Method for verifying design model for technical system
The method addresses the challenge of ensuring stability and certainty in technical systems by validating design models through a systematic process involving test signals, output detection, and parameter comparison, thereby improving the accuracy of model representations in real-world applications.
Patent Information
- Application Number
- JP2024184040
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-10-19
- Filing Date
- 2024-10-18
- Publication Date
- 2025-05-02
AI Technical Summary
Existing methods for verifying design models of technical systems often fail to guarantee stability and certainty in real-world applications, as they rely on assumptions about parameter values that may not accurately represent real conditions.
A method that involves providing a design model for a technical system, applying a test signal, detecting the output, determining the system state and parameter values, identifying disturbances, and validating the design model by comparing parameter and disturbance values to defined tolerances.
This method allows for the verification of design models under real conditions, identifying potential errors and uncertainties, and enabling adjustments to the technical system or design model to ensure stability and certainty.
Smart Images
Figure 2025071060000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a method for validating a design model for a technical system The invention further relates to a computer program, an apparatus and a memory medium for this purpose. [Background technology]
[0002] Prior Art Model-based controller design for technical systems is based on a design model that has been used in the development phase to model the technical system to be controlled. In particular, when attempting to provide stability and robustness guarantees, internal uncertainties and disturbances of the design model can be taken into account, which can be described as a parameter set, which itself can be based on quantities or can be stochastic. In the design phase, the technical system is designed with an assumed parameter set using the design model, in order to be able to guarantee stability and robustness, in particular taking into account the uncertainties and disturbances.
[0003] In general, for parameter identification of technical systems, in particular of control systems, different algorithms exist which estimate nominal parameter values or uncertainty quantifications exist which allow to estimate parameter distributions of uncertain parameters. Summary of the Invention [Problem to be solved by the invention]
[0004] However, in real-world applications of technical systems, assurance of stability and robustness may only be obtained so long as assumptions on the parameters of the design model correspond to behavior under real-world conditions. [Means for solving the problem]
[0005] Disclosure of the Invention The subject of the invention is a method having the features according to claim 1, a computer program having the features according to claim 8, an apparatus having the features according to claim 9 and a computer-readable memory medium having the features according to claim 10. Further features and details of the invention emerge from the respective dependent claims, the description and the drawings, in which case the features and details explained in relation to the method according to the invention naturally also apply in relation to the computer program according to the invention, the apparatus according to the invention and the computer-readable memory medium according to the invention and vice versa, so that the disclosures of the individual aspects of the invention are always or can be cross-referenced.
[0006] The subject of the invention is in particular a method for validating a design model for a technical system, the method comprising the steps described below, which may be performed repetitively and / or in succession. The technical system in particular comprises at least one control circuit. The technical system may be a control system.
[0007] In a first step, a design model is provided, which preferably models the technical system. The design model defines in particular tolerance ranges for at least one parameter of the technical system and for at least one disturbance quantity. The tolerance ranges may for example be a minimum and a maximum value, respectively, for the at least one parameter or for the at least one disturbance quantity, or a probability distribution. The at least one parameter may for example be the mass of the vehicle, the at least one disturbance quantity, for example a disturbance force. The tolerance ranges may be understood as assumptions for the at least one parameter or for the at least one disturbance quantity, and may for example be based on empirical values or may be determined by tests.
[0008] In a further step, the application of a test signal to the technical system is preferably initiated, which may for example be the value of a manipulated variable in the technical system or in a control circuit of the technical system.
[0009] In a further step, preferably triggered by the test signal, a detection of an output of the technical system is started, which may for example be a manipulated variable in the technical system or in a control circuit of the technical system or may be a measured value of a sensor of the technical system.
[0010] In a further step, a state of the technical system is determined, preferably based on the test signal and on the detected output, and a value of each of the at least one parameter of the technical system is determined based on this state, the state being present as a state vector based on or including the current values of the at least one parameter, in particular a number of parameters.
[0011] In a further step, disturbances in the technical system are preferably determined based on the test signals, the detected outputs and the determined states, and a value of each of at least one disturbance quantity is determined. The disturbance may thus be the totality of at least one disturbance quantity, or there may be a plurality of disturbance quantities. The disturbance quantity may be, for example, uncontrolled influences such as driving wind or friction or heating.
[0012] In a further step, a design model for the technical system is preferably verified, and a respective value of at least one parameter of the technical system and a respective value of at least one disturbance quantity are compared with tolerances of the design model, which advantageously makes it possible, on the one hand, to check whether the design model correctly models the technical system under real conditions and, on the other hand, to check whether errors exist in the technical system.
[0013] More advantageously, the method further comprises the steps of: - transmitting the detected output of the technical system to an external data processing device; The external data processing device may for example be a server and may be wirelessly or wired connected to the technical systems to provide two-way communication. The identification of conditions, the identification of disturbances and the validation of the design model may additionally or alternatively be performed by the external data processing device. This advantageously allows outsourcing at least a part of the computing costs. Furthermore, this advantageously allows analyzing multiple outputs from multiple individual technical systems in the external central data processing device, so that for example the design model can be optimized based on the outputs of these multiple technical systems.
[0014] It is further conceivable that the technical system is a vehicle. Thus, the application of the test signal can be performed by at least one actuator of the vehicle and the detection of the output triggered by the test signal can be performed by at least one sensor of the vehicle. This allows the method according to the invention to be used in a vehicle. The vehicle may for example be configured as an automobile and / or a passenger car and / or an autonomous vehicle. The vehicle may for example comprise a vehicle device for providing autonomous driving functions and / or driver assistance systems. The vehicle device may be configured to at least partly automatically control and / or accelerate and / or brake and / or steer the vehicle.
[0015] Advantageously within the framework of the present invention, the method further comprises: Implementing at least one measure affecting the technical system and / or the design model based on the results of the validation. Thus, advantageously, besides the validation, it is also possible to provide a corresponding response, both with respect to the technical system and with respect to the design model, based on the results of this validation.
[0016] Within the scope of the present invention, the implementation of at least one measure - adjusting at least one function of the technical system, thereby increasing the reliability of the technical system; Further advantages can be achieved if the technical system includes: the adjustment can be, for example, a reduction or deactivation of a function of the technical system. It is conceivable that the reduction or deactivation of the function is carried out for a predefined time period. If the technical system is a vehicle, it can be ascertained, for example, within the framework of a validation, that a large disturbance is present, i.e. that a value of at least one disturbance quantity is present that is outside the tolerance range of the design model, and that the vehicle cannot be driven without adjusting the at least one function as long as this disturbance, i.e. in particular this value of the at least one disturbance quantity, is outside the tolerance range of the design model. Furthermore, if the technical system is a vehicle, the adjustment of the at least one function can include an adjustment of the planned trajectory of the vehicle. Thus, for example, it can be ascertained based on the output of the technical system that the design model did not accurately depict reality, so that the planned trajectory can be adjusted precisely, advantageously taking the output into account.
[0017] Within the scope of the present invention, the implementation of at least one measure - adjusting the design model based on results of the validation, whereby the design model for the technical system is individualized. This allows the design model to be advantageously iteratively adjusted and optimized to the behavior of the technical system under real-world conditions.
[0018] For example, the design model may be used as a reference for monitoring the technical system and may be identified based on modeling of the technical system. The design model may describe the technical system based at least on equations and assumptions about at least one parameter and at least one disturbance quantity. Furthermore, the state may represent at least one physical energy storage quantity of the technical system and the disturbance may represent at least one environmental influence on the technical system. In the example where the technical system is a vehicle, the state may be, for example, the speed of the vehicle and the disturbance may be the running wind.
[0019] A computer program, in particular a computer program product, which comprises instructions for causing a computer to carry out the method according to the invention when the computer program is executed by a computer, is likewise a subject of the invention, whereby the computer program according to the invention offers the same advantages as those detailed above in connection with the method according to the invention.
[0020] An apparatus for data processing, adapted to carry out the method according to the invention, is also the subject of the invention. The apparatus may, for example, be a computer for executing the computer program according to the invention. The computer may have at least one processor for executing the computer program. A non-volatile data memory may also be provided, in which the computer program may be stored and from which the computer program may be read for execution by the processor.
[0021] The present invention also relates to a computer-readable memory medium which carries the computer program according to the invention and / or which contains instructions for causing a computer to carry out the steps of the method according to the invention when executed by a computer. The memory medium is configured, for example, as a data memory such as a hard disk and / or a non-volatile memory and / or a memory card. The memory medium may, for example, be integrated into the computer.
[0022] Additionally, the methods of the present invention may be practiced as computer-implemented methods.
[0023] Further advantages, features and details of the invention will become apparent from the following description, in which an embodiment of the invention is explained in detail with reference to the drawings, in which the features mentioned in the claims and in the description may each individually or in any combination be essential to the invention. [Brief description of the drawings]
[0024] [Figure 1] 1 is a schematic visualisation of a method, an apparatus, a memory medium and a computer program according to an embodiment of the invention; [Diagram 2] 1 is a schematic diagram of a technical system, actuators, sensors and an external data processing device according to an embodiment of the invention; [Diagram 3] 2 is a schematic diagram of a method according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0025] In FIG. 1 a method 100, an apparatus 10, a memory medium 15 and a computer program 20 according to an embodiment of the invention are shown diagrammatically.
[0026] FIG. 1 shows in particular an embodiment of a method 100 for validating a design model 2 for a technical system 1, the method 100 comprising the following steps: In a first step 101, a design model 2 is provided which models the technical system 1. The design model 2 then defines tolerances for at least one parameter of the technical system 1 and for at least one disturbance quantity. In a second step 102, application of a test signal to the technical system 1 is started. In a third step 103, detection of an output of the technical system 1 triggered by the test signal is started. In a fourth step 104, a state of the technical system 1 is determined based on the test signal and the detected output, and a respective value of at least one parameter of the technical system 1 is determined based on this state. In a fifth step 105, disturbances in the technical system 1 are determined based on the test signal, the detected output and the determined state, and a respective value of at least one disturbance quantity is determined. In a sixth step 106, the design model 2 for the technical system 1 is verified and each value of at least one parameter of the technical system 1 and each value of at least one disturbance quantity are compared with the tolerance ranges of the design model 2.
[0027] In FIG. 2 a technical system 1, an actuator 3, a sensor 4 and an external data processing device 5 according to an embodiment of the invention are shown diagrammatically.
[0028] One aspect of the present invention according to an embodiment is that parameter estimation algorithms are used during the run-time of a technical system 1, in particular to validate assumptions made during the development phase of a design model 2 for a particular technical system 1.
[0029] If the assumptions for the parameters are violated during run time, in particular the tolerances, then at least one measure can be implemented, for example to prevent malfunction of the technical system 1. If the parameters are still within the tolerances but start to drift (for example due to aging), the technical system 1 can, according to further possible measures, further use this information to optimize the design model 2. The information for each individual system regarding the violation of the assumptions can advantageously also be aggregated for several technical systems 1 and can be forwarded to an external data processing device 5, and thus in particular to a development team, in order to improve the assumptions for the parameters of the design model 2, so that in future updates for several technical systems 1 the output of the technical system 1 can be improved by reconfiguring the design model 2.
[0030] If the technical system 1 is an autonomous vehicle, the design model 2 may be transferred to a trajectory planner or may be used to adjust the vehicle trajectory. In this case, the best adjusted input trajectory for the individual vehicle can be selected, thus ensuring the best output or a high degree of certainty for the individual vehicle. Parameter estimation may already be performed on a prototype of the technical system 1 before the technical system 1 is released in order to identify possible violations of assumed parameters. In case of violation of assumptions, the release can be interrupted, for example, by the latest control software. The information of which assumptions for the parameters were incorrect and the quantitative information can then be used directly to adjust the design model 2.
[0031] Parameter estimation may be performed during run-time on the technical system 1. The determined values of the parameters are preferably compared with assumptions for the parameters of the design model 2 from the design phase, i.e. in particular with tolerance ranges. If the determined values are outside the tolerance ranges of the assumed parameter distributions, the technical system 1 can respond, for example, by measures of adjusting at least one function, i.e. by measures of moving the technical system 1, for example, into a degraded state or is isolated and preventing new switch-ons.
[0032] According to one embodiment, the identified parameters can be uploaded to a development cloud where developers can monitor the parameter distribution in multiple technical systems 1 (e.g. a vehicle platoon) and thereby improve the assumptions of these technical systems 1 about the distribution of parameter values, i.e. in particular the design model 2.
[0033] The design model 2 for the technical system 1 can also be adjusted online during run time for each individual system (e.g. for each individual vehicle) in order to improve the performance for the individual system implementations of the technical system 1. The individual system implementations also vary, for example, over the useful life of the technical system 1.
[0034] The parameter validation method according to one embodiment, and in particular the validation method of the design model 2, can generally be used for all kinds of technical systems 1, where model-based controller design methods are applied.
[0035] To clearly illustrate the method according to one possible embodiment, the method is described below in the context of lateral control of a highly automated vehicle (e.g., SAE Level 4). The design model 2 for lateral control is preferably derived from a linearized single trajectory model. The state vector can be, for example,
number
number
[0036] The resulting state space representation of the technical system 1 in the form of a closed control system is
number
number
number
number
[0037] The system matrix is, in particular, the parameter vector Θ=[c f ,c r ,l f ,l r ,m,J z ,v veh ,...] TIn a real technical system 1, in particular a vehicle, these parameters may for example not be constant but may vary within certain limits, i.e. the technical system 1 may have uncertainties regarding the parameters. Furthermore, the input
number
[0038] 3 shows a schematic diagram of a method according to one embodiment, illustrating the information flow inside a technical system 1, in particular a vehicle. The following steps are applied, preferably cyclically: in a first step 201, a test signal in the form of a control input u is supplied to the technical system 1 in the form of a control system. In a second step 202, a response y of the technical system 1 is measured with the aid of at least one sensor 4, for example an inertial measurement unit (IMU) or other vehicle sensor. In a third step, on the one hand, a state x of the technical system 1 is estimated 203a and, on the other hand, a disturbance y is measured.
number
[0039] The above description of the embodiments has been given by way of example only, and the individual features of the embodiments can of course be freely combined with one another, in so far as this makes sense in technical terms, without departing from the scope of the invention.
Claims
1. A method (100) for validating a design model (2) for a technical system (1), said method (100) comprising: - providing (101) a design model (2) for modelling said technical system (1), said design model (2) defining tolerances for at least one parameter and at least one disturbance quantity of said technical system (1); - initiating (102) the application of a test signal to said technical system (1); - starting (103) a detection of the output of said technical system (1), triggered by said test signal; - determining (104) a state of the technical system (1) based on the test signal and on the detected output, whereby a value of each of the at least one parameter of the technical system (1) is determined based on said state; - a step (105) of identifying a disturbance in the technical system (1) based on the test signal, the detected output and the identified state, in which a value of each of the at least one disturbance quantity is determined; - a step (106) of validating the design model (2) for the technical system (1), in which the respective values of the at least one parameter of the technical system (1) and the respective values of the at least one disturbance quantity are compared with the tolerance ranges of the design model (2); A method (100).
2. The method (100) comprises: - transmitting the detected output of the technical system (1) to an external data processing device (5). Further comprising:
2. The method (100) of claim 1, wherein the identification of states, the identification of disturbances and the validation of the design model (2) are additionally performed by the external data processing device (5).
3. said technical system (1) being a vehicle, the application of said test signal is performed by at least one actuator (3) of said vehicle, 3. The method (100) according to claim 1 or 2, wherein the detection of the output triggered by the test signal is performed by at least one sensor (4) of the vehicle.
4. The method (100) comprises: - implementing at least one action affecting said technical system (1) and / or said design model (2) based on the results of said validation (106); The method (100) of any one of claims 1 to 3, further comprising:
5. The implementation of the at least one measure comprises: - adjusting at least one function of said technical system (1), said adjusting increasing the reliability of said technical system (1). The method of claim 4 , comprising:
6. The implementation of the at least one measure comprises: - adjusting the design model (2) based on the results of the validation (106), by which the design model (2) is individualized for the technical system (1), The method (100) of claim 4 or 5, comprising:
7. said design model (2) being used as a basis for monitoring said technical system (1) and being determined on the basis of a modelling of said technical system (1); The method (100) according to any one of the preceding claims, wherein the state represents at least one physical energy storage of the technical system (1) and the disturbance represents at least one environmental influence on the technical system (1).
8. A computer program (20) comprising instructions for causing a computer (10) to carry out a method (100) according to any one of claims 1 to 7, when the computer program (20) is executed by the computer (10).
9. An apparatus (10) for processing data, configured to perform a method (100) according to any one of the preceding claims.
10. A computer-readable memory medium (15) comprising instructions which, when executed by a computer (10), cause the computer (10) to perform the steps of the method (100) of any one of claims 1 to 7.