Method for estimating tire wear on a driving route

Abstract

The invention relates to a method for estimating tire wear on a driving route, comprising the following method steps: b) Segmenting a digital driving route into a plurality of driving sections, comprising straight-ahead driving sections and curve driving sections, wherein each driving section is assigned route information about the road surface, c) Generating an extended data set from the segmented digital driving route, wherein each driving section is assigned a load profile, d) Estimating the tire wear on the driving route with an electronic data processing device starting from the extended data set, using a machine learning-based straight-ahead driving module and a machine learning-based curve driving module, wherein the electronic data processing device is configured toto input the extended data set into the straight-ahead driving module and the cornering driving module, and to estimate the tire wear on the future route or the previously driven route using the straight-ahead driving module and the cornering driving module.

Description

The invention relates to a method for estimating tire wear on a travel route, to a method for planning a travel route based thereon, and to a method for predicting the profile state of a vehicle tire based thereon, and to a computer program product for use in these methods.There is a continuing interest in the field of tire technology in optimizing the operation of vehicles and the vehicle tires used on them from ecological and economic points of view. A key factor concerning both economic and ecological aspects is tire attrition. The tire abrasion generated by the vehicle tires in use is one of the largest sources of particulate matter in road traffic, and it is believed that this problem will increase even more in the course of increasing electrification of the vehicles, since they often generate higher abrasion. The relevance of the tire abrasion to the fine dust is so pronounced that legal measures are taken in many countries in order to reduce the fine dust as far as possible, as a result of which the requirements for the durability of vehicle tires increase.However, severe tire abrasion is also disadvantageous from an economic point of view, since it leads to the vehicle tires having to be replaced more quickly in order to prevent a disadvantageous effect on driving safety from arising, for example because there is not sufficient tread depth. Accordingly, operators of commercial vehicle fleets in particular have great interest in minimizing tire abrasion in order to keep maintenance costs as low as possible.Numerous approaches are known in the prior art for evaluating the state of the pneumatic vehicle tires used on a vehicle and thus the already carried out abrasion, these approaches ranging from the classic manual measurement of the profile depth to complex, image-based measurement methods.Furthermore, various methods for the indirect estimation and / or calculation of the profile depth are known from the prior art, which methods however have different advantages and disadvantages. Related art is disclosed, for example, in CN 112976956 A, EP 0972658 B1, U.S. Pat. No. 9340211 B1, U.S. Pat. No. 2021 / 0302272 A1 and DE 102018200358 A1.For example, methods are known which allow conclusions to be drawn about the profile depth of the tread of the vehicle tire by measuring the radial acceleration of a vehicle tire with the aid of an acceleration sensor on the inner liner of this vehicle tire. Other methods have been based on the correlation of the vehicle speed, determined for example by means of GPS data, with the rotational speed of the vehicle tire, wherein the profile depth of the vehicle tire is derived from the initial profile depth and the change in the dynamic rolling radius, which can be obtained from the correlation of the vehicle speed with the rotational speed of the vehicle tire. However, these methods usually require an additional sensor system in the vehicle tires or in their periphery. However, this sensor system not only increases the production costs of the vehicle tires, but a sensor system installed in the vehicle tire is also susceptible in many cases to faults which can be caused by the strong mechanical loads occurring during the driving operation. For this reason, many of the vehicle tires which are in use today do not have further sensor systems which would make it possible to use corresponding methods.Against this background, data-based methods for estimating the profile depth of vehicle tires have been proposed, in which the profile depth of vehicle tires is calculated on the basis of vehicle parameters, in particular the vehicle speed or the acceleration values of the vehicle, and a set of different influencing parameters, for example on the nature of the vehicle, the type of tires or the prevailing environmental conditions, wherein the number of influencing parameters entering into the calculation can vary with the respective application scenario and the desired precision of the calculation.Corresponding approaches for estimating the remaining profile depth or the state of the vehicle tires are perceived as advantageous with regard to numerous aspects, in particular since these become increasingly important with the increasing digitalization in the truck fleet store and with the increasing size of truck fleets in order to estimate the current state of the tires of the fleet vehicles without a manual inspection of the tires having to be carried out. This estimation of the current profile depth or the current tire condition is thus very important in order to plan maintenance work pending from an economic point of view or any necessary tire replacement in a timely and time-optimized manner.Many of the methods known from the prior art for estimating the tire condition are, however, considered disadvantageous from other points of view. In particular, many of the simpler approaches are perceived as being in need of improvement with regard to the evaluation quality. The more comprehensive parameter-based methods, which sometimes rely on the use of machine learning, moreover, regularly require a great deal of computing capacity and usually very complicated training of the corresponding machine learning-based modules.In addition to the existing deficiencies in the estimation methods of the prior art, it can also be established that the methods known from the prior art are primarily directed to estimating the already occurring tire abrasion and thus essentially make no contribution to a reduction in the tire abrasion. However, it would be desirable to state solutions that the underlying concepts of the state evaluation should also be able to avoid tire abrasion, thereby achieving environmental protection, in particular by reducing the particulate matter load, and additionally achieving an increase in the durability of the vehicle tires, as a result of which the operating costs of a vehicle fleet could potentially be significantly reduced.The primary object of the present invention was to eliminate or at least alleviate the disadvantages of the prior art.In particular, it was the object of the present invention to specify a method for estimating the tire abrasion on a travel route, which method can be used in an advantageous manner both in estimating the tire abrasion and in the context of route planning for optimizing the tire abrasion on the route.In this case, it was an object of the present invention that the method to be specified should be executable as efficiently as possible with resources, wherein it was a desirable requirement that, in particular, training steps which may be required for a module based on machine learning should be simplified compared with the prior art, such that the need for training data can be reduced.In addition, it was an object of the present invention that the method to be specified should be able to be carried out as extensively as possible on the basis of position data of the vehicle, in particular GPS data, which are very easily available for most vehicles today, wherein it was a desirable requirement that advantageous estimates of the tire abrasion should be possible with the method to be addressed even if only a small amount of telemetry data about the vehicle is available.In a manner linked thereto, it was an object of the present invention to provide a powerful method for predicting the profile state of a vehicle tire.In particular, it was additionally an object of the present invention to provide a method for planning a route, which method is linked to the method to be specified and with which method it should advantageously be possible to make a contribution to minimizing the tyre abrasion, wherein it was a desirable proviso that the corresponding method should be executable in such a way that, in addition to the planning of the route, it is also possible to provide driving instructions which make it possible to further minimize the tyre abrasion on the corresponding route.The inventors of the present invention have now found that the objects described above can be achieved if - in particular in the case of a retrospective evaluation of the tire wear that has taken place - a previously traveled travel route or - in particular in the case of route planning - a future route to be traveled, which are provided in the form of digital travel routes, are segmented on the basis of a curvature criterion into straight and curved sections, each of which is assigned a load profile in which the estimation of the tire wear on the corresponding travel route is carried out using two machine learning-based modules, which are trained to estimate the tire wear on straight sections or on curved sections, as defined in the claims.Advantageously, the corresponding method makes it possible, even when recourse is made to only GPS data, to provide an advantageously precise prediction of the abrasion for a route which has already been driven or is to be driven in the future, which prediction can be used in the state evaluation of a vehicle tire or also in the route planning, wherein the preferred separation of the evaluation into two separate machine-learning-based modules which can be trained independently of one another for two special cases makes it possible to improve the prediction quality considerably and / or to reduce the training requirement for the machine-learning-based modules considerably with the same prediction quality, wherein the preferred concept makes it possible to separate the driving route into different types of partial routes and evaluate them using separate machine-learning-based modules makes it possible in particular to reduce the computation capacities required for the evaluation.The above objects are achieved, accordingly, by the subject matter of the invention as defined in the claims. Preferred embodiments according to the invention are evident from the dependent claims and the following explanations.Such embodiments, which are referred to below as preferred, are combined in particularly preferred embodiments with features of other embodiments referred to as preferred. Combinations of two or more of the embodiments referred to below as particularly preferred are thus very particularly preferred. Also preferred are embodiments in which a feature of one embodiment designated as preferred to any extent is combined with one or more further features of other embodiments designated as preferred to any extent. Features of preferred computer program products result from the features of preferred methods.Particularly preferred embodiments of the invention are disclosed in the exemplary embodiments. Particularly preferred embodiments of the invention have correspondingly two or more, preferably three or more, very particularly preferably four or more, of the preferred features of the invention disclosed below, which are also realized in the exemplary embodiments.The invention relates in particular to a method for estimating tire wear on a travel route, comprising the method steps: a) generating a travel route to be traveled in the future in the course of route planning or providing a previously traveled travel route in the form of a digital travel route, b) segmenting the digital travel route into a plurality of travel sections, comprising straight-ahead travel sections and curve travel sections, wherein each travel section is assigned a piece of route information about the route condition, wherein the piece of route information comprises a piece of route curvature information about the travel route curvature in the travel section, wherein the maximum travel route curvature of the straight-ahead travel sections does not meet a predetermined curvature criterion, and wherein the maximum travel route curvature of the curve travel sections meets the predetermined curvature criterion,c) generating an extended data set from the segmented digital driving route, wherein a load profile is assigned to each of the driving sections, wherein the load profilei) for a future travel route to be travel comprises information about the travel conditions to be expected in the respective travel section, wherein the load profile comprises at least one item of information which correlates with the respective travel speed to be expected, orii) for a previously driven travel route, information about the driving conditions experienced in the respective travel section, wherein the load profile comprises at least one item of information which correlates with the respective driven travel speed,d) estimating the tire abrasion on the future travel route or the previously traveled travel route with an electronic data processing device based on the expanded data record, wherein the estimation of the tire abrasion is carried out using a straight-ahead driving module stored on a memory unit of the electronic data processing device and a curve driving module stored on a memory unit of the electronic data processing device based on machine learning, wherein the electronic data processing device is configured to input the extended data set as an input into the straight-ahead driving module and the curve driving module and to estimate the tire abrasion on the driving route to be driven in the future or the previously driven driving route with the straight-ahead driving module and the curve driving module, wherein the straight-ahead driving module is trained to estimate the tire abrasion in the respective straight-ahead driving section from the route information of a straight-ahead driving section and the associated load profile, wherein the training is carried out with a set of first training data which comprises a plurality of first training sets, wherein the first training sets each comprise, for training straight-line driving sections with known route information and known load profile, abrasion information about the tire abrasion associated with the training straight-line driving section, wherein the cornering module is trained to estimate the tire abrasion in the respective cornering sections from the route information of a cornering section and the associated load profile, wherein the training is carried out with a set of training data which comprises a plurality of second training sets, wherein the second training sets each comprise, for training cornering sections with known route information and known load profile, abrasion information about the tire abrasion associated with the training cornering section.The method according to the invention serves to estimate the tire abrasion on a travel route. The skilled person understands that this is the abrasion that a vehicle tire arranged on a vehicle experiences during travel. Because of the great relevance of tire abrasion for the operation of commercial vehicle liquors, it is preferred in this case to implement the method according to the invention, in particular in the area of commercial vehicles. A method according to the invention is thus relevant in practice, wherein the tire abrasion is the tire abrasion of a vehicle tire arranged on a vehicle, preferably a truck or bus tire, particularly preferably a truck tire.In method step a), a digital travel route must first be obtained, which can be used as the basis for the following method. The skilled person understands that a travel route is formed by a stringing of road traffic routes and that, from the fact that the method according to the invention is followed by a method for estimating the tire abrasion, the travel route is that of a land vehicle fitted with tires and not, for example, that of a ship. This is thus a method according to the invention, wherein the travel route is formed by a plurality of road traffic routes, for example roads, routes and places, and / or wherein the travel route is the travel route for a tire-equipped land vehicle, preferably a truck or bus.An example is a method according to the invention, wherein the digital travel route is a characterized travel route in a digital map material.A great particular feature of the method according to the invention is that in method step a) the required digital travel route can be obtained in principle in two different ways, these ways having a decisive influence on the way in which the method according to the invention can be implemented in other methods. The method according to the invention namely has, in addition to a looking-back aspect in which the tire abrasion that has occurred on a route driven in real fashion is to be estimated, in particular also the aspect that the tire abrasion that is to be expected on a route still to be driven in future is to be estimated. Accordingly, in method step a), either a route to be driven in the future can be generated in the course of route planning, as would be displayed, for example, on the display of a navigation device, or a previously driven route can be used as the basis.It can be seen as an advantage of the method according to the invention that the method according to the invention can be used flexibly for these two types of looking back and future assessment. However, according to the inventors' judgment, the respective partial aspects are also advantageous per se, so that it can be targeted for specific applications with regard to the required computing capacity and the complexity of the corresponding computer program product to focus on only one of the aspects, for example if the method according to the invention is intended to be integrated into its products by a manufacturer of navigation devices, and an estimation of the previous tire wear on already driven travel routes is less relevant than a consideration of the expected tire wear in the route planning. In some cases, a method according to the invention is thus preferred, wherein method step a) comprises generating a future travel route in the course of route planning in the form of a digital travel route, wherein the load profile in method step c) comprises information about the travel conditions to be expected in the respective travel section for a future travel route, wherein the load profile comprises at least one item of information which correlates with the respective travel speed to be expected. A method according to the invention is particularly preferred, wherein the generation of a future travel route to be traveled is carried out during route planning by a navigation device, preferably on the basis of digital map material. In addition or alternatively, a method according to the invention is preferred, wherein method step a) comprises the provision of a previously driven travel route in the form of a digital travel route, wherein the load profile in method step c) comprises information about the travel conditions experienced in the respective travel section for a previously driven travel route, wherein the load profile comprises at least one item of information which correlates with the respective driven travel speed. In addition or as an alternative, a method according to the invention is particularly preferred, wherein the previously driven travel route is provided by providing time-dependent position data, preferably GPS data, of the vehicle equipped with the vehicle tire, wherein the time-dependent position data are preferably correlated with digital map material.Even if advantages can in principle also be achieved with the method according to the invention on short distances of a few hundreds of meters, the skilled person understands that the influence of correspondingly short distances on the tire abrasion will usually be so low that neither an estimation nor a prediction about the future expected abrasion in the context of route planning would be justified by the effect achieved. Accordingly, the advantages of the method according to the invention are evident, in particular, on long distances. A method according to the invention is accordingly preferred, wherein the digital travel route has a length of 2 km or more, preferably 5 km or more, particularly preferably 10 km or more.In method step b), the digital travel route obtained in method step a) is virtually segmented into a plurality of travel sections. The aim of this step is to segment the travel route into a plurality of travel sections, which can be referred to as either a straight travel section or a curve travel section, so that the entire digital travel route can be composed in the simplest case of a plurality of straight travel sections and curve travel sections. Thus, a method according to the invention is relevant later above all, wherein the digital travel route comprises two or more, preferably five or more, particularly preferably ten or more, very particularly preferably twenty or more, straight-ahead travel sections, and / or wherein the digital travel route comprises two or more, preferably five or more, particularly preferably ten or more, very particularly preferably twenty or more, curve travel sections.The method according to the invention is based on the fact that there are at least straight-ahead travel sections and curve sections which are assigned to the digital travel route during the segmentation. According to the inventors' judgment, it is the goal to undertake the segmentation of the travel route in such a way that the total of these is composed as far as possible of straight-ahead travel sections and cornering sections. At the same time, it is to be noted that when land vehicles are travelling, there may certainly be travel sections which, in particular in predictive route planning, cannot be expediently assigned to the straight-ahead travel sections or the curve travel sections, for example journeys in parking spaces or away from predefined paths, or generally for sections to which there is no suitable map material, for example private terrain. In addition, there may be driving sections in which the vehicle itself is conveyed, for example because it is section-loaded onto a train or transported by means of a conveyor over a water route. In principle, it would be possible for this purpose, after the inventors have been assessed, to train further modules based on machine learning, which contribute to the estimation of the tire wear at least for some of these driving sections. However, this is not preferred according to the inventors' judgment in view of the required computational effort and training of the respective machine learning-based modules. This applies in particular since many of these further travel sections, with regard to practical relevance, will regularly only be associated with very little tyre abrasion, in particular during travel or similar transport. Furthermore, other of these further driving sections are very rare to predict after the inventors' judgments and / or very difficult to predict with regard to the actual tire abrasion, in particular "off-road driving sections", which are virtually not provided especially for the relevant commercial vehicles. A method according to the invention is thus preferred, wherein the digital travel route consists to a proportion of 70% or more, preferably 80% or more, particularly preferably 90% or more, very particularly preferably 95% or more, particularly preferably 99% or more, most particularly preferably to substantially 100%, of straight-ahead travel sections and curve travel sections, based on the total length of the digital travel route. Additionally or alternatively, a method according to the invention is conceivable, wherein the digital travel route comprises one or more further travel sections, wherein the further travel sections are selected from the group consisting of parking space travel sections, off-road travel sections and transport sections, for example train transport sections or conveyor transport sections.For segmenting the travel route into the individual travel sections and assigning the individual travel sections to travel straight ahead or travel around a curve, there are in principle many possibilities, of which the skilled person resorts to the one which is most expedient for the application requirements sought by him.To understand the concept, it is the goal to first address the question of when a driving section is considered to be a straight-ahead driving section and when it is considered to be a cornering section. According to the invention, this is established on whether the maximum driving route curvature of the respective driving section is below or above a predetermined curvature criterion. In other words, this means that the assignment takes place depending on how strongly the driving route curvature is in the driving section.It can be seen as an advantage of the method according to the invention that the parameters used for evaluating the route curvature can be selected relatively freely by the person skilled in the art, since in practice it is possible without any restrictions for the person skilled in the art to establish a criterion suitable for him for the curvature of the route. In the development of the invention, it has proven advantageous to set the driving direction vector of the vehicle and to observe its development when driving through the respective driving section, so that the driving route curvature can be described by the change in orientation of the driving direction vector experienced in the driving section or a part thereof, which is advantageously relatively close to the evaluation scale which a human observer would also use when observing a driving route for the driving route curvature. In order to reduce the computing effort and not have to calculate the orientation of the driving direction vector for every non-substantially small part of the driving sections, the inventors consider it to be particularly advantageous to approximate the change in orientation of the driving direction vector by considering the change in orientation between discrete points of the driving route. A method according to the invention is thus preferred, wherein the driving route curvature is the change in orientation of the driving direction vector experienced in a driving plug part. Particularly preferred is additionally or alternatively a method according to the invention, wherein the driving route curvature is obtained by quantifying the change in orientation of the driving direction vector at points of the driving section spaced apart from one another along the driving direction, wherein the points spaced apart along the driving direction are preferably arranged equidistantly one behind the other along the driving direction.Since it has been described above how the allocation of the driving sections takes place on the basis of their driving route curvature, it is expedient to deal with the segmentation. This segmentation can in principle be carried out in any manner which the person skilled in the art would find expedient for his application. According to the inventors' judgment, however, there are above all two particularly targeted variants with which the computing effort can be reduced, namely either in the step of segmenting or in the downstream estimation of the tire wear.One concept with which, in particular, the segmentation step can be carried out in a very resource-saving manner is to choose the length of the driving sections, in particular of the straight-ahead driving sections, to be as equal as possible. In practice, the inventors recommend that the length of the driving sections be selected dynamically at least to such an extent that curves in their entirety are detected in one driving section. For this reason, it is also not preferred to choose driving sections of exactly the same length, since this results in the risk that too much "waste" would be produced. A method according to the invention is accordingly preferred, wherein the length of all travel sections differs from one another by 20% or less, preferably 10% or less, particularly preferably 5% or less, very particularly preferably substantially not at all. In addition or alternatively, a method according to the invention is preferred, wherein the length of the straight-ahead driving sections differs from one another by 20% or less, preferably 10% or less, particularly preferably 5% or less, very particularly preferably substantially not at all, and / or wherein the length of the cornering sections differs from one another by 20% or less, preferably 10% or less, particularly preferably 5% or less, very particularly preferably substantially not at all. By way of example, a method according to the invention is provided, wherein the length of the straight-ahead driving sections and / or of the cornering sections, preferably of the straight-ahead driving sections and of the cornering sections, is in the range from 20 to 200 m, preferably in the range from 50 to 150 m.An alternative embodiment to this, which, although requiring somewhat more computing effort during segmenting, can make the subsequent estimation more efficient, is to dynamically choose the length of the straight-ahead driving sections such that the curvature criterion is not reached at the moment. In other words, this means that long straight routes, for example routes on freeways, are assumed as individual straight-ahead driving sections as far as possible. Thus, as an alternative, a method according to the invention is preferred, wherein the segmenting is carried out in such a way that the length of the straight-ahead driving sections is maximized without the curvature criterion being fulfilled.In addition or as an alternative to the variants described above, it is also possible, in particular in city traffic, to additionally align the segmentation also with traffic measures, so that individual driving sections are bounded, for example by intersections, traffic signals, railroad crossings, zebra stripes or the like, wherein corresponding factors, as further disclosed below, can also be taken into account efficiently within the scope of the route information after the inventors have been assessed.According to the invention, route information is also assigned to each driving section. The term "route information" here broadly first denotes those parameters and information which are suitable for describing the properties of the travel section, wherein the person skilled in the art charged with the implementation of the method according to the invention adapts the route information which he assigns to the travel sections to the respective application requirements of his project just as to the actually available route information. In the simplest embodiments of the method according to the invention, the route curvature can be assumed as the only piece of route information, which, however, results, after the inventors have been assessed, in a method which is more likely to be produced by its resource efficiency than by its prognosis quality.According to the inventors' judgment, it is expedient to also specify at least the length of the respective travel section in the route information. A method according to the invention is thus preferred, wherein the route information comprises route length information about the length of the route in the travel section.On the basis of this, the estimation quality can be improved to a greater extent, the more route information is present in the respective driving sections. This includes in particular information about the type of roadway and the type of road, since the wear to be expected, for example, between freeways and on roads differs the same as that on different road coverings. A method according to the invention is thus preferred, wherein the route information comprises a condition information about the condition of the roadway in the driving section, preferably about the roadway state and / or the roadway subsurface. In addition or alternatively, a method according to the invention is preferred, wherein the route information comprises road type information about the road type of the roadway in the driving section, for example freeway, country road or city traffic.In particular for optimizing the estimation in city traffic, it is also advantageous to provide information about factors that may influence traffic in the driving section, which may be easily derived from digital map material in an advantageous manner and which represent predictable restrictions for unrestricted free driving through the driving section and, on the contrary, cause braking and acceleration processes that may be associated with increased tire wear. For this purpose, a method according to the invention is preferred, wherein the route information comprises traffic information about the number, preferably the number and arrangement, of traffic-influencing factors in the travel section, wherein the traffic-influencing factors are preferably selected from the group consisting of intersections with progress along the travel route, intersections without progress along the travel route, intersections with traffic signals, pedestrian crossings with traffic signals, zebra strips, restricted railroad crossings, unrestricted railroad crossings and circular traffic intersections.In method step c), the digital travel route segmented as explained above is now supplemented by further data, which is referred to within the scope of the present invention as the generation of an extended data set. While the route information described above is static information which is inherent to the route in the travel section and will not change or will only change insignificantly on the time scales relevant to the estimation process, the additional data which are assigned to the digital travel route are those which can dynamically change between multiple executions of the method, for example because they were detected on a drive-specific basis for the routes actually driven in the past, or because they are adjusted for future travel routes to be planned depending on expected influencing factors. Within the scope of the present invention, this additional data is referred to as a "load profile" which is assigned to the respective travel sections. This load profile correlates with which load the vehicle has experienced in this driving section in the previously real driving route, or with the load predicted for future travel by the driving section.In principle, within the scope of the invention, the person skilled in the art can make relatively free decisions as to which influencing factors he keeps sufficiently relevant, that he assigns them to the load profile of the respective driving sections, wherein it is also possible in particular for individual factors to be assigned only to individual driving sections as part of the load profile, because these are less relevant for other driving sections. However, the inventors of the present invention have recognized that it is indispensable for a reasonable estimation of the tire wear to include at least one piece of information in the load profile, which information correlates with the travel speed that was traveled on this travel section in the past or the travel speed that is to be expected in the future. This is due to the fact that the tire abrasion experienced in a driving section depends quite substantially on the driving speed in this section. In principle, it is possible to establish the driving speed information on various parameters, for example the maximum speed, wherein it is very particularly preferred if the corresponding information comprises at least one information about the average speed. According to the skilled person's understanding, this information for a route actually traveled in the past for which the tire abrasion is to be evaluated can be extracted without any necessity from the travel history, in particular from the GPS data collected during the travel. For the future prediction of the load profile in the context of route planning, this value must be subtracted or estimated, as defined above. The inventors propose that information about the permissible maximum speed on the route and / or estimates about the maximum and / or average speeds realizable on the route can be used in an advantageous manner for the prediction at the location of the real speed driven. For example, for the route planning of trucks on most freeways, the average speed to be expected can be estimated quite reliably, since this information is available as comprehensive empirical values, especially in the case of relatively large fleets of vehicles. A method according to the invention is thus preferred, wherein the load profile comprises one or more values which correlate with the respectively expected travel speed and are selected from the group consisting of the realizable maximum speed, the specified directional speed and the expected average speed in the travel section, preferably the expected average speed, and / or wherein the load profile comprises one or more values which correlate with the respectively driven travel speed and are selected from the group consisting of the driven maximum speed, the driven speed at the beginning of the travel section, the driven speed at the end of the travel section and the driven average speed in the travel section, preferably the driven average speed.In addition, the inventors have succeeded in identifying further contributions to the respective load profile which have a noticeable influence on the tyre abrasion and are also in most cases relatively easily accessible in order to be able to be included in the evaluation in the method according to the invention. Here, too, the further influencing factors are aligned either with the actual circumstances of the previously carried out journey or with the frame parameters to be expected for the journey, it being possible during the forecast to resort not only to the parameters of the route planning from the fleet management system, but advantageously also to current weather data, which can be retrieved, for example, on the basis of the GPS coordinates of the planned route. Preference is accordingly given to a method according to the invention, wherein the load profile comprises information about one or more, preferably two or more, particularly preferably three or more, very particularly preferably all, further influencing factors, wherein the further influencing factors are selected from the group consisting of:the expected vehicle type of the vehicle on which the vehicle tire is arranged,the tire pressure to be expected of the vehicle tire,the axle load to be expected acting on the vehicle tire,the weather conditions to be expected in the driving section, andthe ambient temperatures to be expected in the driving section.In this respect, a method according to the invention is additionally or alternatively preferred, wherein the load profile comprises information on one or more, preferably two or more, particularly preferably three or more, very particularly preferably all, further influencing factors, wherein the further influencing factors are selected from the group consisting of:the vehicle type of the vehicle on which the vehicle tire was arranged,the tire pressure of the vehicle tire set during travel,the axle load acting on the vehicle tire during travel,the weather conditions experienced in the driving section, andthe ambient temperatures experienced in the driving section.In method step d), the tire abrasion is now estimated on the corresponding travel route, which is carried out according to the invention using an electronic data processing device which forms the basis for this purpose of the expanded data record. In principle, it is possible here for this processing to take place decentrally in each vehicle, for example via the onboard computer or the navigation device of the vehicle or even via a mobile terminal of the vehicle driver. This embodiment is suitable, according to the inventors' judgment, in particular for implementing the method according to the invention during route planning. In contrast, the inventors find it expedient for the implementation of the estimation of the tire abrasion to driven routes if the required calculations are carried out as extensively as possible in the "backend" of a fleet management system in order to minimize the vehicle-side computing requirement. In some cases, a method according to the invention is thus preferred, wherein the electronic data processing device is a navigation device or a mobile terminal, preferably a mobile telephone. However, a method according to the invention is usually preferred, wherein the electronic data processing device is a central electronic data processing device, preferably a server or a cloud, wherein the central electronic data processing device is particularly preferably part of a fleet management system.The estimation of the tire abrasion is carried out by the electronic data processing device using a straight-ahead module based on machine learning and a cornering module based on machine learning. At least in principle, it is conceivable that the straight-ahead driving module and the cornering module are components of a larger estimation module which has been trained both as a straight-ahead driving module and as a cornering module, as was explained above. However, as contemplated by the inventors, it is preferred for substantially all embodiments if the straight ahead driving module and the cornering module are separately trained modules. A method according to the invention is thus conceivable, wherein the straight-ahead driving module and the cornering module are formed by the same estimation module based on machine learning. However, a method according to the invention is preferred, wherein the straight-ahead driving module and the cornering module are machine learning-based modules trained separately from one another.The concept of machine learning per se is well known to the person skilled in the art. Suitable software solutions which, in view of the present disclosure, can be adapted to the requirements of the method according to the invention by the training as described above are currently commercially available from numerous manufacturers or can be programmed as required by specialized service providers. a method according to the invention is particularly relevant, wherein the training of the straight-ahead driving module and of the cornering module takes place by means of monitored learning. By way of example, a method according to the invention is provided, wherein the straight-ahead driving module and / or the cornering module is based on a machine learning algorithm which is selected from the group consisting of logistic regression, support vector machines, K-nearest neighbor methods, decision tree methods and artificial neural networks, preferably artificial neural networks.The corresponding training can be carried out with training sets which each comprise, for training straight-ahead driving sections or training curve driving sections with known route information and known load profile, information regarding which abrasion is to be expected in a corresponding driving section, which is referred to in the present case as abrasion information. This training enables the straight-ahead driving module or the cornering module to estimate the tire abrasion to be expected for the digital driving route in the past or in the future, based on real or estimated route information and associated load profiles of driving sections. A method according to the invention is preferred, wherein the abrasion information correlates with the amount of gum abraded in micrograms, wherein the abrasion information is preferably the amount of gum abraded in micrograms.The skilled person is without any compulsive to understand that the training of the machine learning-based modules for increasing the estimation quality or for reducing the training effort should be carried out as extensively as possible with such route information or load profiles, which in the later method according to the invention are also given as input into the corresponding machine learning-based modules. It is indeed possible to additionally provide further information for the training, for example in order to provide additional functionality for other application cases, or to set information in part relating to type, which information can be correlated with the route information and / or the load profiles. However, this is much less preferred in view of the required training effort and the estimation quality to be expected. A method according to the invention is thus preferred, wherein the aspects of the route condition characterized in the route information, which are assigned to the driving sections in method step b), correlate at least partially, preferably predominantly, particularly preferably substantially completely with the aspects of the route condition characterized in the first training sets or second training sets by the route information, and / or wherein the route information characterizes the same route condition as the route information assigned in method step b) for at least a part of the first training sets or second training sets, preferably the predominant part of the first training sets or second training sets, particularly preferably for substantially all of the first training sets or second training sets, at least partially, preferably predominantly, particularly preferably substantially completely.In addition or alternatively, a method according to the invention is preferred, wherein the driving conditions characterized in the load profile of the driving conditions assigned to the driving sections in method step c) correlate at least partially, preferably predominantly, particularly preferably substantially completely with the driving conditions characterized in the first training sets or second training sets by the load profile, and / or wherein the load profile characterizes the same driving conditions as the load profile assigned in method step c) for at least a part of the first training sets or second training sets, preferably the predominant part of the first training sets or second training sets, particularly preferably for substantially all of the first training sets or second training sets, at least partially, preferably predominantly, particularly preferably substantially completely.However, for the present invention, the inventors advantageously consider the collection of real training data as not necessarily required and also less preferred. In fact, the inventors propose that the training data can be obtained in an efficient manner by simulation, which is made possible in particular by the fact that comprehensive simulation programs are available in the field of vehicle technology, with which the mechanical load of vehicle tires can be reliably simulated during straight-ahead driving, in particular also acceleration and / or braking maneuvers and also in cornering, as a result of which it is advantageously possible to compile the training sets required for the training in a relatively short time. For example, for the technical realization, numerous journeys can be simulated by a large number of curve types with different speeds, different maximum acceleration values (a x, a y) for different road surfaces and loads using a vehicle simulation tool, it being possible in particular to use curves from real-route routes, it also being possible for this to be combined with additional, synthetic curves in order to achieve the greatest possible variety of curves. A method according to the invention is thus particularly preferred, wherein the first training sets and / or the second training sets are obtained by simulation.For example, the training data of the training sets can be generated by driving simulator data on a plurality of routes, which can be freely routed, for example, and which contain all the manufacturers of interest to the person skilled in the art. Commercially available simulation tools such as, for example, the software "TruckMaker" from IPG Automotive can be used for this purpose. Simulation tools of this type regularly simulate all parameters (e.g. the forces on the wheel or slip) that the person skilled in the art requires to determine the abrasion in order to be able to implement the training of the modules. In the case of curve travel sections, in addition or as an alternative to real-route routes, however, curve types or curvatures can also be artificially generated and simulated in order to be able to include a large number of possible curve embryos in the training.The inventors have recognized that the tire load in the various axle and tire positions in the individual driving maneuvers may be very different.Accordingly, it is particularly preferred if the method according to the invention estimates the wear on a tire-specific basis, i.e. for the different tire positions of the vehicle, wherein the method according to the invention can be advantageously designed to the effect that an averaged tire wear is output.The result of the method according to the invention, i.e. the estimation of the tire abrasion to be expected on the route lying behind or imminent, can be output in particularly simple embodiments of the method according to the invention, for example directly via a display of the onboard computer. In this case, this is a method according to the invention, additionally comprising the method step:e) outputting the estimated abrasion via an output device, preferably via a display.Although it is theoretically conceivable to output the result of the method according to the invention for estimating the tire abrasion as such, the inventors consider it preferable for substantially all embodiments to use the result of the advantageous method according to the invention in one of two subsequent methods.The first variant of these subsequent methods is an advantageous method for predicting the profile state of a vehicle tire. Since the method according to the invention for estimating the tire abrasion comprises information about the estimated tire abrasion, it is advantageously only necessary to provide information about the initial profile state of the vehicle tire in order to predict the profile state, so that the desired information about the profile state obtained at the end of the travel route can be obtained from the information about the initial state and the expected change.The invention thus also relates to a method for predicting the profile state of a vehicle tire, comprising the method steps of the method according to the invention for estimating the tire abrasion on a travel route, comprising the steps: a) determining an initial profile state information item of a vehicle tire before the travel of a travel route, b) estimating the tire abrasion item experienced on the travel route with the method according to the invention, c) predicting the profile state of the vehicle tire while adapting the initial profile state information item while taking into account the estimated tire abrasion item on the travel route.A method according to the invention is preferred, wherein the initial profile state information comprises and the profile depth. In addition or alternatively, a method according to the invention is preferred, wherein the initial profile state information is determined by measuring the vehicle tire, or wherein the initial profile state information is determined by estimating the profile state of the vehicle tire after passing through previous journeys using a data-based estimation method, preferably using the method according to the invention for predicting the profile state of a vehicle tire.A particularly powerful method which builds on the method according to the invention for estimating the tire abrasion is also the method for route planning a travel route. The concept of routing is well known to those skilled in the art. This is based-somewhat shortened-on the fact that possible travel routes between a destination and an end point are calculated and evaluated on the basis of different evaluation criteria in order to select an optimum route which meets the respective requirements, for example for minimizing the travel route, for minimizing the travel time or for minimizing travel on on on country roads. The method according to the invention for estimating the abrasion, which can be used in an advantageous manner to evaluate the expected abrasion on a future route, can now be used in an advantageous manner to integrate the expected tire abrasion into such a route planning as an additional evaluation factor.The invention accordingly also relates to a method for planning a route, comprising the method steps of the method according to the invention for estimating the tire abrasion on a route, wherein two or more route options are generated during a route planning, wherein the tire abrasion on the generated route options is estimated, wherein the selection of the route from the two or more route options is carried out at least partially as a function of the estimated tire abrasion on the route options.A method according to the invention is preferred, wherein a plurality of travel route options are generated during route planning.In turn, a method according to the invention is preferred additionally or alternatively, wherein the selection of the travel route from the travel route options is effected on the travel route with the proviso of a minimum tire abrasion.Preferably, a method according to the invention is also additionally or alternatively, wherein the selection of the driving route from the driving route options is additionally also carried out as a function of one or more further influencing factors, wherein the further influencing factors are preferably selected from the group consisting of the expected driving time and the expected fuel consumption, in particular the expected driving time, wherein the estimated tire abrasion and the further influencing factors are preferably weighted relative to one another.In addition or alternatively, a method according to the invention is also preferred, wherein the electronic data processing device is configured to output the selected travel route, preferably together with an indication about the estimated tire wear, preferably via a display unit, particularly preferably a display.Depending on the consideration of the tire abrasion during the route planning, which is made possible by the present invention, the inventors have recognized that, from the advantageous structure of the method according to the invention, which allows the consideration of dynamic load profiles, it is possible as an advantageous further development not only to provide a route optimized at least partially with respect to the tire abrasion, but also to provide driving instructions for this future route, with which the tire abrasion on the identified route can be further minimized by estimating the tire abrasion for different load profiles in order to propose an optimum load profile, which comprises in particular a proposal for the driving behavior in the respective driving sections, in particular with respect to the maximum and average speed, in particular in curve paths. These proposals for the driving behavior and thus the selection of the load profile obtained in this way can be output directly to the vehicle driver, for example, via the display of a navigation device, but can also be fed as control parameters into a (partially) automatic vehicle controller, for example a driving assistant, in order to configure (partially) autonomous driving on the driving route in a wear-optimized manner, for example.Against this background, a method according to the invention is preferred, wherein the electronic data processing device is configured to carry out the method for the same future travel route twice or more, preferably three times or more, particularly preferably many times, using different possible load profiles and to identify an optimized load profile for the travel on the travel route. Particularly preferred is additionally or alternatively a method according to the invention, wherein the electronic data processing device is configured to identify the optimized load profile on the condition that the ratio of the passage speed through the travel section and the estimated tire abrasion is optimized, wherein a minimum value is preferably predefined for the passage speed. In addition or as an alternative, a method according to the invention is particularly preferred, wherein the electronic data processing device is configured to identify the optimized load profile on the condition that the ratio of the passing speed through the driving section and the estimated tire wear of each vehicle tire of the vehicle is optimized in the best possible manner.A method according to the invention is very particularly preferred, wherein the electronic data processing device is configured to derive a driving instruction for the travel on the travel route from the optimized load profile.In particular, a method according to the invention is preferred, wherein the electronic data processing device is configured to output the driving instruction derived from the optimized load profile to a vehicle driver, preferably via a display unit, particularly preferably a display. In particular, a method according to the invention is also preferred additionally or alternatively, wherein the electronic data processing device is configured to output the driving instruction derived from the optimized load profile as control information to an at least partially automated control system of the vehicle, for example an autopilot or a driver assistance system, so that the control of the vehicle on the driving route takes place at least partially as a function of the control information.Also disclosed is a computer program product comprising instructions which, when the program is executed by an electronic data processing device, cause the latter to execute method step d) of the method according to the invention, wherein the computer program product comprises the machine learning-based straight-ahead driving module and the machine learning-based cornering module.In this respect, a disclosed computer program product is preferred, comprising instructions which, when the program is executed by an electronic data processing device, cause the latter to execute method steps c) and d), preferably method steps b), c) and d), of the method according to the invention, wherein the computer program product particularly preferably comprises instructions which, when the program is executed by an electronic data processing device, cause the latter to cause the method according to the invention to be carried out while generating a driving route to be driven in the future in the course of a route planning.The invention and preferred embodiments of the invention are explained and described in more detail below with reference to the attached figures. The following are shown: FIG. 1 shows a schematic illustration of the steps of the method according to the invention for predicting the profile state of a vehicle tire in a preferred embodiment; and FIG. 2 shows a schematic illustration of the steps of the method according to the invention for route planning a travel route in a preferred embodiment.FIG. 1 schematically visualizes the sequence of the method steps of the method according to the invention for predicting the profile state of a vehicle tire in a preferred embodiment.In the method step marked 500, for this purpose, first an initial profile state information item of the vehicle tire is determined, which characterizes the state of the vehicle tire before the driving of the travel route, wherein in the example shown this is the remaining profile depth, which was previously determined manually on the vehicle tire. As part of method step y marked 600, the estimation of the tire abrasion on the travel route driven by the vehicle is now carried out using the method according to the invention.In method step a) denoted by 100, the travel route actually traveled by the vehicle is provided as a digital travel route in the form of digital map material and is then virtually segmented into a plurality of travel sections in method step b) denoted by 200, wherein, in the example shown in FIG. 1, segmentation is carried out in such a way that the length of the straight-ahead travel sections is maximized as far as possible and curve travel sections are dimensioned with the shortest length which still completely encompasses the curve.The digital travel route can be advantageously obtained from the GPS data. Route information is assigned to each driving section of the digital driving route during segmentation, wherein this information in the example shown comprises, in addition to the information about the driving route curvature in the respective driving section, also the information about the length of the driving section and about the roadway condition and the roadway type, for example for distinguishing freeways and country roads.The data record thus obtained for the previously driven travel route of the vehicle is expanded in method step c, which is marked as 300, to the effect that a load profile is assigned to the travel sections. This load profile can be advantageously obtained from the time-dependent position data, for example GPS data, which were recorded for the vehicle during the past journey. From this, in particular, the average speed in the respective driving sections can be derived as freely as the respective maximum speed in the driving sections and the number or form of acceleration and braking processes. Thus, an extensive and meaningful load profile can be advantageously derived already on the basis of only the easily available GPS data, which load profile can also be advantageously complemented with the respectively prevailing weather conditions.In an advantageous manner, the axle load acting for the vehicle as a result of the loading is also taken into account, wherein this axle load can be advantageously regarded as substantially constant over the respectively considered route in most commercial vehicles if a predefined load is to be transported from A to B. The expanded dataset thus generated is now input into the straight-ahead driving module or cornering module trained as described above in method step d) in order to estimate the tire abrasion on the previously driven travel route using the two modules, wherein in the example shown in FIG. 1 the separate modules were each trained on the basis of simulated training datasets.The information about the tire abrasion thus obtained is combined with the initial profile information in method step z marked as 700) in order to estimate the profile state of the vehicle tire expected at the end of the travel route in the light of the initial state and the estimated change.FIG. 2 visualizes a schematic representation of the steps of the method according to the invention for route planning of a travel route in a preferred embodiment. This is because shown in the example of FIG. 2 in so far that the method steps of the method according to the invention are carried out multiple times, wherein in method step a marked 100), respectively different, generated digital driving routes for possible driving routes are used as starting point, which digital driving routes are evaluated in the subsequent steps of the method according to the invention with regard to the expected tire abrasion.For this purpose, the route information in method step b marked with 200) can be derived analogously to the method described above in FIG. 1 starting from the digital travel routes, wherein it is possible in principle to adapt the dimensioning of the individual travel sections to the effect that the length of straight-ahead travel sections is maximized and the length of curve travel sections is correspondingly minimized.The expanded data set can be generated in method steps c marked as 300) using an estimated load profile, which can be designed in particular taking into account the speed specifications or the applicable guide speeds and the current weather report. The tire abrasion estimated for the different digital driving routes is now used in the method according to the invention for route planning of a driving route as one of a plurality of weighted influencing factors, wherein the other influencing factors represent in particular the entire driving route and in particular the required driving time, wherein the required driving time is weighted as corrective with respect to the tire abrasion in order to prevent the route planning selecting an uneconomically long driving route, in a manner detached from any speed considerations, in order to maximize, for example, trips in 30-order zones.List of reference characters100 Method step a) 200 Method step b) 300 Method step c) 400 Method step d) 500 Method step x) 600 Method step y) 700 Method step z)References included in the specificationThis list of documents cited by the applicant has been produced in an automated manner and is only included for the better information of the reader. The list is not part of the German patent application or utility model application. The DPMA does not take any adhesion for any faults or omissions.Patent Literature citedCN 112976956 A

[0005] EP 0972658 B1

[0005] US 9340211 B1

[0005] US 2021 / 0302272 A1

[0005] DE 102018200358 A1

[0005]

Claims

Method for estimating the tyre abrasion on a travel route, comprising the method steps: a) generating a travel route to be driven in the future in the course of route planning or providing a previously driven travel route in the form of a digital travel route, b) segmenting the digital travel route into a plurality of travel sections comprising straight-ahead travel sections and curve travel sections, wherein each travel section is assigned a piece of route information about the route condition, wherein the piece of route information comprises a piece of route curvature information about the travel route curvature in the travel section, wherein the maximum travel route curvature of the straight-ahead travel sections does not meet a predetermined curvature criterion and wherein the maximum travel route curvature of the curve travel sections meets the predetermined curvature criterion, c) generating an extended data record from the segmented digital travel route, wherein a load profile is assigned to each of the driving sections, wherein the load profile i) comprises information about the driving conditions to be expected in the respective driving section for a driving route to be driven in the future, wherein the load profile comprises at least one item of information which correlates with the driving speed to be expected in each case, or ii) comprises information about the driving conditions experienced in the respective driving section for a previously driven driving route, wherein the load profile comprises at least one item of information which correlates with the driving speed driven in each case, d) estimating the tyre abrasion on the driving route to be driven in the future or the previously driven driving route using an electronic data processing device on the basis of the extended data record, wherein estimating the tyre abrasion using a data record stored on a memory unit of the electronic data processing device, In a further embodiment, the method according to the invention is performed by a machine learning-based straight-ahead driving module and a curve driving module stored on a memory unit of the electronic data processing device, wherein the electronic data processing device is configured to input the extended data set into the straight-ahead driving module and the curve driving module and to estimate the tire abrasion on the driving route to be driven in the future or on the previously driven driving route using the straight-ahead driving module and the curve driving module, wherein the straight-ahead driving module is trained to estimate the tire abrasion in the respective straight-ahead driving section from the route information of a straight-ahead driving section and the associated load profile, wherein the training is performed using a set of first training data which comprises a plurality of first training sets, wherein the first training sets each comprise, for training straight-line driving sections with known route information and known load profile, abrasion information about the tire abrasion associated with the training straight-line driving section, wherein the cornering module is trained to estimate the tire abrasion in the respective cornering sections from the route information of a cornering section and the associated load profile, wherein the training is carried out with a set of training data which comprises a plurality of second training sets, wherein the second training sets each comprise, for training cornering sections with known route information and known load profile, abrasion information about the tire abrasion associated with the training cornering section.Method according to Claim 1, wherein method step a) comprises generating a route to be driven in the future in the course of route planning in the form of a digital route, wherein the load profile in method step c) comprises information about the driving conditions to be expected in the respective driving section for a route to be driven in the future, wherein the load profile comprises at least one item of information which correlates with the respective expected driving speed.Method according to claim 1, wherein method step a) comprises providing a previously driven travel route in the form of a digital travel route, wherein the load profile in method step c) comprises information about the driving conditions experienced in the respective driving section for a previously driven travel route, wherein the load profile comprises at least one information item which correlates with the respective driven travel speed.The method according to any one of claims 1 to 3, wherein the digital travel route is composed of straight traveling sections and cornering sections in a proportion of 70% or more based on the total length of the digital travel route.Method according to one of Claims 1 to 4, wherein the route curvature is the change in orientation of the direction of travel vector experienced in a driving plug part.The method according to any one of claims 1 to 5, wherein the length of all travel sections differs from each other by 20% or less.Method according to one of Claims 1 to 6, wherein the route information comprises route length information about the length of the route in the travel section.Method according to one of Claims 1 to 7, wherein the route information comprises characteristic information about the characteristic of the roadway in the driving section, and / or wherein the route information comprises road type information about the road type of the roadway in the driving section.Method according to one of Claims 1 to 8, wherein the route information comprises traffic information about the number of traffic-influencing factors in the driving section,Method according to one of Claims 1 to 9, wherein the load profile comprises one or more values which correlate with the respectively expected travel speed and are selected from the group consisting of the realizable maximum speed, the specified directional speed and the expected average speed in the travel section, and / or wherein the load profile comprises one or more values which correlate with the respectively driven travel speed and are selected from the group consisting of the driven maximum speed, the driven speed at the start of the travel section, the driven speed at the end of the travel section and the driven average speed in the travel section.Method according to one of Claims 1 to 10, wherein the load profile comprises information relating to one or more further influencing factors, wherein the further influencing factors are selected from the group consisting of: - the vehicle type of the vehicle on which the vehicle tyre is arranged, - the tire pressure of the vehicle tyre to be expected, - the axle load acting on the vehicle tyre to be expected, - the weather conditions to be expected in the driving section, and - the ambient temperatures to be expected in the driving section, or are selected from the group consisting of: - the vehicle type of the vehicle on which the vehicle tyre was arranged, - the tire pressure of the vehicle tyre set during the driving, - the axle load acting on the vehicle tyre during the driving, - the weather conditions experienced in the driving section, and - the ambient temperatures experienced in the driving section.The method of any one of claims 1 to 11, wherein the straight-ahead driving module and the cornering module are separately trained machine learning-based modules.The method according to any one of claims 1 to 12, wherein the first training sets and / or the second training sets are obtained by simulation.Method for predicting the profile state of a vehicle tyre, comprising the method steps of the method for estimating the tyre abrasion on a travel route according to one of Claims 1 to 13, comprising the steps: x) determining an initial profile state information of a vehicle tyre before the travel of a travel route, y) estimating the tyre abrasion experienced on the travel route being traveled with the method according to one of Claims 1 to 13, z) predicting the profile state of the vehicle tyre by adapting the initial profile state information taking into account the estimated tyre abrasion on the travel route being traveled.Method for routing a travel route, comprising the method steps of the method for estimating the tyre abrasion on a travel route according to one of Claims 1 to 13, wherein two or more travel route options are generated during a route planning, wherein the tyre abrasion on the generated travel route options is estimated, wherein the selection of the travel route from the two or more travel route options is effected at least partially on the basis of the estimated tyre abrasion on the travel route options.Method according to claim 15, wherein the electronic data processing device is configured to perform the method twice or more with different possible load profiles for the same future driving route and to identify an optimized load profile for the travel on the driving route.A computer program product comprising instructions which, when the program is executed by an electronic data processing device, cause the computer program product to perform the method step d) of the method of any one of claims 1 to 13, wherein the computer program product comprises the machine learning-based straight-ahead driving module and the machine learning-based cornering module.

Images