these technologies can be implemented without these specific details. In some cases, well-known structures and devices are shown in block diagram form to avoid making the concept of the described embodiments difficult to understand.

So far, optional embodiments of the present disclosure have been explained in more detail with reference to the accompanying figures. However, the exemplary embodiments of the present disclosure are in no way limited to the specific details of the above embodiments, and rather, within the framework of the basic technical ideas of the exemplary embodiments of the disclosure, various variants for the technical solution of the exemplary embodiments of the disclosure are possible, these variants also being part of the scope of protection of the exemplary embodiments belong to revelation.

The foregoing description of the present disclosure will enable one of ordinary skill in the art to make or apply the present disclosure. The various modifications of the present disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the scope of the present disclosure. Therefore, the present disclosure is not limited to the examples and embodiments described herein, but is consistent with the broad scope test the scope of the principles and novel features disclosed herein.

Claims (12)

Method for supporting a vehicle inspection, comprising: Entering one or more parameters from the vehicle into a regression algorithm; Calculating a predicted output value using the regression algorithm based on the one or more parameters, the predicted output value serving to support a test of the overall vehicle control of the vehicle. Procedure according to Claim 1 , wherein the one or more parameters include at least one of the following parameters: target speed of the vehicle; engine speed; Charge level of a 48V battery; accelerator pedal ratio; power of an electric motor; and operating mode of the electric motor. Procedure according to Claim 1 , further comprising: determining, based on an assessment of the deviation of the predicted output value from a target output value from the vehicle, whether the regression algorithm should be adjusted based on a hyperparameter. Procedure according to Claim 3 , wherein the vehicle comprises an electric vehicle, wherein the predicted output value comprises at least one of the predicted values of torque value and current value, and wherein the target output value comprises at least one of the corresponding expected values of torque value and current value. Procedure according to Claim 3 , wherein the vehicle includes a gasoline/electric hybrid vehicle, wherein the predicted output value includes at least one of the predicted values of torque value and current value, and wherein the target output value includes at least one of the corresponding expected values of torque value and current value. Procedure according to Claim 3 , wherein the vehicle includes a fuel-powered vehicle, wherein the predicted output value includes a predicted torque value and the target output value includes an expected torque value. Procedure according to Claim 1 , wherein the one or more parameters comprise a parameter matrix that is extracted from the actual data of the vehicle. Procedure according to Claim 7 , wherein calculating a predicted output value using the regression algorithm based on the one or more parameters comprises: calculating a predicted output value for a first time point and one or more time points before the first time point using the regression algorithm based on the current values of the individual parameters in the parameter matrix at the first point in time and a historical value at one or more points in time before the first point in time. Procedure according to one of the Claims 1 until 4 , wherein the regression algorithm comprises the Gaussian regression algorithm, wherein the testing of the overall vehicle control of the vehicle comprises: testing the software logic of the overall vehicle control of the vehicle. An apparatus for supporting vehicle testing, comprising: at least one processor; a memory that communicates with the at least one processor and in which an executable code is stored, the executable code, when executed by the at least one processor, causing the at least one processor to execute a method according to one of the Claims 1 until 9 executes. A computer-readable storage medium that stores executable code that, when executed, causes a computer to carry out the method according to one of the Claims 1 until 9 executes. Vehicle testing system, comprising: an overall vehicle control to be tested; and a device for carrying out the method according to one of the Claims 1 until 9 or a device to support a vehicle inspection Claim 10 , where the predicted output value is used to support a test of the overall vehicle control.

Images