Procedures for checking and, if necessary, improving a vehicle diagnostic system

Abstract

Method for checking and, if necessary, improving the setting of at least one parameter of a control unit monitoring function of a vehicle diagnostic system, taking into account at least one robustness indicator, in particular taking into account four robustness indicators, - where at least one robustness parameter R z [n] is calculated according to the following equation: R z [ n ] = 1 r max_pos / max_neg ⋅ N 0 ∑ i = n n+N 0 ri - where r max_pos / max_neg the largest positive or largest negative distance value when monitoring against a positive or a negative deviation is, - where N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, - where n is an element of a data series of consecutive feature values, - where r ithe distance of a characteristic value from a positive or negative error threshold, determined on the basis of the raw distance, - where the setting of at least one parameter is retained if the calculated value R z or the calculated values ​​R z above a minimum robustness index R z,min lies / lie - and where the setting of at least one parameter is adjusted when the calculated value R z or the calculated values ​​R z below the minimum robustness index R z,min lies / lie or the calculated value R z or the calculated values ​​R z the minimum robustness index R z,min corresponds / correspond to, - wherein, in order to check the setting of at least one parameter, the results of the control unit monitoring function are calculated by inputting an input data set of a functioning system and an input data set of a non-functioning system into the control unit monitoring function, characterized in that - that, to check the setting of at least one parameter, the results determined by the control unit monitoring function are compared with certain values ​​of a characteristic of a functioning system, and, to monitor against a positive error threshold exceedance, a value R is calculated. z,max_threshold,False Fail to check whether a functioning system is incorrectly identified as non-functional, - that, in order to verify the setting of at least one parameter, the results obtained by the control unit monitoring function are compared with certain values ​​of a characteristic of a functioning system for monitoring against a negative error threshold undercutting by calculating a value R z,min_threshold,False Fail to check whether a functioning system is incorrectly identified as non-functional, - that, in order to verify the setting of at least one parameter, the results determined by the control unit monitoring function are compared with certain values ​​of a feature of a non-functional system for monitoring against a positive error threshold undercutting by calculating a value R z,max_threshold,False Pass to check whether a non-functional system is incorrectly identified as functional, - and that, to verify the setting of at least one parameter, the results obtained by the control unit monitoring function are compared with certain values ​​of a feature of a non-functional system for monitoring against a negative error threshold exceedance by calculating a value R z,min_threshold,False Pass It should be checked whether a non-functional system is incorrectly identified as functional.

Description

[0001] The invention relates to a method according to the features of the independent claim. In particular, the invention relates to a method for checking and optionally improving a vehicle diagnostic system. Preferably, this method checks, on the one hand, whether a functioning system is incorrectly identified as non-functional and, on the other hand, whether a non-functional system is incorrectly identified as functional. Furthermore, the invention relates to a method for improved control of a fault indicator of a vehicle diagnostic system, wherein the checking and / or improvement of the vehicle diagnostic system reduces false diagnoses of the vehicle diagnostic system.

[0002] The field of the invention relates in particular to methods as described in DE 10 2014 115 485 A1 of the applicant.

[0003] The software of a vehicle diagnostic system includes not only functions for controlling and regulating internal engine processes, but also numerous functions for monitoring various systems and components. A key objective of the vehicle diagnostic system is to detect malfunctions that lead to abnormal engine behavior as early as possible and trace them back to the smallest replaceable unit. Since the number of sensors is kept to a minimum for cost reasons, model-based fault detection and diagnostic methods are used to achieve this goal. These methods utilize mathematical models to exploit the dependencies of various measurable signals within a process to gain information about the process state.The overall process of model-based on-board diagnostics (OBD) comprises a certain number of actuators, the actual physical process, and various sensors. Based on input and output signals, characteristic quantities or features are calculated in the model, providing information about the process state.

[0004] A commonly used method for fault detection and diagnostics in the automotive sector is fault detection using parity equations. In this method, models with a pre-defined structure and parameters represent the process behavior. These models are arranged in parallel with the modeled processes. A measured signal from a process is compared to a modeled reference value by means of subtraction or quotient calculation. This characteristic is referred to as the residual e. RThe residual is defined as follows: When the comparison is made by subtraction, it becomes zero; when the comparison is made by quotient, it becomes one if and only if the reference model perfectly represents the process. Conversely, assuming an exact model, the residual can be used to infer deviations in the measured signal and thus an error. The generated characteristic, and in particular the residual, is therefore monitored for deviations from the expected value. This allows changes in the process to be detected and disturbances to be distinguished from the normal state.

[0005] Robustness refers to insensitivity to small deviations from assumptions, particularly from expected values. Applied to OBD, this means the ability to tolerate random fluctuations in monitored characteristic values ​​and to distinguish them from significant, fault-relevant deviations. Random deviations result from the stochastic variation of input signals to the functions of fault detection and diagnostic procedures and can be caused by environmental conditions, driving behavior, aging, tolerances in production processes, defects, and / or measurement inaccuracies.

[0006] In addition to stochastically varying input signals, inaccuracies in the mathematical models used in fault detection and diagnostic procedures also lead to variations in the monitored characteristic values. These uncontrollable variations necessitate the verification and, if necessary, improvement of the parameters of an ECU monitoring function to reduce the risk of misdiagnoses by the vehicle diagnostic system. A misdiagnosis is, on the one hand, the vehicle diagnostic system's assumption that a functioning system is not functioning—a so-called "false fail" error—and, on the other hand, the vehicle diagnostic system's assumption that a non-functional system is functioning—a so-called "false pass" error.

[0007] Conventional fault detection and diagnostic methods use debouncing techniques to increase robustness. This makes it possible to distinguish actual malfunctions of a component or system from false diagnoses. A false diagnosis can occur, for example, if a characteristic value exceeds a threshold only briefly. A commonly used debouncing method defines that an actual malfunction only exists if the characteristic value continuously exceeds a fault threshold for a certain period of time, the debouncing time. Conventional methods for assessing the robustness of fault detection and diagnostic procedures are based on analyzing the maximum duration of the continuous exceedance of a defined fault threshold within a measurement.Conventional methods analyze the absolute frequencies of time periods with error debouncing, i.e., time periods in which the values ​​of a monitored feature were consistently above a defined error threshold. However, these conventional methods fail to provide important information about robustness, particularly information about those parts of the empirical feature values ​​where no error threshold violations occur.

[0008] Furthermore, prior art reveals known error detection and diagnostic methods that utilize debouncing techniques, where a counter reading must reach a defined value. In these conventional methods, the counter is incremented when the error threshold is exceeded and decremented when it falls below the threshold. For robustness analysis, the distribution of the highest and lowest recorded counter readings of the debouncing counter is considered.

[0009] Another method frequently used in engine control unit software to increase the robustness of fault detection and diagnostic procedures and their functions is to increment a counter whenever a feature value exceeds a fault threshold. When the feature value returns to normal, i.e., falls below the fault threshold, the counter is decremented by a certain amount. This allows for faster detection of faults that occur in short time intervals. Conventional fault detection and diagnostic procedures rely on analyzing the maximum debounce counter value within a measurement to assess the robustness of their functions. However, this method fails to provide important information about robustness, particularly information about those parts of the empirical measurements where the fault threshold is not exceeded.

[0010] Among other things, US Patent 7,743,351 B2 discloses a method for verifying the robustness of a model of a physical system. This method involves defining a first model of the physical system with a first set of components and at least one input interface. The first model defines a formal language that describes the behavior and function of each component. This formal language describes the properties that the model of the physical system must fulfill. A second model, corresponding to the first model and additionally featuring an error-entry mechanism, is also described in the formal language. Using formal proofs, the method automatically searches for a combination of entered errors and / or input values ​​that causes the specified property to fail.However, this approach is quite complex and offers no way to objectively compare the robustness of different models.

[0011] Further methods for verifying and, if necessary, improving a vehicle diagnostic system are known from DE 10 2013 202 266 A1 and US 2009 / 0 299 713 A1.

[0012] The object of the invention is to overcome the disadvantages of the prior art. In particular, the object of the invention is to provide a method for checking and, if necessary, improving the setting of at least one parameter of a control unit monitoring function of a vehicle diagnostic system, thereby reducing the probability of incorrect diagnoses by the vehicle diagnostic system. This includes, in particular, reducing both the probability of the vehicle diagnostic system incorrectly assessing a functioning system as non-functional (so-called "false fail" errors) and the probability of the vehicle diagnostic system assessing a non-functional system as functional (so-called "false pass" errors).Furthermore, it is an object of the invention to provide a method for controlling a fault indicator of a vehicle diagnostic system, whereby incorrect diagnoses of a vehicle diagnostic system and thus the incorrect control of the fault indicator, in particular the illumination of a MIL lamp (Malfunction Indicator Light - engine control light) of a vehicle, are reduced.

[0013] The problems according to the invention are solved in particular by the features of the independent patent claims.

[0014] The invention relates to a method for checking and optionally improving the setting of at least one parameter of a control unit monitoring function of a vehicle diagnostic system, taking into account at least one robustness indicator, in particular taking into account four robustness indicators, wherein the at least one robustness indicator R z [n] is calculated according to the following equation: Rz[n]=1rmax_pos / max_neg⋅N0∑i=nn+N0ri where r max_pos / max_neg the largest positive or largest negative distance value when monitoring against a positive or a negative deviation is, where N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, where n is an element of a data series of consecutive feature values, where r i The distance of a feature value from a positive or negative error threshold, determined on the basis of the raw distance, whereby the setting of at least one parameter is maintained when the calculated value R z or the calculated values ​​R z above a minimum robustness index R z,min lies / lie and wherein the setting of at least one parameter is adjusted when the calculated value R z or the calculated values ​​R zbelow the minimum robustness index R z,min lies / lie or the calculated value R z or the calculated values ​​R z the minimum robustness index R z,min corresponds / correspond to, wherein, in order to check the setting of at least one parameter, the results of the control unit monitoring function are calculated by inputting an input data set of a functioning system and an input data set of a non-functioning system into the control unit monitoring function.

[0015] According to the invention, it is provided that, to check the setting of at least one parameter, the determined results of the control unit monitoring function are compared with certain values ​​of a feature of a functioning system, and, to monitor against a positive error threshold exceedance, a value R is calculated. z,max_threshold,False Failto check whether a functioning system is incorrectly identified as non-functional, that, in order to verify the setting of at least one parameter, the results determined by the control unit monitoring function are compared with certain values ​​of a characteristic of a functioning system for monitoring against a negative error threshold undercutting by calculating a value R z,min_threshold,False Fail to check whether a functioning system is incorrectly identified as non-functional, that, in order to verify the setting of at least one parameter, the results determined by the control unit monitoring function are compared with certain values ​​of a characteristic of a non-functional system for monitoring against a positive error threshold undercutting by calculating a value R z,max_threshold,False Passto check whether a non-functional system is incorrectly identified as functional, and that, in order to verify the setting of at least one parameter, the results determined by the control unit monitoring function are compared with certain values ​​of a feature of a non-functional system for monitoring against a negative error threshold exceedance by calculating a value R z,min-threshold,False Pass It should be checked whether a non-functional system is incorrectly identified as functional.

[0016] Optionally, the method according to the invention is characterized in that R is calculated using z,max_threshold,False Fail The following definition is made: Rz,max_threshold,False Fail[n]=1N0∑nn+N0|e[i]−gQpos[i]eQ0[i]−gQpos[i]| e[i]={eQ[i]…eQ0[i]≤eQ[i]≤gQpos[i]eQ0[i]…eQ[i]<eQ0[i]gQpose[i]…eQ[i]> gQpos[i] where N0 is the number of data points in the data series that represent the minimum time, in particular the debounce time, which is used for error detection, where n is an element of a data series of consecutive feature values, where e Q [i] the feature value is, where g Qpos [i] the positive threshold is, where e Q0 [i] the normal value is, and where e[i] is the limited measurement value.

[0017] Optionally, the method according to the invention is characterized in that R is calculated using z,min_threshold,False Fail The following definition is made: Rz,min_threshold,False Fail[n]=1N0∑nn+N0|e[i]−gQneg[i]eQ0[i]−gQneg[i]| e[i]={eQ[i]…gQneg[i]≤eQ[i]≤eQ0[i]eQ0[i]…eQ[i]<eQ0[i]gQnege[i]…eQ[i]> gQneg[i] where N0 is the number of data points in the data series that represent the minimum time, in particular the debounce time, which is used for error detection, where n is an element of a data series of consecutive feature values, where e Q [i] the feature value is, where g Qneg [i] the negative threshold is, where e Q0 [i] the normal value is, and where e[i] is the limited measurement value.

[0018] Optionally, the method according to the invention is characterized in that R is calculated using z,max_threshold,False Pass The following definition is made: Rz,max_threshold,False Fail[n]=1N0∑nn+N0|e[i]−gQpos[i]eQ0[i]−gQpos[i]| e[i]={eQ[i]…gQpos[i]≤eQ[i]≤eQ0[i]eQ0[i]…eQ[i]<eQ0[i]gQpose[i]…eQ[i]> gQpos[i] eQ0[i]={gQpos[i]×(1+m100)…gQpos[i]>0gQpos[i]×(1−m100)…gQpos<0 where N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, where n is an element of a data series of consecutive feature values, where eQ [i] the feature value is, where g Qpos [i] the positive threshold is, where e Q0 [i] the normal value is, and where e[i] is the limited measurement value.

[0019] Optionally, the method according to the invention is characterized in that R is calculated using z,min_threshold,False Pass The following definition is made: Rz,min_threshold,False Fail[n]=1N0∑nn+N0|e[i]−gQneg[i]eQ0[i]−gQneg[i]| e[i]={eQ[i]…eQ0[i]≤eQ[i]≤gQneg[i]eQ0[i]…eQ[i]<eQ0[i]gQnege[i]…eQ[i]> gQneg[i] where N0 is the number of data points in the data series that represent the minimum time, in particular the debounce time, which is used for error detection, where n is an element of a data series of consecutive feature values, where e Q [i] the feature value is, where g Qneg [i] the negative threshold is, where e Q0[i] the normal value is, and where e[i] is the limited measurement value.

[0020] Optionally, the method according to the invention is characterized in that the setting of the at least one parameter is maintained when the calculated value R z,max_threshold,False Fail and the calculated value R z,min_threshold,False Fail above a minimum robustness index R z,min,False Fail lie, or that the setting of at least one parameter is retained when the calculated values ​​R z,max_threshold,False Fail and the calculated values ​​R z,min_threshold,False Fail above a minimum robustness index R z,min,False Fail lie, and that the setting of at least one parameter is adjusted when the calculated value R z,max_threshold,False Fail and the calculated value R z,min_threshold,False Fail below the minimum robustness index R z,min,False Fail lie or the minimum robustness index R z,min,False Fail correspond, or that the setting of at least one parameter is adjusted when the calculated values ​​R z,max_threshold,False Fail and the calculated values ​​R z,min_threshold,False Fail below the minimum robustness index R z,min,False Fail lie or the minimum robustness index R z,min,False Fail are equivalent to.

[0021] The method according to the invention may optionally be characterized by: that the setting of at least one parameter is retained when the calculated value R z,max_threshold,False Pass and the calculated value R z,min_threshold,False Pass above a minimum robustness index R z,min,False Pass lie, or that the setting of at least one parameter is retained when the calculated values ​​R z,max_threshold,False Pass and the calculated values ​​R z,min_threshold,False Pass above a minimum robustness index R z,min,False Pass lie, and that the setting of at least one parameter is adjusted when the calculated value R z,max_threshold,False Passand the calculated value R z,min_threshold,False Pass below the minimum robustness index R z,min,False Pass lie or the minimum robustness index R z,min,False Pass correspond, or that the setting of at least one parameter is adjusted when the calculated values ​​R z,max_threshold,False Pass and the calculated values ​​R z,min_threshold,False Pass below the minimum robustness index R z,min,False Pass lie or the minimum robustness index R z,min,False Pass are equivalent to.

[0022] Optionally, the method according to the invention is characterized in that R z,min,False Fail and / or R z,min,False Pass in a range of 0.5 to 0.9, particularly in a range of 0.6 to 0.8.

[0023] Optionally, the method according to the invention is characterized in that R z,min,False Fail and / or R z,min,False Pass 0.7 is.

[0024] Optionally, the method according to the invention is characterized in that the determined values ​​of the feature are values ​​measured on a system, or that the determined values ​​of the feature are empirically or arbitrarily determined values.

[0025] Optionally, the method according to the invention is characterized in that the vehicle diagnostic system is part of a vehicle. and / or that the vehicle diagnostic system is an on-board diagnostics (OBD) system, and / or that the vehicle diagnostic system is part of an ECU (Engine Control Unit) or an ECU (Engine Control Unit) of a motor vehicle.

[0026] Optionally, the method according to the invention is characterized in that the control unit monitoring function and the checked and optionally improved parameters are included in the vehicle diagnostic system, or that the control unit monitoring function and the checked and optionally improved parameters are transferred to a vehicle diagnostic system.

[0027] Furthermore, the invention relates to a method for controlling a fault indicator of a vehicle diagnostic system, in particular for controlling a MIL lamp “Malfunction Indicator Light - engine control light” of a vehicle, wherein the fault indicator signals a fault when the vehicle diagnostic system diagnoses a fault.

[0028] According to the invention, the method for controlling a fault indicator of a vehicle diagnostic system is characterized in that the setting of the at least one parameter of the control unit monitoring function of the vehicle diagnostic system is carried out as described above.

[0029] If applicable, the method for controlling a fault indicator of a vehicle diagnostic system is characterized by the fact that the vehicle diagnostic system diagnoses a fault: if a deviation between at least one specific value of a characteristic and at least one calculated value of a characteristic is greater than at least one positive error threshold, and / or if a deviation between at least one specific value of a characteristic and at least one calculated value of a characteristic is less than at least one negative error threshold, and / or if at least one release condition is met when this deviation occurs, and / or if a debounce time is exceeded when this deviation occurs.

[0030] The method for controlling a fault indicator of a vehicle diagnostic system may be characterized by: that at least one specific value of a feature is measured, determined, or transmitted to the vehicle diagnostic system, and that at least one calculated value of a characteristic from Vehicle diagnostic system, in particular by the control unit monitoring function, determines or transmits to the vehicle diagnostic system.

[0031] In particular, critical areas can be identified by developing robustness metrics derived from empirical measurement data. The robustness of the ECU monitoring function of a vehicle diagnostic system can be assessed based on the robustness metric. A robustness metric with a value of one means that the tested ECU monitoring function is robust, and therefore incorrect diagnoses by the vehicle diagnostic system are unlikely. Conversely, a robustness metric of zero means that the tested ECU monitoring function is not robust, and therefore incorrect diagnoses by the vehicle diagnostic system are likely. This means that the probability of a "false fail" error (i.e., the assessment of a functioning system as non-functional) and a "false pass" error (i.e., the assessment of a non-functional system as functional) is reduced.

[0032] The robustness index allows for statements not only about the occurrence of error threshold exceedances, but also about those parts of the empirical measured values ​​where no error threshold is exceeded or where the diagnostic release time was shorter than the minimum time required for error detection. The robustness index thus also permits a quantitative evaluation of robustness. Depending on the robustness index(s), parameters of control unit monitoring functions, in particular engine or process parameters, or calibration characteristics of the internal combustion engine, can be modified so that, upon re-execution of the inventive method, the robustness index(s) lies within a desired, predefined range. The robustness index(s) can therefore be used as a control variable for internal combustion engines.

[0033] A common method used in vehicle diagnostic systems to increase the robustness of an ECU monitoring function is to increment a counter whenever the characteristic value exceeds an error threshold. When the characteristic value returns to normal, i.e., falls below the error threshold, the counter is reset.

[0034] In the methods according to the invention, at least one robustness parameter R is used. z calculated for a data series n. The at least one robustness parameter R z This is calculated on the basis of a sum of all distances occurring in a series.

[0035] A distinction is made between monitoring against positive error threshold exceedances and monitoring against negative error threshold exceedances. Monitoring against positive error threshold exceedances refers to monitoring for exceedances of the positive error threshold, while monitoring against negative error threshold exceedances refers to monitoring for exceedances of the negative error threshold.

[0036] A positive maximum distance value can be defined as the difference between a positive error threshold and a defined normal value. The distance can be set equal to the positive maximum distance value if—when the characteristic value exceeds the positive error threshold—the raw distance is greater than the positive maximum distance value. The distance can be set equal to zero if—when the characteristic value exceeds the positive error threshold—the raw distance is less than zero.

[0037] Similarly, a negative maximum distance value is defined as the difference between a defined normal value and a negative error threshold. The distance can be set equal to the negative maximum distance value if—when the characteristic value falls below the negative error threshold—the raw distance is greater than the negative maximum distance value. The distance can be set equal to zero if—when the characteristic value falls below the negative error threshold—the raw distance is greater than zero.

[0038] The respective robustness index R z It is calculated as follows: From the discretely sampled feature value e Q For a monitored feature at position n, the difference to the corresponding error threshold g is calculated. Qpos or g Qneg calculated, which is referred to here as the raw distance r0. This raw distance r0 is advantageously assigned a value of 0 for downwards and r for upwards. max_posor r max_neg limited, which limits the values ​​r i The distance r results. max_pos or r max_neg from a defined error threshold g Qpos or g Qneg is calculated from the difference between the error threshold g Qpos or g Qneg and the normal value e Q0 , which ideally occurs. For diagnostic functions with variable error thresholds, g Qpos or g Qneg dependent on n.

[0039] The robustness index R z is now the sum of the values ​​r i between the currently considered measurement series position n and the end of the debounce time n + N0, with the debounce time N0 and is standardized with the maximum area, which results from the product of maximum distance r max and the debounce time N0. The value r i This corresponds to the maximum distance r max and 0 limited difference between the current error value eQ and the threshold g Q .

[0040] Moreover, the robustness index R z only defined if at least one release condition of the diagnostic function is met for the entire range between n and n + N0.

[0041] Alternatively, and advantageously, the distance of the feature value from the threshold value, determined on the basis of the raw distance, is set equal to the maximum distance value if the release conditions for a diagnosis are not met. Thus, R z It is also defined in areas outside the activation conditions and has a value of 1 in these areas. There is a gradual transition into areas where the diagnostic function is enabled.

[0042] Preferably, the method for checking and, if necessary, improving the setting of at least one parameter of a control unit monitoring function of a vehicle diagnostic system takes into account at least one robustness indicator R. z [n], in particular four robustness parameters R z [n], includes the following steps: - Calculation of the results of the control unit monitoring function by inputting an input data set from a functioning system and an input data set from a non-functioning system, - Verification of the setting of at least one parameter by comparing the determined results of the control unit monitoring function with certain values ​​of a characteristic of a functioning system, whereby a value R is calculated z,max_threshold,False Fail monitoring against exceeding a positive error threshold and by calculating a value R z,min threshold,False FailMonitoring is performed against exceeding a negative error threshold, and thus the probability of a "false fail" error can be quantified. - Verification of the setting of at least one parameter by comparing the determined results of the control unit monitoring function with certain values ​​of a characteristic of a non-functional system, whereby a value R is calculated z,max_threshold,False Pass monitoring against exceeding a positive error threshold and by calculating a value R z,min threshold,False Pass Monitoring is performed against exceeding a negative error threshold, and thus the probability of a "false pass" error can be quantified.

[0043] It may be stipulated that a functioning system consists of functioning components or is a functioning component. It may also be stipulated that a functioning system comprises functioning components or a functioning component.

[0044] It may be stipulated that a non-functional system consists of non-functional components or is a non-functional component. It may also be stipulated that a non-functional system includes non-functional components or a single non-functional component.

[0045] In all embodiments, it can be provided that a vehicle diagnostic system is improved by adjusting the setting of at least one parameter of the control unit monitoring function.

[0046] In all embodiments, it can be provided that the error threshold corresponds to the threshold value.

[0047] The invention is explained in more detail below with reference to a non-limiting embodiment illustrated in the figures. These show: Fig. 1 A schematic representation of a procedure for checking and, if necessary, improving the setting of at least one parameter of a control unit monitoring function of a vehicle diagnostic system, taking into account four robustness indicators R z [n]; Fig. 2a and Fig. 3a Excerpts of evaluation diagrams before the improvement of the setting of a parameter of a control unit monitoring function of a vehicle diagnostic system using the method; Fig. 2b and Fig. 3b Excerpts of evaluation diagrams after the improvement of the setting of a parameter of a control unit monitoring function of a vehicle diagnostic system using the method.

[0048] Fig. Figure 1 shows a method for checking and, if necessary, improving a vehicle diagnostic system. This method can reduce misdiagnoses by the vehicle diagnostic system. Thus, it reduces the false assumption of a fault when it does not exist (so-called "false fail" errors) as well as the failure to detect an existing fault (so-called "false pass" errors).

[0049] The preferred first step of the procedure is to identify a functioning system ("OK System") and a non-functional system ("NOK System"). By identifying the functioning and non-functional systems, the input data sets required to determine the results of the ECU monitoring functions are defined.

[0050] The next step in the procedure involves checking or, if necessary, improving the setting of at least one parameter of an ECU monitoring function. If the procedure has already been performed once and the determined robustness indicators were not above the minimum robustness indicators, at least one parameter of an ECU monitoring function is adjusted or improved in this step.

[0051] Subsequently, to check and, if necessary, improve a vehicle diagnostic system, the setting of at least one parameter of a control unit monitoring function is carried out, taking into account four different robustness indicators R. z [n] checked.

[0052] The procedure involves, on the one hand, calculating a value R. z,max_threshold,False Fail monitoring against a positive error threshold and by calculating R z,min threshold,False Fail Monitoring is performed against a negative error threshold to check whether a functioning system is incorrectly identified as non-functional - a so-called "false fail" error.

[0053] On the other hand, by calculating a value R z,max_threshold,False Pass monitoring against a positive error threshold and by calculating a value R z,min_threshold,False PassA monitoring process is performed against a negative error threshold to check whether a non-functional system is incorrectly identified as functional - a so-called "false pass" error.

[0054] The next step in the procedure for calculating the robustness coefficients, R z,max_threshold,False Fail and R z,min_threshold,False Fail The result of the ECU monitoring function is calculated using the input data set of a functioning system or component. Subsequently, the robustness metrics are calculated by comparing the results obtained from the ECU monitoring function with specific values ​​of a characteristic of the functioning system or component.

[0055] The next step in the procedure for calculating the robustness coefficients, R z,max_threshold,False Passport and R z,min_threshold,False PassThe result of the ECU monitoring function is calculated using the input data set of a functioning system or component. Subsequently, the robustness metrics are calculated by comparing the results obtained from the ECU monitoring function with specific values ​​of a characteristic of the non-functional system or component.

[0056] In this embodiment of the method, the calculation of the robustness indicators takes into account the debounce time and at least one release condition.

[0057] The setting of at least one parameter of the control unit monitoring function of a vehicle diagnostic system is retained if the calculated values ​​for the robustness indicators R z,max_threshold,False Fail and R z,min_threshold,False Fail above a minimum robustness index R z,min ,False Failand the calculated values ​​for the robustness indicators R z,max_threshold,False Pass and R z,min_threshold,False Pass above a minimum robustness index R z,min ,False Pass lay.

[0058] The setting of at least one parameter of a control unit monitoring function of a vehicle diagnostic system is adjusted when the calculated values ​​for the robustness indicators R z,max_threshold,False Fail and R z,min_threshold,False Fail below the minimum robustness index R z,min,False Fail and the calculated values ​​for the robustness indicators R z,max_threshold,False Pass and R z,min_threshold,False Pass below the minimum robustness index R z,min ,False Pass lie or the calculated values ​​correspond to the respective minimum robustness indicators.

[0059] If the setting of at least one parameter of the control unit monitoring function is retained, the review and, if necessary, improvement of the setting of at least one parameter of the control unit monitoring function is completed.

[0060] If the setting of at least one parameter of the control unit monitoring function is adjusted, the procedure is repeated after the improvement.

[0061] In the Fig. 2a, Fig. 2b, Fig. 3a and Fig. 3b shows the monitored signal and the threshold value in hectopascals on the left y-axis, and the robustness index R on the right y-axis. z and time t in seconds is plotted on the x-axis.

[0062] Fig. Figure 2 shows the improvement achieved through the method, taking into account the robustness index R. z,max_threshold,False Fail against a negative error threshold exceedance g Qneg The deviation e is used for this. Q to normal value e Q0 , before the improvement through the process - see Fig. 2a - and after the improvement through the procedure - see Fig. 2b - shown. Before improving the setting of at least one parameter of a control unit monitoring function of a vehicle diagnostic system, the deviation of the monitored signal e must be determined. Q , which is the deviation from a normal value e Q0 is relatively large. As a result, the robustness index R is z,max_threshold,False Fail in large parts of the diagram in the range of 0.6 and thus below the minimum robustness index R z,min,False Fail which in this embodiment is 0.7.

[0063] In Fig. 2b shows that the monitored signal e Q , which of the deviations from the normal value e Q0The result is a smaller deviation, since at least one parameter of a control unit monitoring function of a vehicle diagnostic system has been improved using this method. Consequently, the robustness index is also in the range between 0.95 and 1 across much of the diagram. The probability of a misdiagnosis by a vehicle diagnostic system, a "false fail" error, is thus reduced. This also reduces the probability of an incorrect activation of the fault indicator, such as the erroneous illumination of a MIL (Malfunction Indicator Lamp).

[0064] Also in Fig. 3a and Fig. 3b is the improvement through the procedure taking into account the robustness index R. z,min_threshold,False Pass against a positive error threshold being undercut g QposThe procedure illustrates that the probability of a misdiagnosis by a vehicle diagnostic system, a "false pass" error, is reduced after the improvement. This also reduces the probability of a fault indicator being triggered incorrectly, for example, the false illumination of a MIL (Malfunction Indicator Lamp).

Claims

[1] Method for checking and, if necessary, improving the setting of at least one parameter of an ECU monitoring function of a vehicle diagnostic system, taking into account at least one robustness indicator, in particular taking into account four robustness indicators, - where at least one robustness parameter R z [n] is calculated according to the following equation: Rz[n]=1rmax_pos / max_neg⋅N0∑i=nn+N0ri - where r max_pos / max_neg the largest positive or largest negative distance value when monitoring against a positive or a negative deviation is, - where N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, - where n is an element of a data series of consecutive feature values, - where r ithe distance of a characteristic value from a positive or negative error threshold, determined on the basis of the raw distance, - where the setting of at least one parameter is retained if the calculated value R z or the calculated values ​​R z above a minimum robustness index R z,min lies / lie - and where the setting of at least one parameter is adjusted when the calculated value R z or the calculated values ​​R z below the minimum robustness index R z,min lies / lie or the calculated value R z or the calculated values ​​R z the minimum robustness index R z,min corresponds / correspond to, - wherein, in order to check the setting of at least one parameter, the results of the control unit monitoring function are calculated by inputting an input data set of a functioning system and an input data set of a non-functioning system into the control unit monitoring function, characterized by , - that, to check the setting of at least one parameter, the results determined by the control unit monitoring function are compared with certain values ​​of a characteristic of a functioning system, and, to monitor against a positive error threshold exceedance, a value R is calculated. z,max_threshold,False Fail to check whether a functioning system is incorrectly identified as non-functional, - that, in order to verify the setting of at least one parameter, the results obtained by the control unit monitoring function are compared with certain values ​​of a characteristic of a functioning system for monitoring against a negative error threshold undercutting by calculating a value R z,min_threshold,False Fail to check whether a functioning system is incorrectly identified as non-functional, - that, in order to verify the setting of at least one parameter, the results determined by the control unit monitoring function are compared with certain values ​​of a feature of a non-functional system for monitoring against a positive error threshold undercutting by calculating a value R z,max_threshold,False Pass to check whether a non-functional system is incorrectly identified as functional, - and that, to verify the setting of at least one parameter, the results obtained by the control unit monitoring function are compared with certain values ​​of a feature of a non-functional system for monitoring against a negative error threshold exceedance by calculating a value R z,min_threshold,False Pass It should be checked whether a non-functional system is incorrectly identified as functional. [2] Method according to claim 1 characterized by , that for the calculation of R z,max_threshold,False Fail The following definition is made: Rz,max_threshold,False Fail[n]=1N0∑nn+N0ri|e[i]−gQpos[i]eQ0[i]−gQpos[i]| e[i]={eQ[i]…eQ0[i]≤eQ[i]≤gQpos[i]eQ0[i]…eQ[i]<eQ0[i]gQpose[i]…eQ[i]> gQpos[i] - where -N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, - where n is an element of a data series of consecutive feature values, - where e Q [i] the feature value is, - where g Qpos [i] the positive threshold is, - where e Q0 [i] the normal value is, - and where e[i] is the limited measurement value. [3] Method according to claim 1 or 2 characterized by , that for the calculation of R z,min_threshold,False Fail The following definition is made: Rz,min_threshold,False Fail[n]=1N0∑nn+N0ri|e[i]−gQneg[i]eQ0[i]−gQneg[i]| e[i]={eQ[i]…gQneg[i]≤eQ[i]≤gQ0[i]eQ0[i]…eQ[i]<eQ0[i]gQnege[i]…eQ[i]> gQneg[i] - where N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, - where n is an element of a data series of consecutive feature values, - where e Q [i] the feature value is, - where g Qneg[i] the negative threshold is, - where e Q0 [i] the normal value is, - and where e[i] is the limited measurement value. [4] Method according to any one of claims 1 to 3, characterized by , that for the calculation of R z,max_threshold,False Pass The following definition is made: Rz,max_threshold,False Fail[n]=1N0∑nn+N0ri|e[i]−gQpos[i]eQ0[i]−gQpos[i]| e[i]={eQ[i]…gQpos[i]≤eQ[i]≤eQ0[i]eQ0[i]…eQ[i]<eQ0[i]gQpose[i]…eQ[i]> gQpos[i] eQ0[i]={gQpos[i]×(1+m100)…gQpos[i]>0gQpos[i]×(1−m100)…gQpos[i]<0 - where N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, - where n is an element of a data series of consecutive feature values, - where e Q [i] the feature value is, - where g Qpos [i] the positive threshold is, - where e Q0 [i] the normal value is, - and where e[i] is the limited measurement value. [5] Method according to any one of claims 1 to 4, characterized by , that for the calculation of R z,min_threshold,False Pass The following definition is made: Rz,min_threshold,False Pass[n]=1N0∑nn+N0ri|e[i]−gQneg[i]eQ0[i]−gQneg[i]| e[i]={eQ[i]…eQ0[i]≤eQ[i]≤gQ0[i]eQ0[i]…eQ[i]<eQ0[i]gQnege[i]…eQ[i]> gQneg[i] - where N0 is the number of data points in the data series that represents the minimum time, in particular the debounce time, for error detection, - where n is an element of a data series of consecutive feature values, - where e Q [i] the feature value is, - where g Qneg [i] the negative threshold is, - where e Q0 [i] the normal value is, - and where e[i] is the limited measurement value. [6] Method according to any one of claims 1 to 5, characterized by , - that the setting of at least one parameter is retained when the calculated value R z,max_threshold,False Fail and the calculated value R z,min_threshold,False Fail above a minimum robustness index R z,min,False Fail lay, - or that the setting of at least one parameter is retained when the calculated values ​​R z,max_threshold,False Fail and the calculated values ​​R z,min_threshold,False Fail above a minimum robustness index R z,min,False Fail lay, - and that the setting of at least one parameter is adjusted when the calculated value R z,max_threshold,False Fail and the calculated value R z,min_threshold,False Fail below the minimum robustness index R z,min,False Fail lie or the minimum robustness index R z,min,False Fail are equivalent to, - or that the setting of at least one parameter is adjusted when the calculated values ​​R z,max_threshold,False Fail and the calculated values ​​R z,min_threshold,False Fail below the minimum robustness index R z,min,False Faillie or the minimum robustness index R z,min,False Fail are equivalent to. [7] Method according to any one of claims 1 to 6, characterized by , - that the setting of at least one parameter is retained when the calculated value R z,max_threshold,False Pass and the calculated value R z,min_threshold,False Pass above a minimum robustness index R z,min,False Pass lay, - or that the setting of at least one parameter is retained when the calculated values ​​R z,max_threshold,False Pass and the calculated values ​​R z,min_threshold,False Pass above a minimum robustness index R z,min,False Pass lay, - and that the setting of at least one parameter is adjusted when the calculated value R z,max_threshold,False Pass and the calculated value R z,min_threshold,False Pass below the minimum robustness index R z,min,False Pass lie or the minimum robustness index R z,min,False Pass are equivalent to, - or that the setting of at least one parameter is adjusted when the calculated values ​​R z,max_threshold,False Pass and the calculated values ​​R z,min_threshold,False Pass below the minimum robustness index R z,min,False Pass lie or the minimum robustness index R z,min,False Pass are equivalent to. [8] Method according to any one of claims 1 to 7, characterized by , - that R z,min,False Fail and / or R z,min,False Pass in a range of 0.5 to 0.9, particularly in a range of 0.6 to 0.

8. [9] Method according to any one of claims 1 to 8, characterized by , - that R z,min,False Fail and / or R z,min,False Pass 0.7 is. [10] Method according to any one of claims 1 to 9, characterized by , - that the specific values ​​of the characteristic are values ​​measured on a system, or that the specific values ​​of the characteristic are empirically or arbitrarily determined values. [11] Method according to any one of claims 1 to 10, characterized by , - that the vehicle diagnostic system is part of a vehicle - and / or that the vehicle diagnostic system is an on-board diagnostics "OBD" system, - and / or that the vehicle diagnostic system is part of an ECU (Engine Control Unit) or an ECU (Engine Control Unit) of a motor vehicle. [12] Method according to any one of claims 1 to 11, characterized by , - that the control unit monitoring function and the checked and, if necessary, improved parameters are included in the vehicle diagnostic system, - or that the control unit monitoring function and the checked and, if necessary, improved parameters are transferred to a vehicle diagnostic system. [13] Method for controlling a fault indicator of a vehicle diagnostic system, in particular for controlling a MIL lamp “Malfunction Indicator Light – Engine Control Lamp” of a vehicle, wherein the fault indicator signals a fault when the vehicle diagnostic system diagnoses a fault, characterized by , that the setting of at least one parameter of the control unit monitoring function of the vehicle diagnostic system is carried out according to one of claims 1 to 12. [14] Method according to claim 13, characterized by , that the vehicle diagnostic system has diagnosed a fault: - if a deviation between at least one specific value of a characteristic and at least one calculated value of a characteristic is greater than at least one positive error threshold, - and / or if a deviation between at least one specific value of a characteristic and at least one calculated value of a characteristic is less than at least one negative error threshold, - and / or if at least one release condition is met when this deviation occurs, - and / or if a debounce time is exceeded when this deviation occurs. [15] Method according to claim 13 or 14, characterized by , - that at least one specific value of a feature is measured, determined, or transmitted to the vehicle diagnostic system, - and that at least one calculated value of a feature is determined by the vehicle diagnostic system, in particular by the control unit monitoring function, or transmitted to the vehicle diagnostic system.

Images