Method and vehicle system for determining the condition of the components of a chassis

Abstract

Method for determining the state of the components of an individual chassis of an individual vehicle, characterized by: - Providing an initial dataset of vehicle data, which includes at least load and / or wear data of the same or similar chassis type of the individual vehicle over its entire service life, - Generating an individual second data set in an individual vehicle by recording the vehicle data as a target state up to a predefined first mileage and / or a specified age of the individual vehicle, - Recording of currently measured vehicle data from a predefined second mileage reading and / or a specified age of the individual vehicle, - Performing a comparison of the currently measured vehicle data with the first data set as well as the second data set to determine the condition, whereby - the first data set is provided as a generically generated reference data set by an identical or similar chassis type tested on a test bench in a long-term test, and - the reference data set is generated separately for each of the two axles of the identical or similar chassis type.

Description

[0001] The invention relates to a method and a vehicle system for determining the state of the components of an individual chassis of an individual vehicle. The invention further relates to a vehicle.

[0002] A vehicle typically has onboard diagnostic systems that monitor vehicle components and systems, for example, for routine inspections. These systems compare vehicle data such as the mileage. Furthermore, diagnostic systems have sensors that monitor and report malfunctions in the vehicle's or a component's operation. However, such systems only report a fault if a parameter falls outside an acceptable range.

[0003] There is currently no way to predict a possible malfunction or failure due to, for example, wear and tear.

[0004] However, component wear can lead to serious problems regarding the reliability of the components and the more complex systems into which they are integrated. In fact, excessive wear of a basic component can damage the entire assembly to which it is integrated. Furthermore, excessive wear of, for example, brake pads can lead to dangerous damage to the brake assembly, thereby compromising vehicle safety.

[0005] From US Patent 2016 / 0078690 A1, a method for determining the state of an individual vehicle system of an individual vehicle is known. In this method, when comparing a current state with an expected state, an aging effect is taken into account in the target state, which specifies the aging behavior of the vehicle system as a function of time. Furthermore, an individual second data set is generated for an individual vehicle by recording vehicle data as the target state up to a predefined initial mileage and / or a specified age of the individual vehicle. Currently measured vehicle data is recorded from a predefined initial mileage and / or a specified age of the individual vehicle onwards. The currently measured vehicle data is compared with the modified data, whereby the modified data is determined from the first and second data sets.

[0006] Furthermore, it is known from DE 10 2018 211 047 B4 or WO 2020 / 031133 A1 or DE 10 2018 123 821 A1 that the condition of a road surface can be taken into account or can represent an input parameter.

[0007] DE 10 2018 119652 A1 discloses a vehicle comprising a chassis with at least one chassis component and a driver assistance system for detecting the state of the chassis component, wherein the driver assistance system comprises an evaluation unit and a first acoustic sensor that detects sound generated by the chassis component during the vehicle's operation and converts it into an electrical signal, and the evaluation unit uses the electrical signal to determine at least one piece of information about the state of the chassis component.

[0008] It is therefore an object of the invention to provide a method and a vehicle system in which the condition of the chassis can be determined more precisely, in order to, for example, increase driving safety. Furthermore, it is an object to provide a vehicle.

[0009] This problem is solved by a method having the features of claim 1 and by a vehicle system having the features of claim 9 and a vehicle having the features of claim 14.

[0010] Beneficial training courses, which can be used individually or in combination, are listed in the dependent requirements and in the description.

[0011] The task is solved by a procedure for determining the state of the components of an individual chassis of an individual vehicle, comprising the following steps: - Providing an initial dataset of vehicle data, which includes at least load and / or wear data of the same or similar chassis type of the individual vehicle over its entire service life, - Generating an individual second data set in an individual vehicle by recording the vehicle data as a target state up to a predefined first mileage and / or a specified age of the individual vehicle, - Recording of currently measured vehicle data from a predefined second mileage reading and / or a specified age of the individual vehicle, - Performing a comparison of the currently measured vehicle data with the first data set as well as the second data set to determine the condition.

[0012] The term "chassis" can preferably refer to the two axles of the vehicle as well as their components, or to just one axle as well as its components.

[0013] According to the invention, it has been recognized that current assessments of the condition of a chassis primarily rely on data acquired from acceleration sensors. These data are used to analyze changes in vibration behavior. However, previous investigations have not yielded a single, definitive indicator for accurately assessing chassis wear. The replacement of worn and new chassis components increases the number of possible combinations almost infinitely.

[0014] However, if an incorrect assessment takes place, driving safety may be jeopardized, especially if relevant component parts such as wheel suspension / brakes etc. are directly or indirectly affected.

[0015] The invention recognizes that it is insufficient to make a statement about wear / condition solely based on the currently recorded data. According to the invention, a first data set of vehicle data is provided, which includes at least the load and / or wear data of the same or similar chassis type of the individual vehicle over its entire service life. This data set can be determined on the test field.

[0016] The first data set is provided in a vehicle with a similar or identical chassis.

[0017] According to the invention, it was further recognized that individual learning of the individually equipped vehicle is necessary to define a target state. Therefore, when the vehicle rolls off the assembly line as a new car, the target state is learned by recording the vehicle data. This learning process remains active for a certain period of time and / or a certain number of kilometers driven. During this process, a second data set is generated, for example, from recorded vibrations. It was recognized according to the invention that the target state must be recorded not only generically, but also based on the individual vehicle, since a multitude of individual equipment configurations each generate different target data sets. Subsequently, the current vehicle data is measured or recorded from a predefined second mileage reading and / or a defined age of the individual vehicle. The second mileage reading can, for example, be shortly after the first.

[0018] Furthermore, according to the invention, a comparison is then carried out using the currently measured vehicle data with both the first and the second data set. In other words, according to the invention, the current vehicle data are compared with the target data or target state, as well as the current vehicle data with the first data set representing the actual state during a service life cycle.

[0019] This comparison between the actual state during the service life cycle and the individually defined target state generates a state value that allows for a reliable assessment of the actual condition of a chassis. For example, at the end of the service life cycle, the statement "fully worn" can be more reliably made based primarily on the first data set, with the aid of the second data set. Furthermore, by comparing the two data sets, statements such as "80% fully worn" or "20% as new" can be made.

[0020] The invention enables a clear and reliable assessment of the chassis condition by utilizing both the new condition of the individual vehicle as a starting point and information about the entire service life of a similar or identical chassis. An assessment is made based on the distance from this new condition, as well as a comparison with the initial data set as a representative service life cycle. This allows for a reliable statement regarding the condition.

[0021] In a further refinement, the second dataset is generated by assigning the currently measured vehicle data to a road surface cluster representing the road surface, or by creating a new road surface cluster if the road surface is previously unknown and assigning the currently measured vehicle data to the newly created road surface cluster. To achieve improved matching and comparison, clustering of the road surface is preferably performed. If the road surface has not yet been recorded, a new cluster is created and saved with the recorded vehicle data as a second dataset. This allows for the training of an improved second dataset, enabling a more accurate determination of the chassis condition later on.

[0022] According to the invention, the first data set is provided as a generically generated reference data set by an identical or similar chassis type tested on a test bench in a long-distance test.

[0023] The first data set can be easily determined through a long-distance running test performed on a test bench. This test bench makes it easy to identify specific frequency ranges, i.e., to generate, for example, footprints that can be used for later comparison.

[0024] Preferably, such an initial dataset can be generated using a purely analytical approach, an artificial neural network, or another machine learning method. Generating the initial dataset on a test bench also offers the advantage that component parts can be exchanged. For example, brake pads can be replaced regularly, which then generate different vibrations / frequencies in conjunction with the older, partially worn surrounding components. This allows the initial dataset to be expanded and improved, enabling a better determination of the current chassis condition of the individual vehicle during a later comparison.

[0025] According to the invention, the reference data set is generated separately for each of the two axles. This allows for a more targeted comparison between the current vehicle data recorded at the front of the vehicle and the current vehicle data recorded at the rear of the vehicle.

[0026] In a further embodiment, the currently measured vehicle data and, as a second data set, at least the vibration amplitudes of the vehicle movements on characteristic road surfaces are used. This allows the specific frequency ranges to be generated and assigned to identified road surfaces.

[0027] In a further refinement, an actual state is determined by comparing the currently measured vehicle data with the first data set, and a target state is determined by comparing the currently measured vehicle data with the second data set. The current wear of the chassis components is determined by this target-actual comparison. The comparisons between the data sets and the currently measured vehicle data can also be weighted differently. This allows the current wear of the chassis and its components to be determined. The comparison compares the data with a data set that characterizes the life cycle with corresponding wear, as well as with a data set that represents the new condition. This enables a reliable assessment of the chassis's condition.

[0028] According to further training, the vehicle continuously or adaptively measures current vehicle data. Adaptive measurement can be based, for example, on the vehicle's age, mileage, or the occurrence of specific situations.

[0029] Furthermore, the currently measured vehicle data can include the mileage driven and / or the vehicle's age. This allows for verification of the plausibility of the comparison or the result. For example, the mileage driven can be used to represent average load profiles, and the age can be used to track the aging of components such as rubber bearings.

[0030] Furthermore, the task is solved by a vehicle system for determining the state of the components of an individual chassis of an individual vehicle, including: - a storage unit for providing an initial data set of vehicle data, wherein the initial data set includes at least the load and / or wear data of the same or similar chassis type of the individual vehicle over its entire service life, - the storage unit for providing an individual second data set in an individual vehicle, wherein the individual second data set comprises the measured vehicle data as a target state up to a predefined first mileage and / or a specified age of the individual vehicle by the one or more sensors, - a sensor system for recording currently measured vehicle data from a predefined second mileage reading and / or a specified age of the individual vehicle, - a comparison unit for performing a comparison between the currently measured vehicle data with the first data set as well as the second data set to determine the condition.

[0031] The advantages of the process can also be transferred to the vehicle system.

[0032] The sensor system can consist of several different sensors and different sensor types.

[0033] The comparison unit can be designed as a processor.

[0034] In a further embodiment, the sensor system is designed to detect the road surface. A processor is also preferably provided for generating the second data set by assigning the currently measured vehicle data to a road surface cluster representing the road surface, or by creating a new road surface cluster if the road surface is previously unknown, assigning the currently measured vehicle data to the newly created road surface cluster, and storing the second data set in the memory unit. The processor is connected to the sensor system and the memory unit for communication purposes. Furthermore, the processor, comparison unit, and memory unit can also be configured as a single module.

[0035] In a further embodiment, the comparison unit is designed to determine an actual state by comparing the currently measured vehicle data with the first data set, and to determine a target state by comparing the currently measured vehicle data with the second data set, and furthermore to determine the current wear of the chassis components by means of a target-actual comparison.

[0036] In a further development, the sensor system is designed to continuously or adaptively measure the current vehicle data.

[0037] In addition, the currently measured vehicle data may include the kilometers driven and / or the age of the vehicle.

[0038] Furthermore, the task is solved by a vehicle with a vehicle system as described above.

[0039] Further features and advantages of the present invention will become apparent from the following description with reference to the accompanying figures. These schematically illustrate: Fig. 1: an embodiment of a method according to the invention, and Fig. 2: an evaluation in the frequency range of a control arm over a service life period, and Fig. 3: a vehicle system according to the invention.

[0040] Fig. Figure 1 shows an embodiment of a method according to the invention for determining the state of the components of an individual chassis of an individual vehicle.

[0041] In a preliminary step (S0), an initial data set is generated. This is created based on an identical or similar chassis installed in the individual vehicle. This initial data set is generated through a long-term test on a test track and reflects the chassis's service life cycle from new condition to complete wear. The long-term test on the test track allows for the detection of specific vibrations and frequencies (frequency ranges), each corresponding to a particular condition of the chassis. For example, a louder squealing noise occurs during braking when wear is already present.

[0042] This long-distance running test makes it easy to identify specific frequency ranges, i.e., to generate footprints that can be used for later comparison. The initial data set can be generated using a purely analytical approach or by an artificial neural network or another machine learning method.

[0043] However, worn components can be replaced on the test bench. This replacement, along with the resulting initial data set, primarily containing vibrations and frequencies, is recorded. This allows the initial data set to be expanded and completed. The first data set thus contains information about new components, worn components, and components that are both worn and still usable, comprehensively reflecting the frequencies, vibrations, and load data over the entire service life of the chassis. When compared later with the current chassis condition, this initial data set, combined with the measured vehicle data of the individual vehicle, contributes to a more precise determination of the actual condition of the chassis in that specific vehicle.

[0044] Generating the initial data set on a test bench also offers the advantage that component parts can be deliberately and knowingly replaced. For example, brake pads can be replaced regularly, which then generate different vibrations / frequencies in conjunction with the older, partially worn surrounding components. This allows the initial data set to be expanded and improved. This initial data set essentially represents a generic data set.

[0045] This allows for the simplified and cost-effective generation of generic data. Furthermore, generating such an initial dataset ensures the reliability of creating / implementing the second dataset, for example, by using the first dataset for plausibility checks.

[0046] Fig. Figure 2 shows an evaluation in the frequency range of a control arm over a service life period. The frequency spectrum at the control arm ranges from 3.5 to 4.5 Hz.

[0047] Initially, an 86% increase in frequency and progressive damage are observed. After replacing the tie rod at approximately 100,000 km (time T1), the frequency drops by about 9%. With further mileage, the frequency increases again by 28%. This decrease is reduced by approximately 24% when the engine hydraulic mounts are replaced at approximately 200,000 km (time T2). Over the course of the service life, significant damage occurs, resulting in a further 15% increase in the measured frequency.

[0048] By creating a generic data set on the test bench, individual footprints, i.e., frequency profiles for the respective chassis, can be created for each axle and used for comparison.

[0049] Creating the first data set on the test bench offers the advantage that the replacement of various components is precisely known, and furthermore, it allows for a better determination of how the frequency ranges react to these changes. This also makes it possible, for example, to identify how sensitive the frequencies / vibrations are to the failure of different components.

[0050] Furthermore, creating such an initial data set allows for the cost-effective coverage of an entire chassis life cycle. This initial data set is then integrated into an individual vehicle with an identical or similar chassis.

[0051] In a first step, S1, an individual second data set is generated for an individual vehicle by recording the vehicle data as a target state up to a predefined initial mileage and / or a specified age of the individual vehicle. This takes into account that each / many vehicles have individual features / characteristics that generate individual vibrations / frequencies.

[0052] This ensures an individual learning process based on different equipment options.

[0053] When the vehicle rolls off the assembly line, a learning process is performed to determine the target state and generate the second data set. This learning process remains active for a certain period of time or a specific number of kilometers driven.

[0054] To generate a second dataset that is as representative as possible, the road surface is captured by a sensor system, identified, and divided into clusters. This sensor system can consist of cameras, radar, and other sensors, and can also include information from the navigation system. This allows for the differentiation of gravel roads, tarmac surfaces, and new versus old road surfaces, for example. Ideally, these can be further subdivided according to weather conditions.

[0055] During the learning process, if new vibrations / frequencies are detected, they are assigned to one of these clusters for later comparison. If no cluster representing the road surface exists, a new one is created, for example, by a processor. This allows for the learning of an improved second data set, and thus a more accurate determination of the chassis's condition later on.

[0056] In a second step, S2, sensors record current vehicle data, i.e., vibrations and frequencies. These sensors can be, for example, accelerometers. The current vehicle data is recorded starting from a predefined second mileage reading and / or from a specified age of the individual vehicle. The age and the second mileage reading are determined after the first mileage reading.

[0057] The vehicle can continuously or adaptively collect current vehicle data.

[0058] In a third step, S3, the currently measured vehicle data is compared with both the first and second data sets. This comparison allows the current state of the vehicle to be determined. The wear and tear can be determined by comparing the current vehicle data with the first data set. Essentially, this comparison establishes the current condition of the vehicle components.

[0059] The comparison between the current vehicle data and the second data set corresponds to a comparison between the current state and the target (new) state. Essentially, the difference to the new state is determined, and a target state is established based on a comparison between the currently measured vehicle data and the second data set.

[0060] The current wear of the chassis components is determined by comparing the actual and target values.

[0061] This allows for a target-actual comparison to obtain information about the condition of the chassis. Comparisons are made to the new condition as well as, for example, to the end condition (fully worn) or service life condition. Thus, a result could be, for example, "80% fully worn" or "20% as new".

[0062] Furthermore, the data sets can be weighted differently. For example, with a newer vehicle, the comparison with the second data set can be given more weight in the final result than with an older vehicle.

[0063] Additionally, parameters such as mileage or the vehicle's overall age can be used to verify plausibility. Mileage can be used to represent average load profiles, and age can be used to account for the aging of components such as rubber bearings.

[0064] The method according to the invention enables a clear and reliable assessment of the condition of a chassis. This allows driving safety to be increased or worn components to be replaced in a timely manner.

[0065] Fig. Figure 3 schematically shows the vehicle system 1 according to the invention for determining a state of the components of an individual chassis of an individual vehicle.

[0066] The vehicle system 1 comprises a storage unit 2 for providing the data set of vehicle data, wherein the first data set represents the load and / or wear data of the same or similar chassis type of the individual vehicle over its entire service life. This first data set is preferably generated cost-effectively on the test track in a long-term test.

[0067] Storage unit 2 can, for example, be integrated into a control unit. A second, individual data set is stored in storage unit 2 by a sensor system 3. This data set was generated by the vehicle's own sensor system 3. The second data set is generated up to a predefined initial mileage and / or a specified vehicle age and is then defined as the target state. Depending on the road surface, the second data set is divided into several clusters, reflecting the specific road surface conditions.

[0068] The same sensor system 3 can be used to acquire and measure current vehicle data. This data can be measured continuously or adaptively. Vehicle data generally includes vibrations and their frequencies. These can be detected, for example, by accelerometers.

[0069] In a comparison unit 4, a target-actual comparison can be performed using the first and second data sets. Comparison unit 4 can be configured as a processor. The processor can also be integrated into the control unit.

[0070] Furthermore, an output unit 5 may be provided. This could, for example, be a display or a cockpit display. If a target / actual comparison value exceeds a predefined value, the comparison unit 4 may be configured to issue a warning message via the output unit 5 or to suggest an inspection appointment. Reference sign 1 Vehicle system 2 storage units 3 Sensor system 4 comparison unit 5 output units T1, T2 Time S0-S3 steps

Claims

Method for determining the condition of the components of an individual chassis of an individual vehicle, characterized by: - ​​Providing a first data set of vehicle data, which includes at least load and / or wear data of the same or similar chassis type of the individual vehicle over its entire service life, - Generating an individual second data set in an individual vehicle by recording the vehicle data as the target state up to a predefined first mileage and / or a specified age of the individual vehicle, - Recording currently measured vehicle data from a predefined second mileage and / or a specified age of the individual vehicle, - Performing a comparison of the currently measured vehicle data with the first data set as well as the second data set to determine the condition.wherein- the first data set is provided as a generically generated reference data set by an identical or similar chassis type tested on a test bench in a long-term test, and- the reference data set is generated separately for each of the two axles of the identical or similar chassis type. Method according to claim 1, characterized in that the second data set is generated by assigning the currently measured vehicle data to a road surface cluster representing the road surface or by creating a new road surface cluster if the road surface is previously unknown and assigning the currently measured vehicle data to the newly created road surface cluster. Method according to one of the preceding claims, characterized in that the currently measured vehicle data and the second data set include at least the vibration amplitudes of the vehicle movements on characteristic road surfaces. Method according to one of the preceding claims, characterized in that an actual state is determined by comparing the currently measured vehicle data with the first data set, and a target state is determined by comparing the currently measured vehicle data with the second data set, and the current wear of the chassis components is determined by a target-actual comparison. Method according to one of the preceding claims, characterized in that the current vehicle data are measured continuously or adaptively by the vehicle. Method according to one of the preceding claims, characterized in that the currently measured vehicle data includes the kilometers driven and / or the age of the vehicle. Vehicle system (1) for determining the condition of the components of an individual chassis of an individual vehicle, comprising: - a storage unit (2) for providing a first data set of vehicle data, wherein the first data set includes at least the load and / or wear data of the same or similar chassis type of the individual vehicle over its entire service life, - the storage unit (2) for providing an individual second data set in an individual vehicle, wherein the individual second data set includes the measured vehicle data as a target state up to a predefined first mileage and / or a specified age of the individual vehicle by means of one or more sensors, - a sensor system (3) for acquiring currently measured vehicle data from a predefined second mileage and / or a specified age of the individual vehicle,- a comparison unit (4) for performing a comparison between the currently measured vehicle data with the first data set as well as the second data set for determining the condition, wherein - the first data set is provided as a generically generated reference data set by an identical or similar chassis type tested on a test bench in a long-term test, and - the reference data set is generated separately for each of the two axles of the identical or similar chassis type. Vehicle system (1) according to claim 7, characterized in that the sensor system (3) is designed to detect the road surface and a processor is further provided for generating the second data set by assigning the currently measured vehicle data to a road surface cluster representing the road surface or creating a new road surface cluster if the road surface is previously unknown and assigning the currently measured vehicle data to the newly created road surface cluster and storing the second data set in the storage unit (2). Vehicle system (1) according to one of the preceding claims 7 or 8, characterized in that the comparison unit (4) is configured to determine an actual state by comparing the currently measured vehicle data with the first data set, and to determine a target state by comparing the currently measured vehicle data with the second data set, and furthermore to determine the current wear of the chassis components by means of a target-actual comparison. Vehicle system (1) according to one of the preceding claims 7 to 9, characterized in that the sensor system (3) is configured to continuously or adaptively measure the current vehicle data. Vehicle system (1) according to one of the preceding claims 7 to 10, characterized in that the currently measured vehicle data includes the kilometers driven and / or the age of the vehicle. Vehicle with a vehicle system (1) according to any one of the preceding claims 7 to 11 .

Images