Method, system, neural network, and computer program for performing computational tasks on one or more vehicles

Abstract

The invention relates to a method (V, V1, V2), system (S), a neural network (DNN), and a computer program product for executing computational tasks (R) on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14, ...), in particular in one or more electric vehicles and / or at least partially automated vehicles (10, 11, 12, 13). The method comprises: - Determining (S50) one or more parts of one or more vehicle-external computational tasks (R) depending on the output layer of a neural network (DNN1, DNN2, DNNNZ); and (thereafter):- Causing (S60) the execution of one or more parts of one or more vehicle-unrelated computational tasks (R) on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14, ...).

Description

The invention relates to a method, system, a neural network, and a computer program for carrying out computing tasks in one or more vehicles, in particular in one or more electric vehicles.A modern vehicle typically comprises numerous control units, each having different, in some cases very large, resources for different vehicle functions of the vehicle. The individual control units comprise a large number of diverse very valuable hardware, typically processors, on which respective vehicle-specific or vehicle-function-specific programs are executed.On the one hand, the control units are typically associated with very high costs. On the other hand, a vehicle (e.g. a typical customer vehicle) is usually used only relatively little with regard to the portions of time (for the respective purposes or for executing specific functions). For example, the vehicle is actually used for driving on the average of 1 / 48, 1 / 24, 1 / 12, 1 / 6 of the total time (e.g., per day, week, month, year). In addition, vehicles (in the user's hand) are often parked for a longer period of time.The present document is concerned with the technical object of increasing the benefit of hardware, in particular the cost-benefit ratio of hardware, of a vehicle. In particular, the technical task includes enabling improved, as efficient and / or more reliable and / or more real-time processing of a computing task, in particular a computing task external to the vehicle.The object is achieved by each of the independent claims. Advantageous embodiments are described inter alia in the dependent claims. It is pointed out that additional features of a claim dependent on an independent claim can form a separate invention which is independent of the combination of all features of the independent claim without the features of the independent claim or only in combination with a subset of the features of the independent claim and which can be made the subject matter of an independent claim, a divisional application or a subsequent application. This applies in the same way to technical teachings described in the description, which may form an invention independent of the features of the independent claims.According to a first aspect, a method for executing a computing task on a vehicle (in particular also to be understood as being in or within the vehicle) is provided. The method comprises: determining one or more portions of one or more other-vehicle computing tasks depending on the output layer of a neural network, and (then or depending on the result of the determining): causing one or more portions of one or more other-vehicle computing tasks to be executed on one or more vehicles from a number of vehicles to achieve a result of the computing task.According to a first aspect, a method for executing a computing task on a vehicle (in particular also to be understood as being in or within the vehicle) is provided. The method comprises: training, in particular retraining, a neural network to determine one or more parts of one or more vehicle-external computing tasks, wherein the one or more parts of a vehicle-external computing task are determined to execute the computing task on one or more vehicles from a number of vehicles to achieve a result of the computing task.The features of the (first) method according to the first aspect and features of the (second) method according to the second aspect can be combined with one another. In particular, a resulting method comprises the features of the first method and features of the second method. Therefore, the features described below may relate to the first method and / or the second method. The term method used below is to be understood in particular as the first method, as the second method or as an (advantageous) combination of the methods.The (first and / or second) method preferably further comprises: providing a (the) calculation result of the one or more parts of the vehicle-external calculation task by the one or more vehicles (e.g. to a backend) and / or receiving a (the) calculation result outside the one or more vehicles (e.g. from a backend).The new-tonal network described in the present document is preferably a neural network which is set up and / or trained, in particular appropriately trained, for determining one or more parts of the one or more vehicle-external computing tasks (for execution on one or more vehicles).The at least one neural network can be trained, in particular retraining, and / or be operated to have a computing task executed in the form of one or more parts on one or more vehicles, in particular taking into account a target value corresponding to specific or specific features.This can be done to achieve a result of the computing task. The achievement of a result of the computing task can comprise or be a contribution to the (overall) computing task, in particular the result of the computing task (e.g. ascertaining or the data, in particular data representing the result of the computing task).In this case, the input layer can be controlled as a function of the target value, in particular of one or more target values corresponding to the features described in the present document (in particular to be understood as being the data which characterize the target value).For example, the at least one neural network can be trained or trained overnight to determine one or more parts of a vehicle-external computing task (e.g. those which are initiated at specific vehicles or for respective vehicles or per vehicle) in such a way thatexecuting the vehicle-external computing task on one or more vehicles from a number of vehicles; and / oroptimizing a result of executing the vehicle-foreign computing task.The term "optimizing" or "optimum" is to be understood within the scope of the present document as meaning, in particular, an (expedient) tendency of an optimum or an approach to a (theoretical) optimum and / or the holding of an optimized (improved) value, in particular apart from a certain tolerance value or approach. In this case, an optimization (e.g. in comparison with a conventional or rule-based algorithm and / or in addition to or after execution of a rule-based algorithm.The method may further include: receiving a plurality of results of the execution of the one or more parts of the one or more vehicle-external computing tasks (e.g. individually or from a plurality of vehicles) and providing the result (e.g. a total result combined from partial results) for the one or more vehicle-external computing tasks.The one or more vehicle-external computing tasks can be read in, or provided by, a backend (outside the vehicle), in particular. The result of the execution of the one or more vehicle-foreign computing tasks can be provided to a backend, in particular the backend (the same backend or in a backend networked to the backend).For example, the neural network may be trained, in particular trained at night, to select one or more vehicles from a number of vehicles for executing one or more parts of the vehicle-foreign computing task; and / or to assign one or more parts of a vehicle-foreign computing task to one or more vehicles. It can be trained for this process to be optimized according to the determined or determinable target value.For example, a computing task can be distributed among a plurality of vehicles and / or a plurality of vehicle-external sub-tasks (which can be formed from the computing task in the method) can be distributed among a plurality of vehicles.The determination of one or more parts of a vehicle-foreign computing task can comprise or be a selection of one or more parts and / or a determination of the parts of the vehicle-foreign computing task and / or a division of the (one or in each case more) vehicle-foreign computing tasks into sub-tasks (in particular also to be understood as sub-tasks).Alternatively or additionally, the determination of one or more parts of a vehicle-external computing task can be carried out in such a way that (advantageous or relevant for achieving or maintaining a specific target value) distribution of the parts of the computing task to a plurality of vehicles comprises or is.For example, the determination of one or more parts of a vehicle-external computing task can be carried out in such a way that a number N (e.g. more than three two) of the vehicle-external computing tasks is expediently distributed to a number M (e.g. more than 10, 100, 1000) of the (selected) vehicles, in particular a number M' (e.g. more than 20, 200, 2000) of the (selected) computing units of the vehicles (from a number of the vehicles).One or more steps of the method may be performed depending on the one (or more) neural networks. In particular, this can comprise or be: controlling the execution of the respective step of the method as a function of the operation (also: operation) of the one or more neural networks, in particular as a function of the output layer, in particular of the data of the one or more neural networks output or generated at the output layer.The target value can be a predetermined target value that can be set or controlled more specifically (e.g. case-specific) and / or variably and / or, in particular dynamically and / or during the execution of the method, for specific vehicles, users, conditions, execution of the method, in particular one or more operating cycles (e.g. specific operating cycles) of the method. For example, the target value may vary depending on the vehicle-external computing task or may be controlled in the method.For example, the target value (e.g. by one or more users or operators of the method) can be adjusted, in particular controlled, in particular for one or more specific operating cycles, computing tasks and / or dynamically and / or during the execution of the method. This results in several advantages that are easily understood by the skilled person. For example, a plurality of target values (e.g. by the ordering party of one or more respective vehicle-external computing task R and / or by one or more users of the respective vehicles and / or by an operator of the method) can be taken into account. For example, the at least one neural network is trained and / or operated to aim or maintain an optimum between a plurality of target values.Alternatively or additionally, the target value can be adjusted in the method, in particular depending on one or more (detected) features or the type of the computing task. The adaptation of the target value can be carried out according to a specific or predetermined mathematical relationship (e.g. parameter, formula, characteristic curve, coefficients of a predetermined formula) and / or as a function of predetermined criteria.One or more of the steps of the method can be carried out (e.g. with respect to a plurality of vehicles) in such a way that a target value (which applies and / or is common overall and / or applies to one or more computing tasks, users, vehicles) is sought, in particular (with an expedient approximation) is reached and / or held. This can be effected by the corresponding training and / or operation of the at least one neural network.The computing resources of the respective vehicle described herein may relate (individually or according to a predefined formula) to: the CPU load (relating to general purpose unit); and / or GPU load (relating to graphics processing unit); and / or the RAM (working memory); and / or the nonvolatile memory (e.g. flash memory, EEPROM); and / or data transmission, in particular data transmission capacity, data transmission rate, data packet size, priority, response time or a delay in the data transmission.In this case, the execution of one or more steps of the method can take place depending on available resources of the one or more of the aforementioned types and / or differently for different types of resources.In particular, the execution of the computing task on one or more vehicles is also to be understood as: computing tasks which are to be executed by means of the one or more vehicles (at least predominantly by means of means, resources of the vehicles) from a number of vehicles.For example, the at least one neural network is trained, in particular trained at night, to determine the one or more parts of one or more vehicle-external computing tasks in such a way that the execution of the one or more parts of one or more vehicle-external computing tasks and / or the result of the execution of the one or more parts of one or more vehicle-external computing tasks (e.g. according to one or more predefinable target values) is optimized.For example, this takes place in such a way that a specific and / or (respectively determinable, adjustable) and / or automatically adjustable target value is sought, in particular is reached and / or held.Preferably, all or most of the vehicles described in the present document are (in each case) an at least partially electrically driven (e.g. hybrid, partial hybrid, plug-hybrid, fully electric) and / or automatically drivable (e.g. with the degree of automation of more than level 2 or 3) or parkable or maneuverable vehicle. This results in the advantage that particularly high computing resources of the vehicles that can be driven or parked or can be maneuvered in an automated manner can be used and / or the energy of the traction battery of the vehicle and / or the current from the charging station can be used to perform a computing task that is external to the vehicle. A vehicle-external computing task can thus be processed particularly efficiently (and optionally independently of a further power source).The term "neural network" is to be understood in particular as a deep neural network.The one or more neural networks described herein may include:a generative pre-trained transformer (GPT); and / ora generative, in particular statistical model; and / ora statistical model trained on specific vehicles, users of the vehicles, vehicle-external computing tasks; comprise, be constructed on the basis of such a model and / or be controlled by such a model.Alternatively or additionally, the term neural network is to be understood as meaning another form of the artificial intelligence (in particular comprising a neural network or a comparable structure). In particular, training and / or operating the one or more neural networks is to be understood as meaning an analogous application of other types of machine learning.For example, the at least one neural network, the training and / or operating thereof can comprise or be an appropriate application of one or more circumstances listed below or combinations thereof: wholly or partly supervised learning, wholly or partly unsupervised learning, at least partly reinforcement learning, multi-task learning, classification and regression trees, support vector machines, logical and relational learning, probabilistic graphic models, rule learning, instance based learning, latent representation, bio-aspirated approach.Particularly preferably, a fully or partially reinforced learning can be used. For example, the reinforcement may be dependent on a result of a previous (e.g., on one or more previous operating cycles, days, months, etc.) The execution of the vehicle-external computing tasks (e.g. depending on individual and / or statistical and / or cumulative results) and / or the extent to which one or more target values (e.g. described in the present document) are reached or executed, respectively.The neural networks or artificial intelligence described in the present document (which are based on the one or more models, in particular generative pre-training transformers or the functionalities) is preferably a so-called weak and / or (comparatively closely) specialized artificial intelligence.These (or the corresponding models or neural networks can be specialized for the execution of the (single or multiple) functionalities described in the present document, in particular, in the case of specific vehicles, roads, countries, charging stations, conditions, users, etc. that prevail during the use of the vehicles, and / or optimized to a certain extent for the execution of these and / or limited to these (at least predominantly or essentially). As a result, the method can be realized substantially more easily and / or with limited resources (e.g. using the means of the neural network and / or the vehicles) and / or can be better controlled (in practical handling). For example, this can reduce possible risks in connection with the use (e.g. a relatively universal or strong) artificial intelligence and / or make the result of the method more predictable.The term "training" is to be understood in particular as machine learning, non-rule-based learning and / or deep learning or the execution of machine learning, non-rule-based learning and / or deep learning. The training can comprise or be a (at least partially, completely or predominantly) monitored or (at least partially, completely or predominantly) unsupervised training. The training of the neural network is to be understood in particular (also) as setting up the neural network.The training (or setting up) of the at least one neural network can comprise or be an end-to-end learning (also to be considered: end-to-end learning) and / or a reinforcement learning (also to be considered: reinforcement learning or reinforcement learning). In this case, a plurality of, in particular a predominant number, or (substantially) all intermediate steps necessary for the request for one or more targets can be mapped or integrated in the (at least partially trained) neural network.The training of the neural network is to be understood in particular (also) as setting up the neural network.The operation of the neural network is or comprises in particular a control of the neural network, in particular of the input layer (one or more inputs) of the neural network. Furthermore, a change in the operating mode of the at least one neural network can likewise be counted for operating the neural network.In particular, the at least one neural network is trained and / or operated depending on a combination of at least two parameters (corresponding to one another) (e.g. data of two different types or characteristically different size described in the present document). In this case, one or more multidimensional optima (e.g. for executing the tasks other than the vehicle according to two or more than two different target values) can be sought, in particular can be achieved or held with a certain approximation.The training and / or operation of the one or more neural networks can take place according to one or more predetermined targets or target values (described, for example, within the scope of the present document) that can be determined, adjusted and / or dynamically changed (for example, by a user and / or dispatcher) in the method.The term "target value" can also always be understood within the scope of this document to mean a target value range, for example a value range restricted upward and / or downward and / or standing as a function of a further variable (for example a parameter). The target value can also be understood as an attempt to achieve an (expedient) optimum of the respective dimension. Such an optimum is in particular a local or global minimum or maximum of the respective measure (in particular to be understood as a suitable approximation to the minimum or maximum). The achievement of the one or more target values can therefore also be understood as an optimization. A plurality of target values or an overall optimum can be sought, in particular achieved.It can be one or more specific target values, in particular (for example, selectable per computing task and / or time interval), or a target value which takes into account a plurality of the target values described in the present document.The method can be carried out in such a way that the ascertainment of one or more parts of the vehicle-foreign computing task that can be executed on the one or more vehicles; and / or a formation of the plurality of sub-tasks on the basis of the vehicle-foreign computing task, in particular the division of the vehicle-foreign computing task into a plurality of sub-tasks; and / or the initiation of the execution of a vehicle-foreign computing task on one or more vehicles; is carried out as a function of a target value, in particular a dynamically predefinable target value.In this case, the target value can be taken into account during training, in particular retraining, of the at least one neural network and / or during operation, in particular of the at least one neural network. For example, the neural network can be controlled as a function of a specification, in particular a prompt, (e.g. determinable by a user of the method).The training, in particular retraining, of the one or more neural networks can be carried out, in particular in a plurality of phases, over a long period of time or over a plurality of executions or operating cycles of the method. The training of the at least one neural network can take place at least partially, in particular only or predominantly automatically, during the operation of an already at least partially trained neural network or in a separate process. The method can comprise a change, in particular a longer-term, slow update, of the neural network comprising a plurality of the operating cycles, depending on a plurality of the first data sections.The training (in particular also to be understood as the process of training) can comprise or be:determining training data, in particular on the basis of the data from previous executions of vehicle-external computing tasks; and / ordetermining and / or initiating the backpropagation, in particular on the basis of the data from previous executions of vehicle-external computing tasks; and / ora selection or an adaptation of one or more parameters of the training process, in particular on the basis of the data from previous executions of vehicle-foreign computing tasks.One or more aspects of the training described above may be carried out and / or adapted depending on the (e.g. processed and / or filtered) data from previous executions of vehicle-external computing tasks.The term "operating" of the at least one neural network can be understood to mean using the at least one neural network, in particular also as the neural network, as a function of or on the basis of the data (learning data) of the neural network (e.g. the neural network trained according to one or more features described in the present document), in particular in the active operation.The operation of the at least one neural network can alternatively or additionally comprise or be the determination of one or more parts of one or more vehicle-external computing tasks as a function of the output layer of a neural network; and / or the initiation of the execution of one or more parts of one or more vehicle-external computing tasks on one or more vehicles from a number of vehicles by means of the neural network.The at least one neural network can be controlled during its operation (at its input layer) and / or (directly or indirectly, e.g. including a further data processing step) control the determination of one or more parts of one or more vehicle-external computing tasks.Depending on the variant of the method, the method comprises or is, for example, the training (pretraining or retraining) of the neural network. In this case, it is to be understood as "the neural network is trained (pretraining or retraining)" and / or the operation of the already at least partially trained neural network (it may be a second method). In this case, it is to be understood as "the neural network is already trained (e.g. at least to an appropriate degree) (pretraining or retraining respectively night-trained)". The method can therefore comprise or be the first method and / or the second method.For example, the at least one neural network (e.g. to be used during operation) can be understood as features of or data on the basis of the at least one (e.g. at least partially trained) neural network. For example, the data are generated during training of the at least one neural network on any computing unit, in particular the computing unit of the backend, in order to then make them operable on one or more other computing units, in particular in (e.g. respective, specific) vehicles, of a computing unit that is central or networked. The operation of the at least one neural network can comprise or be a setup of the at least one neural network as a function of the data of in each case at least one corresponding neural network or neural network trained for the corresponding use.Each of the information described in the present document may comprise or be a measure of probability (corresponding to the corresponding parameters or with respect thereto). For example, the at least presumably available computing power of one or more vehicles can be a measure of the probability or (e.g. in addition to a quantitative and / or qualitative measure of the computing power) can comprise a measure of the probability.In an example which can be combined with all the features described in the present document, the ascertainment of a plurality of parts of the vehicle-external computing task, in particular of sub-tasks of the at least one computing task, which can be executed in each case on one of a plurality of vehicles, can be carried out as a function of one or more feedbacks of the one or more vehicles with respect to a progress in the execution of one or more parts of the computing task and / or already executed parts of the computing task. For example, the neural network can be operated, in particular controlled, as a function of feedback.The vehicle-external computing task preferably comprises a program code comprising one or more instructions (also to be understood as an executable program, a binary code, etc.) and / or vehicle-external (to be processed) data, in particular data which are to be processed by means of said program code. For example, the vehicle-external computing task can comprise a program code to be transmitted to the vehicle once or comparatively rarely and (e.g. always new) vehicle-external data to be transmitted to the respective vehicle several times or comparatively frequently (e.g. data packets).The vehicle-external computing task is preferably a computing task that requires considerable computing power. For example, the computational task may have a scope that requires processing by a processor (e.g., laptop processor) (typical or commercially available at the time of application of the method described herein) over several minutes, hours, days, or weeks.The vehicle-external computing task, in particular the program code and / or data of the vehicle-external computing task, can be independent of a vehicle function, in particular independent of all essential or any vehicle functions, of the vehicle.For example, the vehicle-external computing task may (also) be independent of (essential or all) system functionalities and / or user functionalities of the vehicle. Alternatively or additionally, the vehicle-external computing task can at least partially comprise program code and / or data which are independent and / or separate from, in particular from, any program code or data of the vehicle.The vehicle-external computing task can be a computing task to be executed temporarily in the vehicle. The system may preferably be configured to perform various computing tasks, in particular any other vehicle-external computing tasks within specific limit values (e.g. with respect to performance).Furthermore, the system may be configured to distribute various computing tasks, in particular within certain limit values (e.g. with respect to performance), virtually any other vehicle-external computing tasks or computing tasks of any types, to one or more computing components.For example, the at least one neural network can be set up and / or trained, in particular trained at night, to distribute the computing tasks to specific (different) computing components of the vehicles and / or to distribute them within the respective vehicles.The vehicle-external computing task serves in particular to achieve a destination which does not relate to the respective vehicle or to further vehicles and / or to functionalities of the one or more vehicles and / or does not relate or does not relate significantly to these. The non-vehicle computing task, the overall computing task and / or at least one sub-task can be determined or configured for a purpose which is completely or substantially independent of the vehicle, further vehicles, data and / or functionalities of the vehicles (at least predominantly or substantially). In this case, there is no need for any change to the vehicle functionality and / or vehicle data.In particular, the result of the vehicle-external computing task is not used in the vehicle, e.g. for no system functionality and / or user functionality, or is not determined for such use. This results in the advantage that vehicles can be used (as required) for diverse (in particular also confidential) vehicle-foreign computing tasks.For example, the system may be configured to prevent or prevent interaction or mixing of the data and / or the executable program code or the instructions of the vehicle-external computing task with a data or programode of its own vehicle, in particular serving for executing a system functionality or a customer functionality.Particularly preferably, the system is configured to transmit (also to be understood: to be retrieved or to be read in by a further computing unit) the result of the computing task foreign to the vehicle, e.g. the resulting data from the vehicle (from the vehicle, e.g. by means of a so-called push and / or a pull method), in particular to the same or a further computing unit.The method or system described in this document (e.g. each of the variants of the system) can thus be configured to use the computing resources of at least one vehicle, in particular a fleet of vehicles (to be understood as also a specific number of vehicles from a fleet of vehicles), particularly efficiently, to carry out (also to be understood as to be able to execute or to solve) an external computing task which is independent of the vehicle or of a plurality of vehicles and is predefined from the outside (after completion of the vehicles) or can be predefined with a high degree of design freedom. In this case, the benefit and / or the ecological balance of the computing resources (in this way or as installed in the vehicle) (in particular of the one or more control units) of the one or more vehicles can be increased in an efficient manner.The method can comprise causing at least a part of the vehicle-external computing task to be executed on one or more vehicles (also to be understood as: by one or more computing components, in particular by one or more control units of the one or more respective vehicles). In this case, an instruction and / or program code and / or data of the vehicle-foreign computing task (alternatively or additionally to be understood as: for processing the vehicle-foreign computing task) can be transmitted and / or sent to the computing component. The computing task may then be executed by the computing component (to determine a result of the computing task). The result of the computing task can be provided to an external computing unit.The method can furthermore comprise receiving the calculation result (at least) of the part of the vehicle-external calculation task from an (in particular from the) vehicle-external calculation unit. For example, the method may include providing the calculation result (the result of the vehicle-external calculation task, e.g. data corresponding to the result) to or to the backend. However, it can also be a computing unit (e.g. with a specific address) outside the one or more vehicles, which are different from the computing unit from which the computing task was initiated. This can also be counted to the backend.The method can furthermore comprise: executing the (at least one, respective) part of the vehicle-external computing task, in particular in addition to a program code of a vehicle-own computing task, at least partially a new program code, in particular for operating and / or optimizing the computing component of the one or more vehicles, is applied to the computing component of the one or more vehicles, in particular in the form of firmware.In this case, the new program code, in particular the new program code for operating and / or optimizing the computing component, can differ from and / or be independent of the program code, in particular from the program code for operating a vehicle function and / or from the program code which can be executed or is executed by the computing component during a driving operation of the vehicle.The method can comprise applying at least partially new program code, in particular in the form of firmware, to the computing component for executing the part of the vehicle-external computing task. In this case, the program code (alternatively or additionally to instructions to be applied to the data) can comprise parts of an operating system for operating the at least one computing component. This can serve for improved, in particular optimized, effectiveness of the processing of the at least one computing task, in particular of a specific (e.g. expected, jointly transmitted) computing task or a computing task of a specific type.Preferably, at least a portion of a vehicle's own program code (e.g. serving for executing a vehicle functionality) and / or data of the computing component can be deleted at least temporarily and / or copied or moved into another memory area inside or outside the computing component.For example, in particular in the event of a relatively long standstill of the vehicle, at least some of the executable program code and / or data which are used for carrying out tasks of the vehicle's own can be removed or swapped out at least from a particularly performance (e.g. fast) memory area (ROM, EEPROM) of the vehicle. For example, parts of the operating system and / or driver (also to be understood as: hardware-based software) of the computing component can be temporarily replaced by the program code of the vehicle-external computing task, in particular by a part of the operating system and / or driver (also to be understood as hardware-based software) which are intended for processing the computing task.Depending on a recognized or predicted need for the use of the vehicle and / or specifically the relevant computing component, the vehicle-own program codes and / or data (in particular program code and / or data that are optimized or specialized for executing the vehicle-own computing tasks) can be written (e.g. flashed) back into the storage area, in particular into the particularly performance, of the computing component of the vehicle. This can also be done remotely and / or remotely, for example.In this case, the new program code and / or data external to the vehicle (serving and / or optimized for executing the at least one computing task external to the vehicle) can typically be independent of the one or more (in particular independent of all) vehicle functions which are executed (also to be understood processed) by the computing component during the driving operation of the vehicle (e.g. by means of particularly performance means of the vehicle). Thus, more efficient and reliable execution of the vehicle-external computing task can be effected.The method can comprise using one or more computing components of one or more vehicles in a first computing mode during a standstill of the (respective) vehicle (possibly completely) for executing the part of the vehicle-external computing task. Alternatively or additionally, the method can comprise using at least a part of the computing component of one or more vehicles (in each case) in a second computing mode during a driving mode of the (respective) vehicle for executing the part of the vehicle-external computing task.The (at least one) neural network can preferably be trained and / or operated to execute a dynamic and / or flexible distribution of the parts of the vehicle-external computing task to vehicles and / or specific computing components of the (respective) vehicles.As a result, (at least cumulatively) a substantially improved use, in particular the degree of use, of the vehicles available for executing at least part of a vehicle-external computing task and / or (respectively available) computing resources of the vehicles can be made possible.The method thus makes it possible to use the available computing resources of one or more vehicles, in particular a fleet of vehicles (e.g. a fleet of customers), in an efficient and reliable manner for tasks other than vehicles. For example, the computing resources of the one or more vehicles (e.g., at least during their service life), during 23 / 24, 22 / 24, 18 / 24 of the time, may be used for (in each case, at least a part of a vehicle-foreign computing task). In this case, the (at least cumulatively considered) cost efficiency and / or ecological balance of the vehicles or of the vehicle fleet can be increased.According to a further embodiment, the method comprises: predicting one or more respective at least presumably available computing resources on a plurality of vehicles from a number of vehicles as a function of the neural network, and, as a function of the result of the prediction: determining one or more parts of one or more other-vehicle computing tasks as a function of the output layer of a neural network; and / or causing a vehicle-other computing task to be executed on one or more vehicles from a number of vehicles.In particular:determining a (current or according to a current state) at least presumably available computing power of one or more vehicles; and / orpredicting the computing power of one or more vehicles that is at least presumably available, e.g. for one or more specific time intervals, can be executed.In this case, the at least one neural network can be trained, in particular retrained and / or trained, depending on the resources available in the past, in particular at a plurality of different time intervals in the past (e.g. specific time intervals), and / or on data which characterise the applying of earlier predictions with respect to the resources, and / or on the data which characterise the result of the execution of one or more parts of the computing task and / or of the entire computing task.This can be effected or have occurred in particular with respect to the (same) number of vehicles, users of the vehicles, conditions of use of the vehicles, routes, etc., or with respect to those which have a certain degree of overlap and / or similarity and / or similar data patterns with the current number of vehicles, users, conditions, routes prevailing during use. This can be effected as a function of the data from one or more in the past executions of the method, in particular operating cycles (cf., for example, the example description relating to step S 10).The prediction of the available computing power can relate to one or more time intervals. For example, a prediction can relate, for example, overall to several seconds, minutes or hours. The time intervals can be small or very small in comparison with the total (prospective) duration of the execution of the computing task or the sub-tasks (e.g. a few milliseconds, a few seconds or a few minutes). Alternatively or additionally, for the prediction according to the invention, it is also possible to end-to-end train and / or operate the at least one neural network, in which the anticipated future change in the respectively available computing power or corresponding predictive values are already taken into account when forming a number of sub-tasks from one computing task, distributing, in particular redistributing, sub-tasks to a plurality of computing units and / or when controlling the execution of the one or more sub-tasks to (respective) computing units. In this case, the prediction may include or be one without explicit output of the predicted parameters.According to a further embodiment, the method comprises: selecting one or more vehicles for executing one or more respective parts of the vehicle-external computing task from the number of vehicles; and / or assigning one or more parts of a vehicle-external computing task to one or more vehicles.Alternatively or additionally, it can be a selection of one or more specific computing units of the one or more vehicles for executing one or more respective parts of the vehicle-foreign computing task and the assignment of the one or more parts of a vehicle-foreign computing task to one or more selected computing units. For example, the vehicles can be selected from a (optionally specific) group of vehicles (which correspond, for example, to specific features. For example, it may be a group of the vehicles whose users have given a corresponding agreement and / or which are set up to participate in the method.According to a further embodiment, training, in particular retraining, the neural network comprises: predicting one or more respective at least presumably available computing resources on a plurality of vehicles from a number of vehicles; and / or selecting one or more vehicles for executing one or more parts of the vehicle-foreign computing task from a number of vehicles; and / or assigning one or more parts of a vehicle-foreign computing task to one or more vehicles.According to a further embodiment, the one or more parts of the vehicle-foreign computing task are determined in such a way; and / or the one or more vehicles for executing one or more parts of the vehicle-foreign computing task are selected from a number of vehicles in such a way; and / or the one or more parts of a vehicle-foreign computing task are assigned to one or more vehicles in such a way that one or more, in particular adjustable, target values relating to the computing task are sought, in particular reached and / or held, when they are executed on the one or more vehicles.In this case, one or more sub-tasks from the plurality of sub-tasks can be (in each case) assigned to one or more specific time intervals in which they are intended to be executed and / or in a sequence, in particular an execution sequence. In this case, the cause that the number of sub-tasks of the plurality of sub-tasks is distributed to a plurality of the vehicles can be carried out in accordance with the time intervals determined or assigned in this way and / or the sequence.The formation of the plurality of sub-tasks on the basis of the vehicle-external computing task, in particular the division of the vehicle-external computing task into a plurality of sub-tasks, can take place as a function of the at least one or more target values (e.g. applicable for the entire computing task).Alternatively or additionally, the formation of the plurality of sub-tasks on the basis of the vehicle-external computing task, in particular the splitting of the vehicle-external computing task into a plurality of sub-tasks, can be dependent on one or more measures of the probability that at least a certain computing power is available in the one or more (respective) vehicles and / or will be available in a future time interval.Preferably, a prediction of one or more items of state information of the one or more vehicles can be carried out and the formation of the plurality of sub-tasks on the basis of the vehicle-external computing task, in particular the splitting of the vehicle-external computing task into a plurality of sub-tasks, can be carried out as a function of the result of the prediction. In this case, (optionally further, additional) optimization can take place, in particular dynamically.In this case, the at least one neural network can be trained, in particular retraining, depending on the data (e.g. from different time intervals or operating cycles of the method from the past), in particular with respect to data which are characteristic of:parameters of the (in the past) parts of the vehicle-external computing task determined (generated) in the method; and / or the vehicles selected from a number of vehicles (in the past in the method);and / or the assignments (executed in the method in the past) of one or more parts of a vehicle-external computing task to one or more vehicles; and / or the reaching and / or holding (e.g. of a measure of the reaching or holding) of one or more target values with respect to the computing task.This can be effected or have occurred in particular with respect to the (same) number of vehicles, users of the vehicles, conditions of use of the vehicles, routes, etc., or those which have a certain degree of overlap and / or similarity with the current number of vehicles, users, conditions which prevail during the use of the vehicles, routes or the corresponding data patterns.According to a further embodiment, the neural network is trained, in particular trained at night, is or is, depending on the use of computing resources and / or on the available computing resources during operation: a specific number or a number of the vehicles corresponding to a specific feature, in particular the one or more vehicles; and / or one or more further vehicles operated under similar conditions as the one or more vehicles; and / or one or more of the same users or users with respectively similar user profiles.According to a further embodiment, the one or more target values relate to a cost required for executing at least a part of the computing task, in particular a number of sub-tasks, in particular: to the power costs for executing the one or more parts of the computing task; and / or to the data transmission costs for transmitting at least a part of the computing task, in particular a number of the plurality of sub-tasks, to one or more vehicles (10, 11, 12, 13); and / or to the data transmission costs for transmitting at least a part of the one or more computing results from one or more vehicles.This is to be understood in particular in such a way that one or more target values relate explicitly and / or implicitly to the (respective costs, in particular those mentioned here) (optionally in each case).For example, a target value may relate to costs per measure of computing power or as a cost of executing the (overall) computing task.According to a further embodiment, the one or more target values relate to the execution of a plurality of, in particular all essential, parts of the computing task, in particular all essential sub-tasks of the one or more vehicle-external computing tasks, on one or more vehicles, in particular of the vehicles selected from a number of vehicles.According to a further embodiment, the method comprises: ascertaining or predicting one or more parameters with respect to one or more vehicles from a number of vehicles, and (thereupon): taking into account the ascertained or predicted parameters in the method, in particular operating the neural network as a function of the parameters. The parameters can be characteristic here for: a (current) connection of the respective vehicle to a charging station; and / or the (respective) state of charge of the battery of the respective vehicle; and / or the energy requirement of the respective vehicle with respect to the use of the vehicle.For example, the method can take account of the fact that, depending on whether the vehicle is connected to a charging station and / or will be connected in a future time interval (e.g. with at least a certain first probability) or will not be connected (e.g. with at least a certain second probability), different costs (e.g. cumulative for the vehicle) arise or are calculated.Alternatively or additionally, the state (e.g. age, charging capacity, operation of the battery between the BoL (=begin of life) or remaining life until the time at which predefined values are undershot by aging, e.g. EoL (= end of life) of the battery of the vehicle, can be determined and taken into account.The (one or more) computing task or sub-tasks can then be distributed over a plurality of vehicles in such a way that, taking into account the respectively expected costs, the most favorable possible processing of the computing task, in particular overall, results.According to a further embodiment, the method comprises: determining a property parameter relating to one or more properties of one or more vehicles from the number of vehicles. The parameters may be characteristic of: one or more computing resources (e.g. computing resources of specific, possibly different types) of the one or more vehicles; and / or a degree of utilization of the computing resources of the one or more vehicles, in particular for a further computing task with respect to one or more specific time intervals; and / or a set of the at least presumably available computing resources of the plurality of vehicles with respect to one or more specific time intervals. Training, in particular retraining, and / or operating (e.g. controlling, prompting) the neural network can then be carried out depending on the one or more property parameters determined (also to be understood as: data dependent on the property parameters).According to a further embodiment, the method comprises using one or more computing components of the respective vehicle in a first computing mode during a standstill, in particular charging of the respective vehicle for executing the part of the vehicle-foreign computing task; and / or using one or more computing components of the respective vehicle in a second computing mode during a driving operation of the respective vehicle for executing the part of the vehicle-foreign computing task.According to a further embodiment, the method comprises ascertaining or predicting state information of the one or more vehicles as to whether the respective vehicle is at a standstill or in a driving mode or will at least be expected to be located, wherein the ascertaining or predicting is carried out in particular by means of at least one neural network, in particular the neural network; and forming the plurality of sub-tasks on the basis of the vehicle-foreign computing task, in particular dividing the vehicle-foreign computing task into a plurality of sub-tasks, depending on the one or more ascertained or predicted state information. In this case, the formation of the plurality of sub-tasks can be carried out in particular as a function of the output layer of the at least one neural network.According to a further embodiment, the at least one neural network is based on a foundation model or comprises a foundation model, wherein the foundation model takes into account, in particular also comprises and / or represents, the statistical dependencies, in connection with: a use of vehicles, in particular relating to a number of vehicles corresponding to a specific feature or a specific feature, in particular by specific users; and / or with the use of computing resources of one or more vehicles and / or of the available computing resources of one or more vehicles. These can be vehicles from a specific number or a number of the vehicles corresponding to a specific feature, in particular the one or more (selected) vehicles.According to a further configuration, the method comprises receiving a plurality of results from the plurality of sub-tasks, and combining results to form an overall result for the vehicle-foreign computing task; and providing the overall result for the vehicle-foreign computing task.According to a further embodiment, the method comprises a movement of one or more sub-tasks from a first vehicle to one or more second vehicles, depending on the initial layer of at least one (optionally further, separate) neural network, in particular the neural network (described in the independent claims), wherein the neural network is trained, in particular retraining, and / or controlled, depending on parameters which are characteristic of: a driving mode of the vehicle; and / or a road on which the respective vehicle is driving; and / or a current position of the vehicle; and / or a current and / or predicted degree of use of a vehicle function which is generated by one or more computing components of the respective vehicle.According to a further embodiment, the method comprises: making, in particular (also) negotiated an agreement with respect to a reward and / or a counter-performance for the execution of the vehicle-foreign computing task with the vehicle-external device, in particular by means of the neural network; and / or selecting a vehicle-foreign computing task from a number of vehicle-foreign computing tasks to be executed, in particular by means of the at least one neural network, in particular depending on a reward and / or counter-performance for the execution of the respective vehicle-foreign computing task; and / or depending on the determined or predicted available computing resources of one or more vehicles.For example, a calculation, accounting, compensation of costs and / or bids of one or more vehicles can be carried out by means of a micropayment (in particular also to be understood as: micro transactions, millipayment, nanopayment or picopayment) or by means of a data processing linked to a micropayment and / or configured according to one or more features of the micropayment. For example, the MPTP (=Micro Payment Transfer Protocol) can be used in the method.Preferably, compensation of advantages and / or disadvantages (alternatively or in addition to a currency or money units) which have been produced or expected on account of the execution of one or more steps of the method can be effected in specific (e.g. exchangeable or convertible) units. The units can comprise or be value units, e.g. money units.Alternatively or additionally, the units may identify an ecological characteristic, e.g. an at least prospective ecological balance (e.g. with respect to the result of at least a part of the computing task and / or per computing power), discounts and / or rights. For example, a calculation, accounting, compensation can be carried out on the basis of a vehicle or the number of vehicles.The method can be operated (in particular also to be understood as: controlled or executed) by means of (suitably set-up, in particular trained) bots (e.g. negotiation, optimization, accounting, accounting, compensation of units, division of the vehicle-external computing task into a plurality of sub-tasks and / or the formation of a group of vehicles for executing a computing task, using (in particular, also to be understood as: controlled or executed) a negotiation, optimization, accounting, accounting, accounting, compensation of units.The bots can be configured to determine and / or agree (e.g. by means of an operative connection to one another or a message exchange) a solution which is respectively improved according to predefined target values.According to a further aspect, a neural network is described. The neural network is configured, in particular trained and / or retraining or is retraining, to determine one or more parts of one or more vehicle-external computing tasks, wherein the one or more parts of one or more vehicle-external computing tasks are executable on one or more vehicles from a number of vehicles to achieve a result of the computing task.Furthermore, the neural network can be set up, trained or retrainable for (directly or indirectly) causing execution of a vehicle-external computing task on one or more vehicles from a number of vehicles in order to achieve a result of the computing task.According to a further aspect, a (road) motor vehicle (in particular a passenger car, transporter or a truck) is described which is configured to carry out the method (e.g. together with a device external to the vehicle, or the further part of the system, e.g. backend).For example, the one or more vehicles comprise a vehicle device for carrying out the method. The vehicle device can be configured to execute at least part of a vehicle-foreign computing task and / or to control and / or monitor the execution of the vehicle-foreign computing task. Furthermore, the vehicle device can be configured to provide, in particular to transmit or have it retrieve, a calculation result of the part of the vehicle-foreign calculation task (e.g. to the backend). Alternatively or additionally, the vehicle device may be configured to provide property information relating to one or more properties of the computing component.According to a further aspect, a (first) system for executing a computing task on one or more vehicles is described. The (first) system is configured to: determine one or more portions of one or more non-vehicle computing tasks depending on the output layer of a neural network; and (e.g., then depending on the result of the determining): cause execution of a non-vehicle computing task on one or more vehicles from a number of vehicles to achieve a result of the computing task.The (first) system is in particular further configured to carry out the method according to one or more features described in the present document. The (first) system can comprise one or more vehicles (e.g. with the corresponding vehicle devices) and at least one vehicle-external device, in particular a backend.According to a further aspect, a (second) system for executing a computing task on one or more vehicles is described. The (second) system is configured to: train, in particular train night, a neural network to determine one or more parts of one or more vehicle-external computing tasks, wherein the one or more parts of the one or more of a vehicle-external computing tasks are executed on one or more vehicles from a number of vehicles in order to achieve a result of the computing task (also to be understood as being executable).The (second) system is in particular further configured to carry out the method according to one or more features described in the present document. The (second) system can comprise one or more vehicles (e.g. with the corresponding vehicle devices) and at least one vehicle-external device, in particular a backend.For example, the system may be configured to set up, train and / or operate a foundation model, wherein the foundation model takes into account the statistical dependencies in connection with: the use of vehicles, in particular concerning a specific number of vehicles corresponding to a specific feature or a specific feature, in particular by specific users; and / or with the use of computing resources and / or of the available computing resources of one or more vehicles from a specific number or a number of the vehicles corresponding to a specific feature.Alternatively or additionally, the system can be configured to read in, operate and / or train night such a foundation model (e.g. from a further source). This can involve retraining with respect to the data which are at least predominantly relevant or representing for specific vehicles (the vehicles), users, and / or the conditions, routes of the vehicles prevailing during use and / or a measure of reaching or holding one or more target values with respect to the computing task.The at least one neural network may be configured, trained, and / or operated to achieve a measure of achieving or holding one or more target values (e.g., reducing costs, a faster and / or more reliable result of the computing task. This can be carried out, for example, for specific or specific features, corresponding computing tasks, vehicles, users of the vehicles, conditions during the use of the vehicles.The system may comprise the first system and / or the second system, in particular any combination of the systems described in the present document.According to a further aspect, a computer program, in particular a computer program product comprising a computer program, is described, wherein the computer program is designed, when it is executed on at least one arithmetic unit, to execute the method described in this document. The computer program can comprise, for example, the program code with which one or more vehicles are set up (during their production or later, in particular already after the delivery of the vehicle to the user) for executing the method. This computer program differs in particular from the program code (which can be used temporarily or depending on a current demand) of the vehicle-external computing tasks.For example, the computer program (e.g. by means of the new program code described in this document, which can differ from the program of the vehicle-foreign computing task) is configured to change, in particular temporarily or temporarily executable, an operating system or part of basic software of the at least one computing component of one or more vehicles (also to be understood as meaning: of the specific vehicle), in particular to operate and / or optimize the computing component for one or more vehicle-foreign computing tasks.In this case, the computer program can be designed to set up or optimize a computing component, which is set up or optimized during vehicle operation for processing the (typically specific) vehicle-own computing tasks, for processing the vehicle-foreign computing task (which can typically be of a different type, in particular of a specific type). For example, the computer program can have the effect that instead of vehicle-specific data, e.g. sensor data, the data of the vehicle-external computing task (which can be transmitted to the vehicle with the computing task) is processed, in particular in a particularly optimum manner.For example, the computer program or the computer program product can be provided to a user of the vehicle (e.g. to the vehicle and / or via a user device), with which the vehicle of the user is set up and / or optimized to execute one or more variants of the method.The computer program product can be designed as an update of a previous computer program, which comprises the parts of the computer program or of the corresponding program code, for example, in the context of a function extension, for example, in the context of a so-called "remote software update". The computer program product comprises, in particular, a medium readable by the data processing device, on which the program code is stored, or at least one encrypted file. According to a further aspect, a program product is described which comprises an authorized access right to stored data of the computer program product.It should be appreciated that the methods, systems, and systems described herein may be used both alone and in combination with other methods, devices, and systems described herein. Furthermore, any aspects of the methods, systems described in this document can be combined with one another in a variety of ways. In particular, the features of the claims can be combined with one another in many ways.The invention is described in more detail below on the basis of exemplary embodiments without limiting generality. The following are shown: FIG. 1 illustrates a vehicle 10.This is an at least partially, in particular fully electrically drivable vehicle 10, for example a hybrid vehicle or a fully electrically drivable vehicle 10. The information about this can be determined and taken into account in the method.Alternatively or additionally, the state of charge and / or charging process (e.g. a parameter of the state of charge and / or of the charging process) and / or of the energy requirement can be predicted and taken into account for a time interval in the future. For example, if (e.g. for the next day) a high electrical energy requirement is predicted and, in particular, if the vehicle 10 is not connected to a charging station (also to be understood as: power network, energy network), the computing task R or sub-tasks R 1, R 2, R 3 can be initiated preferentially to other vehicles 11, 12, 13 or only one (tendingly small) sub-task R 1, R 2, R 3 can be initiated to the vehicle 10.The vehicle 10 has hardware (within the scope of the document, in particular to be understood as meaning or to be equivalent to computing units of different types and / or data memories). For example, they are one or more processors, in particular GPP and / or specialized or highly parallel processors, e.g. graphics processors.One or more parameters of the hardware installed in the one vehicle and / or resources of the hardware available for executing at least part of the computing task external to the vehicle are determined. In particular, these can be queried from the backend BE or read in there and taken into account.The vehicle 10 is configured for an operative connection (on both sides) with the backend BE for carrying out the method 100.FIG. 2 illustrates an example of the system S including a backend BE and some of the plurality of vehicles 11, 12, 13, 14... The vehicles 11, 12, 13, 14,... can each (optionally) have the features discussed with the aid of FIG. 1. For example, vehicles 11, 12, 13 may be connected to the charging station (e.g. in each case at different time intervals), have different state of charge and / or current energy requirement or energy requirement predicted for a time interval in the future.The backend BE comprises a device K which is configured to divide a vehicle-external computing task R into a plurality of sub-tasks R 1, R 2, R 3. The division is carried out as a function of the information determined in the method (e.g. in step 100).The backend BE comprises a device A which is configured to combine the partial results E 1, E 2, E 3 to form an (at least preliminary) overall result E.The backend comprises a neural network DNN (described herein). This is also to be understood in such a way that it is (remotely) controlled) by the neural network DNN (e.g. at least partially spatially swapped).For example, the device K used for dividing a vehicle-external computing task R and / or the device A used for combining the partial results E 1, E 2, E 3 to form an (at least preliminary) overall result E is controlled by the neural network DNN.The individual devices, e.g. K, A, can be operated spatially together or separately from one another.The system S is configured for dynamic and / or adaptive operation (e.g. with respect to a change in the data processed in the method). For example, the division of the vehicle-external computing task R into a plurality of sub-tasks R 1, R 2, R 3 and / or the initiation of the execution of at least one part of the computing task, in particular of the sub-tasks R 1, R 2, R 3, is changed, in particular adapted, dynamically, e.g. depending on the always new ascertained information.FIG. 3 shows an exemplary flow chart or sequence of the method according to the invention on the basis of exemplary steps S 10, S 20, S 30, S 40, S 50, S 60, S 70 of method V, V 1, V 2. The method V, V 1, V 2 can be executed cyclically, with a plurality of successive operating cycles.The solid arrows indicate a preferred proposed sequence of steps. Further, the arrows may also mean that the step pointed to by the arrow is executed depending on the result of the step at which the arrow starts. The dash-dot arrows each mean an influence during an operating cycle, in particular a control of the execution of a step pointed to by the arrow, depending on the result of one or more steps (from a preceding operating cycle) from which the arrow or the arrows originate. The method can comprise in particular only a part of the sequence shown and / or also further steps.In step S 10, the data are ascertained with respect to one or more of the (previous) executions of one or more of steps S 20, S 30, S 40, S 50, S 60, S 70 or (appropriate, expected, assumed or predicted) initial values are used. These may (in each case) be qualitative and / or quantitative measures, in particular typically the execution of one or more of the steps mentioned.The data can be merged and / or processed, e.g. aggregated, from a plurality of vehicles and / or operating cycles executed in the past (e.g. in step S 10 and / or in a further step not shown in FIG. 3 ), in particular according to a predetermined mathematical dependency.The term "data" is also understood within the scope of the present document in particular as data determined as a function of the data.Then, depending on the data:the training data are determined; and / orthe back propagation (data for executing the back propagation) can be determined and / or back propagation can be initiated or controlled; and / orselecting or adjusting one or more parameters of the training process.Alternatively or additionally, the merged or aggregated and / or processed data (in step S 20) can be provided or used for operating S 30 the at least one neural network DNNN.The operation S 30 of the at least one neural network DNN can (also) comprise or be a case-specific control S 30 or control S 30 relating to one or more (specific) operating cycles.The operation S 30 of the at least one neural network DNN can be, in particular, a (e.g. case-specific, target value-specific) prompt.In step S 20, the training and / or retraining S 20 (e.g., according to the above descriptions) may be performed.In step S 30, the operation of the one or more neural networks DNN may be performed (e.g. as described above). For example, the input layer of the one or more (corresponding) neural networks DNN is executed depending on the data determined in step S 10.In step S 40, the determination of one or more parts of one or more other-vehicle computing tasks R; and / or the causing of the execution of one or more parts of one or more other-vehicle computing tasks R on one or more vehicles 10, 11, 12, 13 from a number of vehicles 10, 11, 12, 13, 14, 15,... can be carried out as a function of the output layer of the neural network DNN (correspondingly set, trained S 20 or night-trained S 20).In step S 50, one or more parts of one or more other-vehicle computing task R can be caused to execute. In this example, this is effected by means (e.g. respectively selected control units) of the vehicles 10, 11, 12, 13 (e.g. selected depending on the output layer of the neural network DNN from a higher-order number of the vehicles 10, 11, 12, 13, 14, 15... ).Alternatively or additionally to the causing, a control S 60 and / or distribution S 60 of the parts of one or more vehicle-external computing task R can take place. This can (also) take place during the execution of the (current) operating cycle of the method V, V 1, V 2, e.g. the handling of a vehicle-external computing task R.In step S 70, the one or (in each case) a plurality of parts of the vehicle-external computing task R is executed on one or a plurality of vehicles 10, 11, 12, 13,... The execution S 70, in particular the operation of the execution, can (also) be controlled S 60 depending on the neural network DNN, in particular dynamically and / or during operation.Thereupon, one or more results (calculation results) of one or more parts of the vehicle-external calculation task can be provided by the one or more vehicles. The provision can be made to the backend BE. A plurality of results can be received by the latter and / or fused outside the one or more vehicles 10, 11, 12, 13, in particular in the backend BE, to form an overall result.Furthermore, the data can be determined, in particular updated and / or aggregated, as characteristic of the execution of steps S 20, S 30, S 40, S 50, S 60, S 70 (for the first time or repeatedly) in the (next) step S 10.(further) optional features or steps of the method V, V 1, V 2 are described below, which features or steps can be combined both with one another and with all further steps described in the present document.In an example which can be combined with all the features described in the present document, the ascertainment of a plurality of parts of the vehicle-external computing task R, in particular of sub-tasks R 1, R 2, R 3 which can be executed in each case on one of a plurality of vehicles 10, 11, 12, 13, can be dependent on:one or more feedbacks of the one or more vehicles 10, 11, 12, 13 with respect to a progress in the execution of one or more parts of the computing task R; and / oralready executed parts of the computing task R.The expression "feedback of the one or more vehicles 10, 11, 12, 13 with respect to a (e.g. current) progress in the execution of one or more parts of the computing task R" is also to be understood as a feedback that characterizes (e.g. characterizes, describes or is characteristic of) the progress in the execution of one or more parts of the computing task R.The "target value with respect to the computing task R" is also to be understood in particular as the (concrete) target value with respect to the one or more specific computing tasks, vehicles, users or use of the vehicles, etc. The target value can comprise, for example, information which characterizes, describes or is characteristic of the target of the execution of the computing task, in particular characterizes (e.g. qualitatively and / or quantitatively).In a further example which can be combined with all the features described in the present document, it is possible to:determining S 40 one or more parts of the vehicle-external computing task R that are executable on the one or more vehicles 10, 11, 12, 13; and / orforming S 40 the plurality of sub-tasks R 1, R 2, R 3 on the basis of the vehicle-external computing task R, in particular dividing the vehicle-external computing task R into a plurality of sub-tasks R 1, R 2, R 3; and / orcausing S 50 the execution of a vehicle-external computing task R on one or more vehicles 10, 11, 12, 13;depending on an, in particular dynamically predefinable, target value.One or more vehicles 10, 11, 12, 13, in particular a part of the so-called customer fleet (i.e. vehicles 10, 11, 12, 13) in use in a plurality of end customers, preferably comprising one or more electric and / or a highly automated driving vehicles 10, 11, 12, 13, can be configured and / or used to carry out external or externally receivable computing tasks R or sub-tasks R 1, R 2, R 3. Such vehicles are actively used (for driving) on average (e.g., as opposed to commercially used fleets).The individual computing tasks R or R1, R2, R3may be tasks which are substantially independent of vehicle functions. For example, it may be a so-called mining. Calculations necessary for handling blockchain (e.g., on multiple vehicles 11, 12, 13, 14... ) can be calculated. For example, at least a part of the computing task R comprises mining for generating and / or handling a cryptocurrency. For example, the method V, V 1, V 2 is a method for generating and / or handling a cryptocurrency.It is particularly important that a system S can be or is built up (at least temporarily), that the system is set up, in particular optimized, to use a plurality of vehicles 10, 11, 12, 13 in a dynamic and / or flexible manner (including large) vehicle-external computing tasks R or a plurality of vehicle-external computing tasks R.An overlapping computing task R can be distributed and / or controlled (if appropriate during the execution of respective parts of the computing task, depending on the result of the ascertainment, in particular or prediction of an at least presumably available computing power of one or more vehicles 10, 11, 12, 13 to a plurality of vehicles 10, 11, 12, 13 and / or to a plurality of different time intervals.In this case, one or more parts of the vehicle-external computing task R can be determined, which can be executed on the one or more vehicles 10, 11, 12, 13, in such a way that a target value with respect to the computing task R (which preferably corresponds substantially to the entire computing task or comprises all essential or computing-intensive parts thereof) is at least presumably reached. In this case, the parts of a vehicle-external computing task R can be initiated S 50 on one or more vehicles 10, 11, 12, 13.FIG. 4 shows the operation S 30 of the (e.g. pre-trained, pre-trained, pre-trained) neural network DNN.These can be a plurality of neural networks DNN operated or operable (possibly in the backend and / or in one or more vehicles 10, 11, 12, 13,... These may be interconnected or crosslinked with one another. The method is preferably carried out by means of a central neural network DNNZ. This can be pre-trained S20 during baking and / or in the vehicles 10, 11, 12, 13,... and / or re-trained S20' depending on vehicle data.In this case, the one or more executions S 70 of the vehicle-external computing tasks R are operated S 60 (in particular also to be understood as: controlled or influenced in their operation) as a function of the operation S 30 of the one or more neural networks DNN(Z).The at least one neural network DNN(Z) can be trained S 20 and / or operated S 30 (in particular also to be understood as: used) for executing (even only) a part of the method, in particular one or more steps of the method.Operating S 20 the at least one neural network DNN, DNN 10, DNN 20, DNNZ can comprise or be the execution of at least one step of the method by means of the at least one neural network DNN 1, DNN 2, DNNZ. For example, the neural network DNN 1, DNN 2, DNNZ can form part of the system S according to the invention.The method can comprise or be (also only or with a combination of further features from the present document) training S 20 and / or operating S 30 the at least one neural network DNN for carrying out one or more steps of the method.For different vehicles 10, 11, 12, 13, 14, 15 or selection of vehicles 10, 11, 12, 13, a respectively different, in particular specific training S20, in particular retraining S20' or different neural networks DNN10, DNN20,... or neural networks operated with different learning data, respectively. DNNZ can be used. For example, the at least one neural network DNN 10, DNN 20, DNNZ trained S 20, S 20' and / or operated S 30 can aim, in particular achieve, a target or target value relating to the vehicle 10, one or more further vehicles 11, 12, in particular relating to a number of vehicles 10, 11, 12, 13, 14, 15.For example, the at least one neural network DNN 10, DNNZ in the backend BE can be trained S 20, S 20' and / or operated S 30. Alternatively or additionally, the at least one neural network DNN 10, DNN 20, DNN 30, DNN 40, DNN 15 in the respective vehicle 10, 11, 12, 13, 14, 15 and / or depending on the data of respective vehicles 10, 11, 12, 13,... can be trained S 20, in particular retrained S 20' and / or operated S 30.The at least one neural network DNN 10, DNN 20, DNNZ can be understood as data (e.g. topology, learning data, checkpoint, weight information, bias terms) for establishing the neural network DNN 10, DNN 20, DNNZ by means of at least one neural network DNN 1, DNN 2, DNNZ (e.g. on a data processing device). Therefore, a further neural network DNN 30, DNN 40, in the sense of the present document, is the same neural network DNN, DNNZ if it is or is at least predominantly depleted on a different computing unit.For example, the data are trained S20 on a data processing device in order to then make them S20' and / or operable S30 trainable on one or more other data processing devices, in particular on (specific) vehicles 10, 11, 12, 13,... or backend BE and / or for the execution of vehicle-foreign computing tasks R on vehicles 10, 11, 12, 13, 14, 15,... and / or on specific vehicles 10, 11, 12, 13.The at least one neural network DNN, DNN 10,... DNNZ may include, be, or include, one or more features of a convolutional neural network (CNN). The CNN is also to be understood in particular (according to corresponding German terms) as a convolutional neural network, a convolutional layer-comprising or so-called convolutional neural network. For example, the at least one neural network DNNDNZ can comprise one or more convolutional layers (also to be understood as: convolutional layer).In training S 20, the core of the matrix may be learned. In this case, the core (for example with regard to the dimensioning) can be smaller than the input parameters. This core can be used for a plurality of cases, positions, various circumstances described herein, etc.The at least one neural network DNN, DNN 10, DNN 20, DNNZ can recognize or learn influences, in particular structures of influences, that occur independently of case, location, and / or independently of one or more specific vehicles and / or specific vehicle-external computing tasks.During operation S 30, these can be used, for example, for new cases or circumstances in each case. This can be carried out by means of a locally meshed layer (e.g. configured for this purpose), in particular the convolutional layer (convolutional layer).Furthermore, the information about the influences can be compressed in a matrix or transferred into a more abstract representation. In this case, at least predominantly in each case strongest features of a matrix can be prioritized or strengthened or the weaker features can be discarded. This can be done by means of the locally meshed layer (e.g. configured for this purpose), in particular pooling layer or subsampling layer. This can lead to a further reduction of the data to be further processed.Furthermore, the results from the two layers of the above-mentioned types mentioned above can be processed with one another, in particular can be combined analogously. In this case, the values received and / or the values received from the two slices can be processed together, in particular be calculated. This can be done in the third layer (or in the third step within the neural network DNN, DNN 10, DNN 20,... DNNZ). This can be an at least predominantly, in particular completely meshed layer of the at least one neural network DNN, DNN 10, DNN 20, DNNZ.Each of these optional features (e.g. of specific layers, the matrix of the topology, etc.) of the one or more neural networks DNN, DNN 10, DNN 20, DNNZ can be configured, dimensioned and / or (at least to a certain degree) optimized for the execution of at least one (corresponding) part of the method.The term "parameter" is also to be understood in the context of this document analogously as a parameter value or a measure of the change or deviation of a corresponding parameter value, for example of a change or deviation from a preceding value, a standard value or setpoint value. For example, a parameter can characterize a measure of the change in a parameter (to be understood as also: parameter value) over time and / or in relation to a (spatially variable) variable relating to the space, e.g. direction and / or distance. In particular, a parameter corresponds to a pattern (e.g. recognizable or executable as such) of parameters or parameter values.The term "characterizing" is also to be understood in particular as "representing" within the scope of the present document.The determination of one or more data mentioned in this document can in particular also be a prediction of the respective data, in particular with respect to one or more specific time intervals, e.g. of a few minutes, hours, days. Preferably (both) a determination (e.g. related to a value valid in the past or presence) and (also) a prediction (related to a future time interval, e.g. related to a specific time interval) are carried out.For example, the training S 20, S 20' and / or operation S 30 of the at least one neural network DNN 1, DNN 2, DNNZ takes place as a function of data, in particular parameter(s) which are determined with respect to one or more (e.g. specific) time intervals and are predicted for one or more (e.g. specific) time intervals.Figure 5 illustrates a variant of the invention.In this variant, the training S 20 of the at least one neural network DNNZ takes place in the backend BE.In this case, the training data can be determined on the basis of the first data and / or learning data of the at least one neural network DNN in the backend BE (the neural network is denoted there as DNN(Z). These can in turn be transmitted (directly or indirectly) to the vehicle 10 (e.g. in response to the transmission of a multiplicity of data) for use in the vehicle 10. The learning data (in particular to be understood as data of the at least one neural network DNN or for setting up the neural network DNN 1 can also be loaded and / or loaded onto the first vehicle 10 and / or (optionally) onto further respective vehicles 11, 12, 13, 14, 14 in order to operate the at least one neural network DNN therein depending on the learning data.This or such a variant is advantageous, for example, because a substantially larger quantity of the first data is used as output data for generating the training data. For example, first data can be processed (only or primarily) to form training data which are in a (temporal and / or causal relationship with (particularly) highly pronounced driver errors and / or driver errors detected with high probability. In this case, one or more predetermined goals of training or goals determinable in the method, in particular with respect to the achievement of target values (e.g. better and / or faster and / or more precise), can be achieved.Each of the steps of the method described herein (e.g., S 10, S 20, S 30, S 40) may be executed dynamically (e.g., cyclically and / or event-driven). For example, the triggering S 50 of the execution of a vehicle-external computing task R on one or more vehicles 10, 11, 12, 13 can take place as a function of the respectively valid prediction of an at least presumably available computing power of one or more vehicles 10, 11, 12, 13 and / or a (respectively current) target value.For example, this can comprise ongoing changes, in particular adaptations depending on new and new ascertained and / or predicted information.The one or more vehicles 10, 11, 12, 13 are preferably electric vehicles. An electric vehicle 10, 11, 12, 13 is distinguished, for example, in that it connects a relatively large portion of the time at which the vehicle 10, 11, 12, 13 is not used for the driving operation to an electrical grid (also to be understood as: charging station), so that the electrical energy supply of the computing components of the (respective) vehicle 10, 11, 12, 13 is ensured.Alternatively or additionally, an electric vehicle 10, 11, 12, 13 typically has electrical energy reserves (e.g. a high-voltage energy store) which enable operation of the computing components of the vehicle over a relatively long period of time when the vehicle 10, 11, 12, 13 is at a standstill.The one or more computing components of the vehicle 10, 11, 12, 13 can be computing units (hardware) of the vehicle 10, 11, 12, 13 which are installed or usable for at least partially automated driving.The vehicle 10, 11, 12, 13 may be configured to play software onto the one or more computing components of the vehicle 10, 11, 12, 13 in a comparably flexible manner. The software for operating the vehicle 10, 11, 12, 13 can be stored as a backup, while the one or more computing components are used to solve a computing task R. After the completion of the computing task R, the software for the operation of the vehicle 10, 11, 12, 13 can then be installed again. The software can be transmitted via the communication unit of the vehicle 10, 11, 12, 13. In this case, the charging cable for charging an electrical energy store of the vehicle 10, 11, 12, 13 and / or a wireless communication connection (e.g. 3G, 4G or 5G) can be used as the data connection.By means of a data exchange, a computing task R, in particular input data to a computing task R (also: for executing the computing task), can be read in, in particular called up, from outside the vehicle 10, 11, 12, 13. The computing task R can then be executed in the vehicle 10, 11, 12, 13, and the result E or E1, E2, E3 of the computing task R can be transmitted (e.g. to the same computing unit V by which the computing task R or R1, R2, R3 has been initiated or to a computing unit A networked thereto).The one or more vehicles 10, 11, 12, 13 can be configured to perform an (information-related) operative connection with the vehicle-external computing unit (also: central unit).In this case, an identifier and / or information for the available computing resources, e.g. nonvolatile memory, volatile memory, can be communicated (overall and / or in each case per control device) in the (respective) vehicle 10, 11, 12, 13. Alternatively or additionally, an identifier and / or information for the degree of use of the available resources for vehicle functions can be communicated. This can also be understood as information about the available resources. Alternatively or additionally, a forecast and / or time planning can be communicated with available resources of the (respective) vehicle.Furthermore, the method can comprise dividing a part of a computing task R 1, R 2, R 3 (also) within the vehicle 10, 11, 12, 13 (transmitted to the respective vehicle) (also to be understood as forming sub-tasks which each serve for processing the vehicle-external computing task or its parts). In particular, one or more read-in parts of the computing task R 1, R 2, R 3 can be divided among a plurality of computing units or computing components of the respective vehicle 10, 11, 12, 13. This can be done using the (communication) on-board power supply system of the vehicle 10, 11, 12, 13. In this context, the respective available computing resources in the individual computing units (e.g. control units) can be determined, e.g. by means of a query within the vehicle 10, 11, 12.The division of a computing task R between the different vehicles 10, 11, 12, 13 and / or computing components of the (respective vehicle 10, 11, 12, 13 can then be carried out in accordance with the type of required computing operations (e.g. GPU= Graph Processing Unit), parallel processing, graphics processing, neural processing).The vehicle 10, 11, 12, 13 can be configured with a first calculation mode which can be used in the stationary state, for example at night, in the case of a longer stationary state of the vehicle 10, 11, 12, 13, in the vacation state, in the case of deregistration of the vehicle 10, 11, 12, 13, etc. The first computing mode can comprise the flashing of at least partially new firmware onto one or more computing units (e.g. control units) of the respective vehicle 10, 11, 12, 13. In this case, the resources of one or more EEPROMs (electrically erasable programmable read-only memory) of the vehicle 10, 11, 12, 13 can also be used (directly or indirectly).The one or more computing units of the vehicle 10, 11, 12, 13 can be used, in particular, for an executable program code which is different from the vehicle function software. The program code can be part of the computing task R (or a sub-task R 1, R 2, R 3). Input data relating to the computing task R (or a sub-task R 1, R 2, R 3) can be stored on the one or more computing units. Alternatively or additionally, the one or more computing units can be used as a medium-term or long-term buffer or for memory management for the (global) computing task R. One or more vehicles 10, 11, 12, 13 can thus be used for storing (vehicle-external and / or vehicle-independent) data.The one or more vehicles 10, 11, 12, 13 can be configured or designed to execute a second computing mode which can be used during vehicle operation, in particular during driving operation. The second calculation mode can be enabled, for example, during driving operation on specific roads, locations, countries, etc. The second calculation mode can be provided depending on the degree of use and / or on the prediction of use of vehicle functions of the (respective) vehicle 10, 11, 12, 13. It is thus possible to allow at least a portion of the computing resources of a vehicle 10, 11, 12, 13 to be used during vehicle operation.The one or more vehicles 10, 11, 12, 13 can be configured to allocate a portion or a region of a ROM or working memory and / or a processor computing power (e.g. one or more cores of the processors) for the computing task R or R 1, R 2, R 3 in the second computing mode. Alternatively or additionally, the one or more vehicles 10, 11, 12, 13 can be configured to carry out, in the second computing mode, the management, in particular planning, displacement, deceleration, aborting, resuming, etc., of an external computing task R or R 1, R 2, R 3 within the vehicle 10, 11, 12, 13, in particular per control device or within the control devices.The one or more vehicles 10, 11, 12, 13 can be configured to carry out dynamic management, in particular planning, relocation, deceleration, aborting, resuming, etc., of an external computing task R or R 1, R 2, R 3 within the respective vehicle 10, 11, 12, 13 in the second computing mode.Alternatively or additionally, the one or more vehicles 10, 11, 12, 13 can be configured to move a part of the computing task R 1, R 2, R 3 from one computing unit of the vehicle to another computing unit of the respective vehicle 10, 11, 12, 13. This can be effected, for example, depending on:• a specific road, location, or country on or in which the vehicle 10, 11, 12, 13 is in driving operation; and / or• the degree of use and / or the prediction of use of one or more vehicle functions of the vehicle 10, 11, 12, 13.The method can comprise, in particular in the second computing mode described further, being configured to determine a time prediction, a risk of a break-off or a pause of a computing task, in particular a contiguous sequence of the computing tasks R, by one or more vehicles. A distribution of the sub-tasks R 1, R 2, R 3 to one or more vehicles 10, 11, 12, 13 and / or an acceptance of one or more sub-tasks by the one or more vehicles 10, 11, 12, 13 can then be changed. Alternatively or additionally, the handling of one or more computing tasks R or sub-tasks R 1, R 2, R 3 within one or more vehicles 10, 11, 12, 13 can then be changed.Alternatively or additionally, depending on the output of the at least one neural network DNN, a forwarding, in particular an under-ordering, of a computing task R or a sub-task R 1, R 2, R 3 from one vehicle 10, 11, 12, 13 to another vehicle 10, 11, 12, 13 can be carried out (if applicable thereupon).The method may comprise a "brokerage" and / or "contracting" with respect to a computing task and / or sub-tasks R 1, R 2, R 3. In particular, the method may include comparing one or more offers from one or more vehicles 10, 11, 12, 13 with respect to a reward. A comparison or balancing, in particular an optimization, of the acceptance of a computing task R or sub-task R 1, R 2, R 3 can take place. For example, it can be weighed whether a larger computing task R is accepted for a longer time for less counter-power (e.g. money units, counter-power units, triggerable conditions) per computing power or a smaller computing task or sub-task R1, R2, R3 with more counter-power or money per computing power.For example, the at least one neural network DNN, DNN 10, DNN 20, DNNZ can be set up, in particular trained, and / or operated to perform the brokerage and / or contract with respect to a computing task and / or sub-tasks R 1, R 2, R 3. In this case, it can optimize it to a target value (for example, changeable or predefinable by a user).Alternatively or additionally, the one or more vehicles 10, 11, 12, 13 can be configured to send and / or adapt a separate offer for accepting a computing task R or a sub-task R 1, R 2, R 3 (e.g. to the central unit BE). Furthermore, the backend BE and / or the one or more vehicles 10, 11, 12, 13 can be configured to bring about an automatic completion of an agreement with respect to the execution of a computing task R or of one or more sub-tasks R 1, R 2, R 3.The invention or measures described in this document can substantially increase the utilization of the computing resources of a vehicle 10, 11, 12, 13.Thus, the effective costs of a vehicle, in particular costs for operating a vehicle or total costs, can be reduced (e.g. by marketing and / or using the computing resources for vehicle-independent computing tasks). Furthermore, the measures described can also be used to use computing resources of a vehicle which are (if appropriate still) not required for the provision of a vehicle function.The measures described in this document are particularly advantageous in the case of at least partially electrically driven vehicles, for example during a charging process and / or on account of the availability of an electrical energy store. The one or more vehicles are preferably vehicles which can be driven at least partially in an automated manner. Such vehicles have substantially more powerful hardware (computing units, memories, etc.).The present invention is not limited to the exemplary embodiments shown. In particular, it should be noted that the description and the figures are intended to illustrate, by way of example only, the principle of the proposed methods, apparatuses and systems. All features identified in the exemplary embodiments shown can be freely combined with all other features or claims.

Claims

Method (V, V1) for executing a computing task (R) on one or more vehicles (10, 11, 12, 13), comprising: - determining (S40) one or more parts of one or more other-vehicle computing tasks (R) depending on the output layer of a neural network (DNN), and - causing (S50) the execution of one or more parts of one or more other-vehicle computing tasks (R) on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14, 15,..) to achieve a result of the computing task (R).Method (V, V2) for executing a computing task (R) on one or more vehicles (10, 11, 12, 13), comprising: - training (S20), in particular retraining (S20'), a neural network (DNN) to determine (S40) one or more parts of one or more other-vehicle computing tasks (R); wherein - the one or more parts of an other-vehicle computing task (R) are determined for executing (S60) the computing task (R) on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14...) in order to achieve a result of the computing task (R).Method (V, V1) according to Claim 1, comprising - predicting one or more respective at least presumably available computing resources on a plurality of vehicles (10, 11, 12, 13, 14...) from a number of vehicles as a function of the neural network (DNN), and: - determining (S40) one or more parts of one or more other-vehicle computing tasks (R) as a function of the output layer of a neural network (DNN); and / or - causing (S50) the execution of an other-vehicle computing task (R) on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14...); as a function of the result of the prediction.Method (V, V1) according to claim 1 or 3, comprising: - selecting one or more vehicles (10, 11, 12, 13) for executing one or more respective parts of the non-vehicle computing task (R) from the number of vehicles (10, 11, 12, 13, 14...); and / or - assigning one or more parts of a non-vehicle computing task (R) to one or more vehicles (10, 11, 12, 13).Method (V, V2) according to Claim 2, wherein the training (S20), in particular retraining (S20') the neural network (DNN), comprises - predicting one or more respective at least presumably available computing resources on a plurality of vehicles (10, 11, 12, 13, 14...) from a number of vehicles (10, 11, 12, 13, 14...); and / or - selecting one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14...) for executing one or more parts of the vehicle-foreign computing task (R); and / or - assigning one or more parts of a vehicle-foreign computing task (R) to one or more vehicles (10, 11, 12, 13).Method (V, V1) according to one of the preceding claims, wherein - the one or more parts of the vehicle-foreign computing task (R) are determined (S40) in such a way; and / or - the one or more vehicles (10, 11, 12, 13) for executing one or more parts of the vehicle-foreign computing task (R) are selected from a number of vehicles (10, 11, 12, 13, 14...) in such a way; and / or - the one or more parts of a vehicle-foreign computing task (R) are assigned to one or more vehicles (10, 11, 12, 13) in such a way; that one or more, in particular adjustable, target values relating to the computing task (R) are aimed at, in particular reached and / or held when they are executed on the one or more vehicles (10, 11, 12, 13).Method (V, V2) according to one of the preceding claims, wherein the neural network (DNN) is trained (S20), in particular retraining (S20'), depending on the use of computing resources and / or on the available computing resources during operation: - a specific number or a number of the vehicles (10, 11, 12, 13,...) corresponding to a specific feature, in particular of the one or more vehicles (10, 11, 12, 13,...); and / or - one or more further vehicles (10', 11', 12', 13') operated under similar conditions to the one or more vehicles (10, 11, 12, 13); and / or - one or more vehicles (10, 11, 12, 13,... used by the same users or users with respectively similar user profiles. 10', 11', 12', 13').Method (V, V1, V2) according to claim 6, wherein the one or more target values relate to a cost required for executing at least a part of the computing task (R), in particular a number of sub-tasks (R1, R2, R3), in particular: - power costs for executing the one or more parts of the computing task (R); and / or - data transmission costs for transmitting at least a part of the computing task (R), in particular a number of the plurality of sub-tasks (R1, R2, R3), to one or more vehicles (10, 11, 12, 13); and / or - data transmission costs for transmitting at least a part of the one or more computing results (E1, E2, E3) from one or more vehicles (10, 11, 12, 13).Method (V, V1, V2) according to Claim 6 or 8, wherein the one or more target values relate to the execution of a plurality of, in particular all essential, parts of the computing task (R), in particular all essential part tasks (R1, R2, R3) of the one or more vehicle-external computing tasks (R1, R2, R3), on one or more vehicles (10, 11, 12, 13), in particular of the vehicles (10, 11, 12, 13, 14...) selected from a number of vehicles (10, 11, 12, 13).Method (V, V1, V2) according to one of the preceding claims, comprising - determining or predicting one or more parameters relating to one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14...), and: - taking into account the determined or predicted parameters in the method, in particular operating (S30) the neural network (DNN) as a function of the parameters, wherein the parameters are characteristic of: - connecting the respective vehicle (10, 11, 12, 13, 14...) to a charging station; and / or - charging state of the battery of the respective vehicle (10, 11, 12, 13, 14...); and / or - energy requirement of the respective vehicle (10, 11, 12, 13, 14...) relating to the use of the vehicle.Method (V, V1, V2) according to one of the preceding claims, comprising: - determining a property parameter relating to one or more properties of one or more vehicles (10, 11, 12, 13,...) from the number of vehicles (10, 11, 12, 13, 14,...), wherein the parameters are characteristic: - one or more computing resources of the one or more vehicles (10, 11, 12, 13,...); and / or - a degree of utilization of the computing resources of the one or more vehicles (10, 11, 12, 13,...), in particular for a further computing task (R) with respect to one or more specific time intervals; and / or - a set of the at least presumably available computing resources of the plurality of vehicles (10, 11, 12, 13,...) with respect to one or more specific time intervals; and, thereupon: training (S20), in particular retraining (S20') and / or operating (S30) the neural network (DNN), depending on the property parameters.Method (V, V1, V2) according to one of the preceding claims, comprising - using one or more computing components of the respective vehicle (10, 11, 12, 13,...) in a first computing mode during a standstill, in particular charging of the respective vehicle (10, 11, 12, 13,...) for executing the part of the vehicle-foreign computing task (R); and / or - using one or more computing components of the respective vehicle (10, 11, 12, 13,...) in a second computing mode during a driving operation of the respective vehicle (10, 11, 12, 13,..) for executing the part of the vehicle-foreign computing task (R).Method (V, V1, V2) according to one of the preceding claims, comprising: - determining or predicting a state information item of the one or more vehicles (10, 11, 12, 13,...) as to whether the respective vehicle (10, 11, 12, 13,...) is at a standstill or in a driving mode or will at least be expected to be located, wherein the determining or predicting is carried out in particular by means of at least one neural network (DNN), in particular the neural network (DNN); and - forming the plurality of sub-tasks (R1, R2, R3) on the basis of the vehicle-external computing task (R), in particular the splitting of the vehicle-external computing task (R) into a plurality of sub-tasks (R1, R2, R3), depending on the one or more ascertained or predicted state information; wherein the forming of the plurality of sub-tasks (R1, R2, R3) is carried out in particular depending on the output layer of the at least one neural network (DNN).Method (V, V1, V2) according to one of the preceding claims, wherein the at least one neural network (DNN) is based on a foundation model or comprises a foundation model, wherein the foundation model takes into account the statistical dependencies in connection with: - use of vehicles, in particular relating to a specific number of vehicles (10, 11, 12, 13, 14...), in particular by specific users, which number corresponds to a specific feature; and / or - use of computing resources of one or more vehicles and / or of the available computing resources of one or more vehicles, from a specific number or a number of the vehicles (10, 11, 12, 13...), in particular of the one or more vehicles (10, 11, 12, 13).Method (V, V1, V2) according to one of the preceding claims, comprising: - receiving a plurality of results (E1, E2, E3) from the plurality of sub-tasks (R1, R2, R3), and combining results (E1, E2, E3) to form a total result (E) for the vehicle-foreign computing task (R); and - providing the total result (E) for the vehicle-foreign computing task (R).Method (V, V1) according to one of the preceding claims, comprising - moving one or more sub-tasks (R1, R2, R3) from a first vehicle (10, 11, 12, 13) to one or more second vehicles (10, 11, 12, 13), depending on the initial layer of at least one neural network (DNN), in particular the neural network (DNN), wherein the neural network (DNN) is in particular trained (S20) and / or controlled (S30), depending on parameters which are characteristic of: - a driving mode of the vehicle (10, 11, 12, 13,...); and / or - a road on which the respective vehicle (10, 11, 12, 13,...) is driving; and / or - a current position of the vehicle (10, 11, 12, 13,...); and / or - a current and / or predicted degree of utilization of a vehicle function generated by one or more computing components (RES) of the respective vehicle (10, 11, 12, 13,... ).Method (V, V1) according to one of the preceding claims, comprising: - making an agreement with respect to a compensation and / or a counter power for the execution of the vehicle-external computing task (R) with the vehicle-external device (BE), in particular by means of the neural network (DNN); and / or - selecting a vehicle-external computing task (R) from a number of vehicle-external computing tasks (R, R', R") to be executed, in particular by means of the at least one neural network (DNN), in particular depending on a compensation and / or counter power for the execution of the respective vehicle-external computing task (R, R', R"); and / or depending on the determined or predicted available computing resources (RES) of one or more vehicles ( 10, 11, 12, 13,... ).Neural network (DNN), wherein the neural network (DNN) is configured, in particular trained (S20) and / or retraining (S20') or retraining (S20') to determine (S40) one or more parts of one or more other-vehicle computing tasks (R), wherein the one or more parts of one or more other-vehicle computing tasks (R) are executable on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14, 15,..) to achieve a result of the computing task (R).A system (S) for executing a computing task (R) on one or more vehicles (10, 11, 12, 13,...), wherein the system (S) is configured to determine (S40) one or more parts of one or more other-vehicle computing tasks (R) depending on the output layer of a neural network (DNN); and to cause (S50) execution of an other-vehicle computing task (R) on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14,...) to achieve a result (E) of the computing task (R).System (S) for executing a computing task (R) on one or more vehicles (10, 11, 12, 13), wherein the system (S) is configured to determine (S40) one or more parts of one or more other-vehicle computing tasks (R) for training (S20), in particular retraining (S20'), a neural network (DNN), wherein the one or more parts of the one or more of an other-vehicle computing task (R) are executed on one or more vehicles (10, 11, 12, 13) from a number of vehicles (10, 11, 12, 13, 14...) in order to achieve a result (E) of the computing task (R).Computer program product, comprising a computer program, wherein the computer program product is configured, when executed on at least one arithmetic unit, in particular on one or more arithmetic units of the system according to claim 19 or 20, to execute one or more steps of the method (V, V1, V2) according to one of claims 1 to 17.

Images