A method for generating a virtual prototype of a vehicle with multiple wheels.
The method uses road measurements to iteratively simulate and adapt tire models, addressing the inefficiencies of conventional tire behavior analysis by creating accurate virtual prototypes of multi-wheeled vehicles.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- AVL LIST GMBH
- Filing Date
- 2024-05-24
- Publication Date
- 2026-06-23
AI Technical Summary
Conventional methods for analyzing vehicle tire behavior require extensive real-world test drives, which are costly and time-consuming, and existing simulation tools lack the ability to accurately generate virtual prototypes of multi-wheeled vehicles without extensive physical testing.
A method and system for generating a virtual prototype of a multi-wheeled vehicle using road measurements to indirectly determine Pacejka parameters through iterative simulation, comparing measured and simulated traction parameters, and adapting tire models to match actual tire behavior.
Enables the creation of a high-quality virtual prototype of a vehicle with multiple wheels at a lower cost and in a shorter time, accurately simulating tire traction behavior without the need for additional test runs.
Smart Images
Figure 2026520427000001_ABST
Abstract
Description
[Technical Field]
[0001] The present invention relates to a method for generating a virtual prototype of a multi-wheeled vehicle based on data from road measurements, a computer program or storage medium including instructions for performing such a method, and a system for generating a virtual prototype of a vehicle based on data from road measurements. [Background technology]
[0002] In vehicles, the wheels ensure the necessary power transmission to the roadway. Particularly relevant to power transmission are the tires, usually made of rubber, which form the contact between the vehicle and the roadway. Conventional technology is known for analyzing the behavior of wheels and tires based on physical data.
[0003] In the vehicle-road system, tires play a crucial role as the connecting element between the roadway and the vehicle, transmitting all forces and moments. Their power transmission and behavior significantly impact the overall driving behavior, comfort, and safety of the vehicle. Pneumatic tires are primarily supported by trapped gas under high pressure, with only a small portion of the wheel load directly supported by the tire structure. Tire characteristics are influenced by the shape of the contact patch, structural design, and material use. The development of tires for passenger cars and trucks is strongly influenced by the ever-changing and increasing demands on automobiles. Electric vehicles, in particular, generally experience increased tire wear due to the heavier weight of their batteries and the higher torque and rotational gradient from their electric motors. Tire performance characteristics represent the individual characteristics of the tire. Road measurements are performed to determine these performance characteristics. Theoretical explanations of tire characteristics are generally known through the work of Pacejka. In this context, the following academic papers are referenced as examples: Pacejka, HB; Besselink, IIM: Magic Formula Tire Model with Transient Properties. Lisse, the Netherlands, Swets & Zeitlinger BV, 1997, pp. 234-249
[0004] To analyze the behavior of a vehicle using real tires under as many relevant driving conditions, road conditions, and environmental conditions as possible, a very large amount of test driving distance must be covered.
[0005] Furthermore, such real-world test drives cannot be performed during vehicle development and can only be carried out in the later stages of vehicle development. Conventional technology generally suggests the possibility of performing virtual test drives using vehicle simulation tools. However, these vehicle simulation tools require a virtual prototype of the vehicle. [Overview of the project] [Problems that the invention aims to solve]
[0006] The object of the present invention is to provide a virtual prototype of a vehicle with multiple wheels. In particular, the object of the present invention is to automate the generation of a virtual prototype of a vehicle with multiple wheels as much as possible. [Means for solving the problem]
[0007] This problem is solved by teaching in the independent claim. A favorable configuration is claimed in the dependent claim.
[0008] A first aspect of the present invention is a computer implementation method for generating a virtual prototype of a multi-wheeled vehicle based on data from road measurements, in particular for indirectly measuring the values of Pacejka parameters, comprising the following work steps, i.e. S1) A work step of providing a tire database that includes multiple tire datasets containing Pacejka parameters, S2) A work step that provides a vehicle model including a digital twin of the vehicle and a tire model that can be adapted using a tire dataset, S3) A work step of providing a tire dataset for the tire model from the tire database, S4) A work step in which a measurement run is performed using a vehicle, wherein the measurement run includes a load event, and during the load event, the measured value of the traction parameter of at least one wheel is determined. S5) A work step of simulating a load event using a vehicle model, wherein at least one simulated value of the traction parameter of at least one wheel is output as a target amount. S6) A work step in which the values of the traction parameters measured in work step S4 are compared with the values of the traction parameters simulated in work step S5, S7) The work step includes modifying the tire dataset so that the simulated values of the traction parameters match the measured values of the traction parameters by changing the values of the Pacejka parameters, Work steps S5 to S7 are repeated until the termination condition is met, and then, S8) Work step to output the values of the Pacejka parameters of the tire model. This includes methods.
[0009] A second aspect of the present invention relates to a method for analyzing a vehicle tire set, wherein the vehicle tire set is simulated by a virtual prototype of a vehicle generated by the method of any one of the above claims.
[0010] A third aspect of the present invention relates to a system for generating a virtual prototype of a vehicle based on data from road measurements, in particular for indirectly measuring the values of Pacejka parameters, the system comprising means for parameterizing a tire model of the virtual prototype including Pacejka parameters, wherein the means for parameterizing is designed to iteratively and sequentially determine the values of Pacejka parameters by comparing simulated values of traction parameters with measured values of traction parameters determined by road measurements, in a simulation loop in which the parameters of the tire model are optimized based on measured values from road measurements, particularly by cascading software-in-the-loop simulation.
[0011] A fourth aspect of the present invention is a system for generating a virtual prototype of a vehicle based on data from road measurements, particularly according to claim 19, comprising means for parameterizing a tire model, wherein the means for parameterizing is as follows: A means for calculating at least one measured value of a tire traction parameter based on the measured values recorded during a test run, A means for calculating at least one value relating to the tire slip ratio based on the measured values recorded during the test run, A means for simulating a vehicle using a tire model, wherein the tire model has at least the following physical characteristics of the vehicle, namely Vehicle weight, wheelbase, track width, center of gravity, and steering gear ratio This is incorporated as a parameter. At least the traction parameter value is output as the target amount. means and Means for comparing at least one value of a measured traction parameter with at least one value of a simulated traction parameter, By changing the Pacejka parameters, a means is provided to adapt the vehicle model so that the simulated traction parameters match the measured traction parameters determined based on road measurements. Includes an interface for outputting values for Pacejka parameters of the tire model, The parameterization method is designed to adapt the tire model until a termination condition is reached. Regarding the system.
[0012] Further aspects of the present invention relate to a computer program and a storage medium that, when executed by a computer, includes instructions causing a computer to perform a method according to the present invention.
[0013] In this invention, road measurement is preferably on-site measurement, that is, measurement performed during the actual driving operation of a vehicle. The part on which the wheel rolls is called the tire. The wheel refers to the entire unit consisting of the rim and the tire.
[0014] Software-in-the-loop simulation in the sense of the present invention is preferably a simulation in which a component represented by software is tested in a virtual model world.
[0015] In the sense of the present invention, traction parameters preferably represent the characteristics of a tire. In particular, the traction parameters of a tire are characteristics that represent the behavior of the tire during power transmission. Traction parameters are, in particular, grip, i.e., the coefficient of friction, slip ratio or slip angle, or quantities derived therefrom. Grip G or coefficient of friction is defined as a force F parallel to the road plane. P And a force F perpendicular to the road surface. z The ratio is given, that is, G = F P / F zThe slip angle represents the ratio between the rolling direction of the tire and the moving direction of the tire. The ratio of grip to slip angle, or the ratio of the force in the X direction to the slip angle, or the ratio of the force in the Y direction to the slip angle, are also examples of traction parameters.
[0016] The Pacejka parameters are a set of tire model parameters used in vehicle dynamics to describe the forces and moments generated between the tire and the road surface. These are used to predict the behavior of the vehicle at various speeds, load conditions, and road surfaces. The Pacejka parameters are based on actual measurements and are suitable for modeling the behavior of the vehicle. The mathematical formula for describing the forces and moments generated between the tire and the road surface is called the Magic Formula or Pacejka formula, and was developed by the Dutch engineer Hans B. Pacejka. The Pacejka parameters describe characteristics such as tire stiffness, coefficient of friction, and tire shape, and model the behavior of the tire.
[0017] The Pacejka parameters are different for each tire. A tire data set contains multiple Pacejka parameters related to a tire or set of tires. There are various Pacejka parameter sets that can differ depending on the application. Therefore, the number of Pacejka parameters can be different for different tire data sets. In the simplest case of the Pacejka formula, the Pacejka parameters only include six different values. The Pacejka formula, which is constantly being developed, has more than 100 parameters in its current form. Tire data sets also preferably have a large number of such parameters as well. The conventional Pacejka formula basically represented only the static case, i.e., the static and constant slip behavior within the steady state region, but the improved version of the latest Pacejka formula can also represent dynamic tire characteristics.
[0018] This method is not only used for the purpose of generating a virtual prototype, but can also be equivalently represented as an indirect measurement method, especially for indirectly measuring the Pacejka parameters. The physical state of a tire is defined by the measurable physical properties of an object at a specific point in time. According to the present invention, a method of performing a simulation using the actual measurement data of a vehicle, especially the speed, acceleration, rotational speed, rotation number, and torque of the tire, is described. These data are used as inputs, and the simulation determines the Pacejka parameters of the tire model as outputs.
[0019] The Pacejka parameters represent the physical characteristics of a tire. For example, the Pacejka parameter A represents the lateral stiffness, which indicates how the tire reacts to lateral forces, the Pacejka parameter B represents the lateral peak coefficient, which shows the degree of non-linearity of the lateral force according to the slip angle, or the Pacejka parameter C represents the lateral shape coefficient, which affects the shape of the side force curve. For all other Pacejka parameters, there are corresponding physical relationships.
[0020] Therefore, the claimed method utilizes the measurement of a real object as an input, provides the physical state of an existing object in reality, and each step thereof contributes to the technical realization of the method.
[0021] The digital twin of a vehicle means the digital representation of a physical vehicle in the real world in the digital world. Whether the vehicle already exists in the real world or will be realized in the future is not important. The digital twin enables comprehensive data exchange, consists of models of individual elements of the vehicle, and can also include simulations, algorithms, and services that describe the characteristics or behaviors of the vehicle.
[0022] A load event refers to a time-limited event in which at least the vehicle's tires are subjected to a load different from that of a stationary vehicle or a vehicle moving without acceleration. Load events may include, or may include, acceleration maneuvers such as full-load acceleration, braking maneuvers such as full-load deceleration, cornering with a constant or variable radius, or other driving maneuvers.
[0023] In step S4, if the measured traction parameters of at least one wheel are determined, this could mean that the traction parameters of one wheel are determined, or that the traction parameters of multiple wheels are determined, or that a single traction parameter is determined from a combination of wheels, for example, two wheels mounted on the same axle. In particular, in step S4, the intention is to determine the measured traction parameters of at least one tire. The tire is an important wheel component in this step. The same applies to the output of the simulated traction parameters in step S5. The same parameter, namely the ratio of grip to slip ratio, is used for both the measured and simulated traction parameters, but the specific measured or simulated values of the two are different.
[0024] In particular, an IMU (Inertial Measurement Unit) is used to determine the measured values of traction parameters during the test run. The IMU is equipped with a gyroscope that can determine the vehicle's rotational speed in three axes, an accelerometer that can determine the vehicle's acceleration in three directions, and a GPS system that can determine the vehicle's position in three dimensions. However, the measurement of at least the vehicle's z-direction velocity, z-direction acceleration, and z-direction rotational speed (i.e., perpendicular to the roadway) is optional. Further measurements during the test run include the wheel rotation speed and torque.
[0025] Further advantages are achieved if, in step S4), the measured values of the traction parameters of at least one wheel are determined by the speed, acceleration, and rotational speed of the vehicle parallel to the roadway, as well as the rotational speed and torque of the tire. In a particular embodiment of the present invention, all steps of the method performed in relation to the wheel are performed in relation to the tire.
[0026] Preferably, rather than the individual values of each measured or simulated traction parameter, multiple values of each measured traction parameter are determined, or multiple values of each simulated traction parameter are output. This means that, with respect to the values of measured traction parameters, several measured traction parameters are determined during the load event. With respect to the values of simulated traction parameters, this means that several values of the simulated traction parameters over the load event are output as target quantities.
[0027] This can improve the accuracy of fitting the modeled Pacejka parameters.
[0028] The tire dataset fitting in step S7 can be performed by completely replacing the tire dataset and / or fitting the individual Pacejka parameters within the tire dataset. The goal of this step is to match the simulated traction parameters with the calculated traction parameters, which can be done with or without a defined goal. In particular, the fitting in step S7 of the loop is intended to enable the simulation of load events by the vehicle model using each tire dataset in the tire database.
[0029] In particular, termination conditions may include the number of iterations of steps S5-S7, or, in the case of simulating all tire datasets available in the tire database, reaching the final simulation.
[0030] In step S8, the Pacejka parameters that best demonstrate the comparison results in step S6 and / or the Pacejka parameters for which the censorship conditions are met are output. The output of the Pacejka parameter values for the tire model is equivalent to the selection of parameters for the tire model. Thus, the values of the Pacejka parameters are selected with respect to the tire model. This output thus creates a virtual prototype of a vehicle with multiple wheels.
[0031] The present invention is based on an approach that allows the Pacejka parameters of a vehicle's tires to be determined by an iterative simulation method relating to a virtual prototype. In this way, the driving behavior of the vehicle's tires can be simulated without requiring further test runs using a test vehicle. Thus, with respect to the tires, a vehicle model can be created at low cost, in a short time, and with verifiable high quality. Here, depending on the type of tire, the traction behavior, i.e., the tire grip, can be simulated with particular accuracy. The method according to the present invention allows for the automatic creation of a vehicle model based on measurement data from road measurements.
[0032] Preferably, in the method according to the first embodiment, the traction parameters are intended to include grip, and / or slip ratio, and / or slip angle.
[0033] Grip refers to the ratio of the force acting on a tire parallel to the road to the force acting perpendicular to the road. Slip ratio is the ratio of the speed at which the tire moves on the road to the speed at which the vehicle as a whole moves forward or backward. A high slip ratio means the tire rotates faster than the vehicle. Slip angle indicates the angle between the direction of wheel rotation and the direction in which the wheel is moving. In particular, traction parameters derived from this quantity may also be considered.
[0034] Further advantages are achieved if steps S4 to S8 are performed for each wheel, especially each driven wheel.
[0035] Explicitly, this means that in step S4, the measured traction parameters are determined for each wheel, in particular each driven wheel. With respect to step S5, this means that at least one simulated traction parameter from each wheel, in particular each driven wheel, is output as a target quantity. With respect to step S6, this means that the traction parameter values measured in work step S4 are individually compared with each other's traction parameter values simulated in work step S5 for each wheel.
[0036] A good compromise between accuracy and cost for some target setting is to determine only the measured traction parameters from each driven wheel and output only the simulated traction parameters from each driven wheel as the target quantity, especially when the load event is acceleration, particularly full-load acceleration.
[0037] In the special embodiments described last of the present invention, it is preferable that the conformance quality is calculated by the least squares method.
[0038] In the least squares method, the set of data points for simulated traction parameters is fitted to be as close as possible to the set of data points for measured traction parameters. Therefore, in this particular embodiment of the present invention, in step S4, multiple measured traction parameter values are determined from one wheel, and in step S5, multiple simulated traction parameter values for at least one tire are output as target quantities.
[0039] In a further advantageous embodiment, the method further includes the step of adapting the vehicle control function based on the value output in step S8.
[0040] In a more advantageous form, this method further includes a step of controlling and / or adjusting the vehicle based on the value output in step S8.
[0041] The output values of Pacejka parameters can function as control parameters in a vehicle, or influence control parameters in a vehicle. This allows for the configuration of vehicle functions that enable particularly efficient operation of the vehicle.
[0042] More preferably, the vehicle model is intended to take into account the vehicle's suspension.
[0043] The suspension is a component within a vehicle that connects the wheels to the chassis or body and controls the vertical, lateral, and horizontal movement of the wheels. Composed of various parts such as spring elements, dampers, control arms, and axles, the suspension works to absorb road irregularities and maintain contact between the wheels and the road, enabling stable and controlled driving. Because the suspension affects the contact area and angle between the tire and the road, it directly impacts tire grip. A properly fitted suspension can improve grip by maximizing contact between the tire and the road and keeping the tire in the optimal position to transmit longitudinal and lateral forces to the road.
[0044] Considering the suspension in vehicle models allows for the more optimized generation of virtual prototypes of vehicles.
[0045] Preferably, paving is intended to be further considered in step S5.
[0046] Further advantages are achieved if the load event involves acceleration. The acceleration may be, in particular, full-load acceleration.
[0047] In a further specific embodiment of the present invention, the load event is intended to have a delay. The delay may, in particular, be a full load delay.
[0048] Particularly preferably, the load event is intended to include cornering with a constant radius and speed increase.
[0049] Further benefits are achieved if the censorship condition is reaching a minimum, particularly local or absolute, of the deviation between the measured value of the traction parameter and the simulated value of the traction parameter.
[0050] More preferably, in order to determine the measured values of the traction parameters in step S4, the measured quantities of vehicle speed, vehicle acceleration, vehicle rotational speed, wheel rotational speed, and wheel torque are intended to be acquired.
[0051] In a further preferred embodiment of the present invention, the fitting of the tire dataset in step S7 is intended to include selecting a tire dataset for a tire model from a tire database.
[0052] This allows all Pacejka parameters to be changed simultaneously, which, as a general rule, enables a rough optimization of the simulation quality.
[0053] In a further advantageous embodiment of the present invention, the fitting of the tire dataset in step S7 is intended to include fitting of the individual Pacejka parameters of the selected tire dataset.
[0054] This allows for particularly precise optimization of simulation quality. In particular, by selecting the tire dataset, it is also possible to perform fitting of individual values from the selected tire dataset after optimization.
[0055] Further advantages can be achieved in vehicle models when vehicle weight, wheelbase, track width, center of gravity, and steering gear ratio are taken into consideration.
[0056] In a system according to a fourth aspect of the present invention, the means may preferably further include means for calculating at least one value relating to the lateral force of the tire based on the values of the measured quantities recorded during the measurement run.
[0057] Furthermore, in a particular embodiment of the present invention, the term "includes" can also mean "is".
[0058] Further features and advantages are evident herein in conjunction with the drawings, which are shown at least partially schematicly. [Brief explanation of the drawing]
[0059] [Figure 1] This figure shows an exemplary embodiment of a method for generating a virtual prototype of a vehicle. [Figure 2] This figure plots the measured and simulated values of traction parameters for an unfitted tire dataset. [Figure 3] Figure 2 shows a plot of the traction parameters of the tire dataset, comparing the measured and simulated values fitted to each other. [Figure 4] This figure shows an exemplary embodiment of a system for generating a virtual prototype of a vehicle. [Modes for carrying out the invention]
[0060] Figure 1 shows an exemplary embodiment of method S0 for generating a virtual prototype of vehicle 10 based on data from road measurements.
[0061] In step S1, a tire database 12 is provided that includes multiple tire datasets 14 containing Pacejka parameters.
[0062] In step S2, a vehicle model 15 is provided, which includes a digital twin 16 of the vehicle and a tire model 18 that can be adapted by the tire dataset 14.
[0063] In step S4, a measurement run is performed using vehicle 10, which includes a load event during which the values of traction parameters from multiple tires are determined. The determination of these values is performed by the measurement and calculation steps. These values are called the "measured values" of the traction parameters. The load event is acceleration. The traction parameter is the ratio of the force acting on both tires of the driven axle to the slip acting on both tires of the driven axle. For this purpose, two tires of the driven axle are combined. The measured values can be obtained either via a data interface or directly by the sensors during the measurement run.
[0064] Steps S1, S2, and S4 are independent of each other in their order. Step S3 requires the tire database 12 and the tire model 18. Therefore, step S3 is performed after steps S1 and S2.
[0065] In road measurements, the vehicle 10 is used to perform measurement runs on a driving surface, particularly on a road. For this purpose, the vehicle 10 is equipped with measuring instruments and sensors. In particular, the vehicle 10 has an Initial Measurement Unit (IMU) for measuring rotational speed in three axes, acceleration in three directions, and the position of the vehicle in three dimensions. During the measurement run, the wheel rotation speed, vehicle speed in the longitudinal and lateral directions, vehicle rotational speed in the longitudinal and lateral directions, and torque acting on each tire are further determined. In road measurements, the following vehicle parameters are further assumed: the vehicle's static weight, wheelbase, tread width, center of gravity in three dimensions, and steering gear ratio, i.e., the ratio of steering wheel rotation to wheel rotation on the ground. Further optional measurements include the vehicle's speed, acceleration, and rotational speed in the direction perpendicular to the roadway.
[0066] Following steps S1 to S4, in step S5, the load events that occurred during the measurement execution using the vehicle 10 are simulated by the vehicle model. Here, the simulated value 24 of the traction parameter from at least one tire is output as the target amount.
[0067] In a vehicle model, the vehicle's weight, wheelbase, track width, center of gravity, and steering gear ratio are taken into consideration. The vehicle model also further considers the suspension. All of these parameters influence slip and the forces acting on the wheels in various directions, and are represented in the vehicle model.
[0068] Therefore, in steps S4 and S5, the same parameters are determined once as measured values in the test run and once as simulated values in the simulation. Thus, the simulated values 24 of the traction parameters are directly comparable to the measured values 22 of the traction parameters.
[0069] Following step S5, in step S6, the traction parameter value 22 measured in step S4 is compared with the traction parameter value 24 simulated in step S5.
[0070] For comparison, conformance quality is calculated. The calculation is performed using the least squares method.
[0071] Further details regarding the comparison will be discussed in relation to Figure 2.
[0072] In step S7, the tire dataset is adapted so that the simulated value 24 of the traction parameter matches the measured value 22 of the traction parameter by changing the value of the Pacejka parameter of the tire model. To do this, a different tire dataset for the tire model is first selected from the tire database. Using this new tire dataset, steps S5 and S6 are repeated, that is, the simulation of load events using the vehicle model and the comparison of the simulated value 24 of the traction parameter obtained therefrom with the measured value 22 of the traction parameter determined in work step S4 are repeated. Work step S4, i.e., the execution of a measurement run using the vehicle 10, is not repeated here. The measured values obtained during a single measurement run are referenced.
[0073] Steps S5-S7 are repeated until the termination condition is reached. In the first selected example, all tire datasets 14 present in the tire database 12 are used to simulate load events using the vehicle model, and the termination condition is reached as soon as the simulated values 24 of the traction parameters determined from this simulation are compared with the measured values 22 of the traction parameters.
[0074] The fit quality is determined in each comparison. This determination is performed using the least squares method. Then, the tire dataset with the highest fit quality is selected.
[0075] In step S8, the output Pacejka parameters form the tire dataset of the tire model 18 of the vehicle model, thus completing the method at this point. In this way, the generation of a virtual prototype of the vehicle 10 is completed.
[0076] However, as an alternative or addition, work steps S5-S7 can be repeated further before outputting the values of the Pacejka parameters, which may be intended to be done by further fitting one or more individual values of the Pacejka parameters in the tire dataset 14 with the highest fit quality. Using the fitted Pacejka parameters, steps S5 and S6 are run again to perform further optimization of the fit of the simulated traction parameters to the measured traction parameters. These fittings can also be repeated many times in work steps S5-S7 until a censorship condition is reached.
[0077] This censorship condition is determined, in particular, by an optimization problem. Preferably, such a censorship condition may be reaching a local or absolute minimum of the deviation between the measured value 22 of the traction parameter and the simulated value 24 of the traction parameter.
[0078] Furthermore, the termination condition may also be reaching the limit value of the simulated traction parameter 24, particularly when the simulated value 24 of the traction parameter changes only slightly.
[0079] Alternatively, during the measurement run, the following driving maneuvers can also be performed depending on the traction parameters to be determined, namely, Tip-In, Tip-Out, full-load acceleration, partial-load acceleration, uphill driving, and downhill driving.
[0080] Figure 2 shows a plot of measured and simulated values of traction parameters for a tire dataset 14 that is not fitted with measured values. Measured traction parameters 22 are represented by dots. Simulated traction parameters 24 are marked with an "x".
[0081] In Figure 2a, the traction parameter, the force in the x-direction, i.e., the force in the direction of travel, is plotted against the traction parameter, the slip ratio. The measured values were recorded during load events of a test run using the vehicle. The traction parameters measured are the force in the x-direction, the force in the z-direction, and the grip. The force in the x-direction and the force in the z-direction are forces acting on both wheels of the driven front axle, respectively. The measured values are determined by measuring the vehicle's position, speed, and rotation in multiple directions, and the vehicle's weight, its wheelbase, tread width, center of gravity, and steering gear ratio are used as input variables in the calculations.
[0082] In Figure 2a, the longitudinal force on the tire pair is plotted against the slip ratio.
[0083] Here, the slip ratio S is given by S = (Ω·R C It is given by ) / v-1. Here, Ω represents the angular velocity of the wheel, and R C This is the effective radius of the freely rolling tire, and can be calculated from the total number of wheel rotations per kilometer. Quantity v represents the vehicle's forward speed.
[0084] The slip ratio indicates how strongly or slowly the wheels are spinning relative to the speed of the vehicle 10. A slip ratio of 0 means the wheels are not slipping and are rotating at the same speed as the vehicle 10, while a slip ratio of 1 means the wheels are rotating at twice the speed of the moving road surface.
[0085] Regarding the measured values in Figure 2a, the vertical force F increases with increasing slip ratio. x We can see that the force increases. In contrast, the force in the z direction decreases with increasing slip ratio. The ratio of these two quantities corresponds to the grip plotted in Figure 2c, and the grip also increases with increasing slip ratio. In Figures 2a, 2b, and 2c, in addition to the measured values, simulated values of the force and grip in the x and z directions are also plotted for the first tire dataset.
[0086] In the selected example, it can be seen that the measured x-direction force and grip are basically higher than the simulated x-direction force or the simulated grip. For the z-direction force, the behavior is reversed.
[0087] In FIGS. 3a to 3c, the same measured values 22 of the traction parameters (F x , F z , and grip) are plotted against the slip ratio. The simulated values of the traction parameters were created using another fitted tire data set. The simulated values 24 of the traction parameters shown were created by simulating the load event using the vehicle model, similar to FIG. 2. Comparing with the example shown in FIG. 2, it can be seen that the deviation between the measured values 22 of the traction parameters and the simulated values 24 of the traction parameters is quite small. The simulated values of all three traction parameters are located at the center of the variation of the measured values 22 of the traction parameters over the entire range of the measured slip ratio. When such a match is achieved between the measured value and the simulated value of the traction parameter, the threshold of the fitting quality may be reached. This corresponds to reaching the termination condition. After reaching the termination condition, the best-fitted values of the Pacejka parameters of the tire model, that is, the tire data set used in this simulation, are output. Through the output, a virtual prototype of the vehicle 10 is created. By the steps of the above method, it is guaranteed that the tire model of the virtual prototype corresponds to the actual vehicle 10.
[0088] Figure 4 shows an exemplary embodiment of a system 40 for generating a virtual prototype of a vehicle 10 based on data from road measurements, the system 40 comprising means 41, 42, 43, 44, and 45 for parameterizing a tire model 18 of the virtual prototype. Here, the parameterizing means 41, 42, 43, 44, and 45 are designed to determine the values of the Pacejka parameters of the tire model 18 in a simulation loop based on the measured values from road measurements by a cascading software-in-the-loop simulation. In the simulation loop, the parameters of the tire model 18 are optimized so that the simulated values 24 of the traction parameters iteratively and sequentially match the measured values 22 of the traction parameters determined by road measurements.
[0089] In particular, system 40 is designed to perform the method shown in Figure 1. Preferably, but not limited thereto, system 40 includes means 41 for calculating at least one value relating to the longitudinal force of the tire based on the values of the measured quantities recorded during the measurement run.
[0090] Furthermore, the system 40 preferably includes means 42 for calculating at least one value relating to the lateral force of the tire based on the values of the measured quantities recorded during the measurement run.
[0091] More preferably, the system 40 includes means 43 for calculating at least one value relating to the tire slip ratio based on the values of the measured quantities recorded during the measurement run.
[0092] More preferably, the system 40 includes means 44 for simulating a vehicle using a tire model 18, the tire model 18 including at least the following physical characteristics of the vehicle as parameters: vehicle weight, wheelbase, tread width, center of gravity, and steering gear ratio, where at least the traction parameter value is output as a target quantity.
[0093] More preferably, the system 40 includes means 45 for comparing at least one measured value of a traction parameter determined based on road measurements with a simulated value 24 of the traction parameter.
[0094] More preferably, the system 40 includes means 46 for adapting the vehicle model 18 to match the simulated value 24 of the traction parameter to the measured value 22 of the traction parameter by changing the Pacejka parameter.
[0095] The system 40 preferably further comprises an interface 47 for outputting values relating to the Pacejka parameters of the tire model. The parameterization means is preferably designed to adapt the tire model until a censorship condition is reached.
[0096] The means 41, 42, 43, 44, 45, and 46 of system 40, as well as interface 47, are preferably part of a data processing device. Preferably, method S0 is performed automatically by such a data processing system and / or in a computer implementation.
[0097] The means 41, 42, 43, 44, 45, 46 and interface 47 described above are also designed, among other things, to run multiple simulation loops of method S0.
[0098] It should be noted that the exemplary embodiments are merely examples and do not limit in any way the scope of protection, uses, and structure. Rather, the above description provides guidance to those skilled in the art for carrying out at least one exemplary embodiment, and various modifications can be made, in particular with respect to the function or arrangement of the described components, without departing from the scope of protection obtained from the claims and combinations thereof equivalent features. [Explanation of symbols]
[0099] 10 vehicles 12 Tire Database 14 Tire Datasets 16 Digital Twin 18-inch tire model 22. Measurement values of traction parameters 24. Simulated values of traction parameters 40 Systems 41. Means for calculating at least one value relating to the longitudinal force of a tire. 42. Means for calculating at least one value relating to the lateral force on a tire. 43. Means for calculating at least one value relating to the slip ratio of a tire. 44. Means for simulating a vehicle using a tire model. 45. Means for comparing at least one value of a traction parameter calculated based on road measurements with a simulated value of a traction parameter. 46. Means for adapting vehicle models 47 Interfaces S0 method
Claims
1. A computer implementation method (S0) for generating a virtual prototype of a multi-wheeled vehicle (10) based on data from road measurements, in particular for indirectly measuring Pacejka parameters, comprising the following work steps: S1) A work step of providing a tire database (12) that includes multiple tire datasets (14) containing Pacejka parameters, S2) A work step of providing a vehicle model including a digital twin (16) of the vehicle (10) and a tire model (18) that can be adapted by a tire data set (14), S3) A work step of providing a tire dataset (14) for the tire model (18) from the tire database (12), S4) A work step in which a measurement run is performed using the vehicle (10), wherein the measurement run includes a load event, and during the load event, a measured value (22) of the traction parameter of at least one tire is determined. S5) A work step of simulating the load event using the vehicle model, wherein at least one simulated value (24) of the traction parameter of at least one of the tires is output as a target amount, S6) A work step in which the value of the traction parameter (22) measured in work step S4 is compared with the value of the traction parameter (24) simulated in work step S5, S7) The work step of adjusting the tire dataset (14) so that the simulated value (24) of the traction parameter matches the measured value (22) of the traction parameter by changing the value of the Pacejka parameter, Work steps S5 to S7 are repeated until the termination condition is met, and then, S8) A work step of outputting the values of the Pacejka parameters of the tire model (18) Methods that include...
2. The method according to claim 1, wherein in step S4), the measured value (22) of the traction parameter of at least one of the wheels is determined by measuring the speed, acceleration, and rotational speed of the vehicle parallel to the roadway, as well as the rotational speed and torque of the wheel.
3. The method according to claim 1 or 2 (S0), wherein the traction parameters include grip, and / or slip ratio, and / or slip angle.
4. The method according to any one of claims 1 to 3 (S0), wherein steps S4 to S8 are performed for each of the driven wheels.
5. The method according to any one of claims 1 to 4 (S0), wherein the termination condition includes reaching a threshold of conforming quality.
6. The method according to claim 5 (S0), wherein the conforming quality is calculated by the least squares method.
7. The method (S0) according to any one of claims 1 to 6, further comprising the step of controlling and / or adjusting the vehicle (10) based on the value output in step S8.
8. The method (S0) according to any one of claims 1 to 7, wherein the suspension of the vehicle (10) is taken into consideration in the vehicle model.
9. The method according to any one of claims 1 to 8, wherein paving is further considered in step S5.
10. The method according to any one of claims 1 to 9 (S0), wherein the load event has acceleration.
11. The method according to any one of claims 1 to 10 (S0), wherein the load event has a delay.
12. The method according to any one of claims 1 to 11 (S0), wherein the load event includes curved driving with a constant radius and speed increase.
13. The method according to any one of claims 1 to 12 (S0), wherein the termination condition is reaching a local or absolute minimum of the deviation between the measured value (22) of the traction parameter and the simulated value (24) of the traction parameter.
14. The method according to any one of claims 1 to 13 (S0), wherein for the determination of the measured value (22) of the traction parameter in step S4, the measured quantities of vehicle speed, vehicle acceleration, vehicle rotational speed, wheel rotational speed, and wheel torque are acquired.
15. The method according to any one of claims 1 to 14, wherein the fitting of the tire dataset (14) in step S7 includes selecting a tire dataset (14) for the tire model (18) from the tire database (12).
16. The method according to any one of claims 1 to 15 (S0), wherein the fitting of the tire dataset (14) in step S7 includes fitting of the individual Pacejka parameters of the selected tire dataset (14).
17. The method according to any one of claims 1 to 16 (S0), wherein the vehicle model takes into consideration the weight of the vehicle (10), the wheelbase, the tread width, the center of gravity of the vehicle (10), and the steering gear ratio.
18. A method (S0) for analyzing a vehicle tire set, wherein the vehicle tire set is simulated by a virtual prototype of the vehicle (10) generated by the method according to any one of claims 1 to 17.
19. A computer program or storage medium that, when executed by a computer, includes instructions causing the computer to perform the method (S0) according to the present invention as described in any one of claims 1 to 18.
20. A system for generating a virtual prototype of a vehicle (10) based on data from road measurements, in particular for indirectly measuring Pacejka parameters, the system including means for parameterizing a tire model (18) of the virtual prototype including Pacejka parameters, wherein the parameterizing means is designed to iteratively and sequentially determine the values of the Pacejka parameters by comparing a simulated value (24) of at least one traction parameter with a measured value (22) of the at least one traction parameter determined by the road measurements, in a simulation loop in which the parameters of the tire model (18) are optimized based on measured values from the road measurements, particularly by a cascaded software-in-the-loop simulation.
21. A system (40) according to claim 20, comprising means for parameterizing the tire model (18), in particular for generating a virtual prototype of a vehicle (10) based on data from road measurements, wherein the means for parameterizing is as follows: A means (41) for calculating at least one value relating to the longitudinal force of the tire based on the measured values recorded during the test run, Means (43) for calculating at least one value relating to the slip ratio of the tire based on the measured values recorded during the measurement run, A means (44) for simulating the vehicle (10) using a tire model (M), wherein the tire model (18) is provided with at least the following physical characteristics of the vehicle (10), namely The weight, wheelbase, tread width, center of gravity, and steering gear ratio of the aforementioned vehicle (10) This is incorporated as a parameter. A means (44) outputs the value (24) of at least one simulated traction parameter as the target quantity, Means (45) for comparing at least one measured value of the at least one traction parameter with at least one simulated value (24) of the at least one traction parameter, Means (46) for adapting a vehicle model so that the simulated value (24) of the at least one traction parameter matches the measured value (22) of the at least one traction parameter determined based on the road measurements, by changing the Pacejka parameter, The tire model (18) includes an interface (47) for outputting values for the Pacejka parameters, The parameterization means is designed to adapt the tire model (18) until a termination condition is reached. System (40).