Computing device, in particular central computing device, of a motor vehicle

Abstract

Computing device (1), in particular central computing device (2), of a motor vehicle (100), designed and configured in such a way as to - to establish at least one communication between functions (A, B, ..., N) and at least one hyper-agent stack (HAS) (6) in such a way as to supply requirements (21, 22) of the functions (A, B, ..., N) to at least one function agent in order to carry out an evaluation of at least one of the requirements (21, 22) of the functions (A, B, ..., N); and to - to establish at least one communication between at least one signal generator (16) and the at least one hyper-agent stack (HAS) (6) in such a way as to supply requests (21, 22) and / or states of the signal generator (16) to at least one signal generator agent in order to carry out an evaluation of at least one of the requests (21, 22) of the signal generator (16) and / or to carry out an evaluation of at least one of the states of the signal generator (16); wherein in the event of a conflict (50) of requests (21, 22) for influencing a signal generator (16), a hyper-agent (15) of the HAS (6) decides, based on the evaluations of the requests (21, 22) and / or the states, how the conflict (50) is to be resolved.

Description

The invention relates to a computing device. The invention relates to a motor vehicle. The invention relates to a method for resolving conflict of requests. The invention relates to a computer program product.Known vehicle architectures often have a multiplicity of control units (ECUs) which communicate with one another via different bus systems. A plurality of a wide variety of actuators and reactions can be connected to these bus systems or control units connected thereto. The regulation of these actuators, i.e. the application of the various functions in the vehicle, is complex and multi-layered. The control units are developed separately from one another, as a result of which a large number of problems can arise in combination when all the control units come together, for example conflicts for the manipulated variables.The document DE 10 2019 219 464 B3 relates to a method for operating a self-driving motor vehicle which uses a plurality of control units and program codes for the autonomous driving operation. In the semi-autonomous mode, the vehicle functions are prioritized and weighted. When executing the functions, a distribution of the functions to different control units (application placement) is evaluated and optimized by means of a target reaching level. The functions are enabled depending on the achieved target reaching level.The publication US 2020 / 0 128 082 A1 relates, according to summary, to a vehicle arbitration system which is configured to coordinate a plurality of requests from different application units by means of a main manager and to determine corresponding control requests for specific vehicle components. Additionally, the system includes a plurality of sub-managers configured to arbitrate requests determined by the main manager and requests from other application units different from the first application units, and control the vehicle component based on the arbitration result.The document DE 10 2016 222 091 A1 relates to an electronic control unit which prioritizes software applications when a plurality of processes require resources simultaneously. Priority evaluation distributes the resources so that higher priority processes are preferentially executed to ensure optimum system performance.According to the summary, the post-published document DE 10 2022 208 004 A1 describes a method for controlling the access of a plurality of applications to a safety-relevant vehicle component. In this case, the accesses of the applications are first evaluated and each application is classified qualitatively on the basis of the evaluations. Subsequently, the access of the applications to the safety-relevant component is prioritized, applications with a higher quality rating being preferred in order to ensure safety.The publication DE 10 2011 117 116 A1 describes a control device for the at least partially autonomous operation of a vehicle. This control device comprises at least two arithmetic units which form a decentralized system and execute algorithms which communicate with one another. Two wired networks are available for communication: a first network allows the computing units to communicate directly with one another, especially within a middleware environment, while a second network connects the computing units to sensors and serves as a backup for the communication between the computing units if the first network fails or is overloaded.The solutions described are not regarded as satisfactory.It is the object of the invention to improve the regulation.The object is achieved in particular by a computing device having the features of claim 1. the object is achieved in particular by a motor vehicle having the features of claim 9. the object is achieved in particular by a method having the features of claim 10. Features and details which are described in connection with the computing device according to the invention also apply in connection with the method according to the invention, the computer program product according to the invention and the motor vehicle according to the invention. This also applies in each case in reverse, so that with regard to the disclosure reference is or can always be made to the individual aspects of the invention reciprocally.According to one aspect, the object is achieved in particular by a computing device having the features of claim 1.In this case, a computing device, in particular designed as a central computing device, of a motor vehicle, can be designed and configured in such a way as to configure at least one communication between functions and at least one hyper-agent stack-hereinafter referred to for short as HAS-, in particular in the form of a hyper-agent stack function-hereinafter referred to for short as HASF-, in such a way as to supply requests for the functions to at least one function agent in order to carry out an evaluation of at least one of the requests for the functions. The computing device can be designed and configured to form at least one communication of at least one signal transmitter and the at least one HAS, in particular in the form of an HASF, in such a way as to supply requests and / or states of the signal transmitter to at least one signal transmitter agent in order to carry out an evaluation of at least one of the requests of the signal transmitter and / or an evaluation of at least one state of the signal transmitter. In the event of a conflict of requests for acting on a signal transmitter, a hyper-agent of the HAS, in particular of the HASF, can decide how to resolve the conflict, based on the evaluations of the requests and / or of the states.As a result, a large part of the problems existing at the registration time or at the priority day and the existing complexity in vehicle communication, which are based on conflicts, i.e. if, for example, different control units wish to make different decisions, can be solved. The problem can also be solved that functions and control units often originate from different suppliers or developers and are therefore not mutually completely known, but rather function as a kind of black box. These conflicts can be identified and implemented, thereby allowing them to be resolved. These conflicts of goals can be identified and implemented in particular in a non-automated manner. The achievement of these conflicts can be influenced by tasks set in this case and the requirements placed in this respect on functions and signal transmitters, e.g. relating to safety, right, efficiency, comfort or performance.The vehicle architecture described here and elsewhere may have a single (high-performance) computing device, in particular central computing device, in which a plurality or all of the control devices may be combined, which may also be referred to as a central computer. The conflicts which may still exist between the individual functions can, however, be easily and better resolved here, in particular in the central computer, using a corresponding computing device as described here.Individual functions can be implemented in the computing device, in particular designed as a central computing device and / or as a central computer. The functions communicate in particular with a hyper-agent stack function (e.g. via DDS data distribution service of a motor vehicle electrical system for communication), which can implement a hyper-agent stack (HAS). In this case, one agent or in this case a plurality of agents can be provided which map corresponding tasks which are intended to be fulfilled via a function. Each request for a function runs through at least one such agent. Each function is not mandatory to provide or pass through its own agent, but rather one or more agents may represent one task or the plurality of tasks, as described elsewhere herein. These task / tasks of one agent or of the plurality of agents can be defined as desired. In an agent, the requirements of the functions can be evaluated. This evaluation can be based in particular on an evaluation function.Signal generators can also be represented via their own agent representation, wherein the agent of the signal generator represents its requirements and / or its state.In this case, it can also be provided in particular that the agents for one or more signal generators represent or represent the tasks which are intended to be implemented via the signal generators and / or via their interaction with vehicle components.For example, if two different functions attempt to control the same signaling device, such as an actuator, with different values, a hyper-agent, which may be configured or mapped via the HASF, may decide how to resolve the conflict based on the evaluations from the agent evaluating the function requirements and the agent evaluating the signaling requirements, such as to perform a corresponding task.In this case, in particular the at least one agent which evaluates the function requirements and the at least one agent which evaluates the signal transmitter requirements, in particular together with the hyper-agent, form a so-called hyper-agent stack-referred to here for short as HAS. By means of the hyper-agent, function requirements and signal generator states, in particular in the form of actuator states, can be evaluated centrally, whereby conflicts can be resolved there centrally, securely and unambiguously.The agents may represent evaluation instances, in particular designed as evaluation algorithms, which are in particular together part of a computer program product, as described elsewhere herein that an evaluation architecture provides on a computing device, in particular a central computing device, when it is executed on the computing device. The evaluation instances or the evaluation architecture allows at least one selected from requirements or states of at least one selected from a function (or a set of functions) or a signal transmitter (or a set of signal transmitters) to be evaluated, in particular with regard to a task to be fulfilled or to be carried out.In this case, a decision entity, in particular designed as a decision algorithm, can be provided as the hyper-agent in order to take over information from the described evaluation architecture, wherein the information originates from the two types of agents, those for evaluating the requests and / or states of the functions and those for evaluating the requests and / or states of the signal generators and transport their evaluations. In this case, the hyper-agent can be provided as part of a computer program product described elsewhere herein, which, together with the at least one agent which evaluates the function requirements and the at least one agent which evaluates the signal transmitter requirements, can form an evaluation decision architecture, a so-called hyper-agent stack or, following this, also referred to briefly as HAS.According to one aspect, the at least one function agent can be a partial agent of a function agent stack-referred to below in short as FAS-, wherein the FAS is in particular designed to map a plurality of tasks. Alternatively or additionally, the at least one signaling agent can be a partial agent of a signaling agent stack-referred to below in short as SAS-wherein the SAS is in particular designed to map a plurality of tasks.The architecture described above can thus be extended to any number of functions and / or signal generators. In this regard, the individual advantages, features, technical effects and architectural structures as described above can also be applied to the case of architectures of any desired size. Here, for the sake of convenience and compactness, repetition of these descriptions is omitted.In one aspect, at least one of the following agents may be embodied and arranged in at least one of the FAS or the SAS. The agent may be a performance agent which is designed and configured to adapt, in particular maximize, a sportiness or performance of driving. The agent may be an efficiency agent which is designed and configured to adapt, in particular optimize, an efficiency. Alternatively or additionally, it can be designed and configured to adapt, in particular minimize, consumption. The agent may be a legal agent configured and arranged to observe at least one selected from a set approval, legal regulation or homologation. The agent may be a safety agent sin, which is designed and configured to implement a safety-relevant aspect. As a result, a series of different tasks can be implemented during driving of a motor vehicle.The performance agent may attempt to generate the maximum possible performance and / or sportiness. The efficiency agent may attempt to achieve the best possible efficiency and / or the lowest possible consumption of fuel or energy. The legal agent implements, in particular, permissions such as restrictions in driving operation, laws such as with respect to fuel consumption or others, and / or homologation. Homology is in particular the grant of approval by an official authority. This may be a jurisdiction, government department, or academic or professional establishment that can operate according to certain rules or standards to determine whether such approval should be granted. An example here would be a TUV test station as a professional device, which could possibly make specifications. A safety agent implements safety-relevant aspects, such as acceleration limits, speed limits, brake accelerations, activities of safety devices, such as airbags, and others.In one aspect, the functions may include an evaluation function to be used for evaluating at least one request in communication with the ADAS. As a result, misallocatedness of evaluation functions can be avoided.Here and elsewhere, evaluation functions represent, in particular, weighting functions which can weight at least one of a request or a state in order to be able to perform an evaluation of how the request and / or the state "contracts" with a task being set. Thus, the evaluation functions allow requirements to be defined that could be implemented to perform a task. Alternatively or additionally, the evaluation functions also make it possible to classify states as to whether the task to be performed is compatible with a state currently or in the future. In this case, actuating parameters for a signal transmitter that would be required to be able to perform a task can be determined or output via evaluation functions, wherein, on the one hand, the requirements of a corresponding function and a requirement and / or a state of a signal transmitter can be included in the evaluation, in the respective agents, as described elsewhere.According to one aspect, the signal generators can have at least one selected from one of an actuator, a control variable or a control device. A plurality of different signal generators can thus be used. The term signal transmitter is to be understood here in particular in such a way that it transmits a signal via a communication connection from a motor vehicle component, which signal transmits information about at least one of a state or a request to the computing device, as is described elsewhere herein.An actuator is in particular a device which can influence a movement, an adjustment and / or a state of a motor vehicle component. For this purpose, the actuator can receive control signals and translate them into a corresponding action.During control, a control device attempts to influence the output variables by means of one or more control variables. During the regulation, a regulating device continuously compares the desired value and the actual value and tries to approximate the actual value to the desired value by changing the manipulated variable. Accordingly, the control variables can be transmitted as part of information from a signal transmitter configured accordingly, for example the control device (ECU) itself.In one aspect, the sub-agents may be arranged in at least one of a FAS or a SAS to be traversed in an ordered communication pattern. This allows a hierarchy to be introduced in assessing the requirements and / or conditions, such as to perform the various tasks represented in the agent.In this case, in particular a series can be provided in which, from a first agent to an nth agent, the requests and / or the states of the functions and / or the requests and / or the states of the signal generators run through a FAS or a SAS in a consecutive sequence.Alternatively, parallel connections can also exist, but also feedback loops, in order to be able to build up complex evaluation architectures in this way, which take into account the complexity, in particular also interacting signal generators and / or functions.Alternatively, the sub-agents may be located in at least one of a FAS or a SAS to be traversed in any communication pattern. In this case, the architecture can be designed with a particularly low level of specifications-in relation to the structure and the communication paths between the agents, which communication paths may need to be observed.In particular, the arrangements of the agents or the path of the information through them can also be designed dynamically. This also allows information to be passed on, which can be adapted to the situation.According to one aspect, the computing device can have the functions. The computing device can thus function as a storage device or it can have such a device in order to store the functions there. The functions can thus be kept ready for retrieval and are stored in particular on site in the computing device (in a part which can function as a storage device, at least for a short time).Alternatively, a pool of functions may also be maintained in a storage device, such as organized in a accessible network. As a result, the functions can be adapted independently of the computing device. Functions may also be reserved here which are not necessary for a corresponding motor vehicle, but for other motor vehicles. This also enables a fleet of vehicles to be recorded with the appropriate functions depending on the tasks. An update structure can also be implemented thereby, for example also in that an above-described local storage device in a described computing device or a local storage device which is connected to a computing device can be recorded with new functions and / or with updated functions.In one aspect, the functions may communicate with each other. Thereby, the functions can exchange information with each other, such as states of their associated signal generators, requests, and other information.The computing device can be described by the features, properties and advantages of the motor vehicle, the method and the computer program product. This also applies accordingly across the category limits of method, device and system. The motor vehicle, the method and the computer program product can thus also be described by the features, properties and advantages of the computing device. Repetition of all of these features, characteristics and advantages is omitted for the sake of convenience and compactness.According to an independent aspect, a motor vehicle can have an on-board power supply system in order to establish communication between at least one signal transmitter and at least one computing device, as described elsewhere herein. In this case, the computing device can be designed and configured to form at least one communication between functions and at least one hyper-agent stack (HAS), in particular in the form of a hyper-agent stack function (HASF), in particular in such a way as to supply each request for the functions to at least one function agent in order to carry out an evaluation of at least one of the requests for the functions. In this case, the computing device can be designed and configured to form at least one communication from at least one signal transmitter to the at least one HAS, in particular in the form of the HASF, in such a way as to supply each request and / or each state of the signal transmitter to at least one signal transmitter agent in order to carry out an evaluation of at least one of the requests and / or at least one of the states of the signal transmitters. In this case, the computing device can be designed and configured to initialize the output of a control signal to a signal transmitter in order to resolve a conflict of requirements and / or states.Initializing the control signal can be understood in particular in such a way that the computing device can transmit a signal in order to have a control signal output by a control device. In this case, it can be provided in particular that the computing device itself does not distribute any control signals in embodiments. In other embodiments, it can be provided that the computing device itself can act as a control device in order to output corresponding control signals.The motor vehicle can be described by the features, properties and advantages of the computing device, the method and the computer program product. This also applies accordingly across the category limits of method, device and system. The computing device, the method and the computer program product can thus also be described by the features, properties and advantages of the motor vehicle. Repetition of all of these features, characteristics and advantages is omitted for the sake of convenience and compactness.According to an independent aspect, a method can be designed and configured to resolve a conflict between requirements. The method comprises in particular the step of communicating between functions and at least one hyper-agent stack (HAS) in such a way as to supply requests for the functions to at least one function agent in order to perform an evaluation of at least one of the requests for the functions. The method has in particular the step of communicating between at least one signal transmitter and the at least one HAS in such a way as to supply requests from the signal transmitter and / or states of the signal transmitter to at least one signal transmitter agent in order to perform an evaluation of at least one of the requests from the signal transmitters and / or in order to perform an evaluation of at least one of the states of the signal transmitters. The method has in particular the step of resolving a conflict of requests for acting on a signal transmitter by the hyper-agent of the HAS, in particular based on the assessment of the requests and / or the assessment of the states.The method can be described by the features, properties and advantages of the motor vehicle, the computing device and the computer program product. This also applies accordingly across the category limits of method, device and system. The motor vehicle, the computing device and the computer program product can thus also be described by the features, properties and advantages of the method. Repetition of all of these features, characteristics and advantages is omitted for the sake of convenience and compactness.According to an independent aspect, a computer program product may be designed and configured to perform a method as described elsewhere when executed on a computing device, in particular as described elsewhere.The computer program product can be described by the features, properties and advantages of the motor vehicle, the computing device and the method. This also applies accordingly across the category limits of method, device and system. The motor vehicle, the computing device and the computer program product can thus also be described by the features, properties and advantages of the computer program product. Repetition of all of these features, characteristics and advantages is omitted here for the sake of compactness and readability.The computer program product may be embodied and configured to be executed on a machine. In this case, when the computer program product is executed on the machine, a method can be carried out as has been described elsewhere. The computer program product is in particular machine-readable code and / or an electrical signal which is / are configured to be read by a machine in order to transmit working instructions to a machine, such as for example in order to carry out a method of the type described elsewhere. A computer program product may be designed, in particular, as machine-readable code, in particular as an algorithm, which may enable and perform corresponding communication between agents and agent stacks.A storage medium may include a computer program product as described. Alternatively or additionally, the computing device as described can have a computer program product as described. The computer program product is designed and configured to be read from the storage medium by the computing device in order to carry out the described method. The storage medium and / or the computing device can thus be correspondingly described by the features, properties and advantages as have been described and presented for the methods and / or for the computer program product. This applies correspondingly also in the opposite sense. Repetition of all of these features, characteristics and advantages is omitted here for the sake of compactness and readability.Further advantages, features and details of the invention will become apparent from the following description, in which exemplary embodiments of the invention are described in detail with reference to the drawings. The features mentioned in the claims and in the description can be essential to the invention individually or in any combination. They show schematically: FIG. 1 shows a schematic illustration of an embodiment of a computing device; FIG. 2 shows an example conflict of requirements of two functions; FIG. 3A shows an example resolution of a conflict of requests of two functions in an HAS; FIG. 3B shows an exemplary catalog of evaluation functions; FIG. 4A shows an example with an actuator agent; and FIG. 4B shows an example with actuator failure.FIG. 1 shows a schematic representation of an embodiment of a computing device 1, which can be designed as a central computing device 2 of a motor vehicle 100. Functions A, B,..., N in the computing device 1 can be temporarily stored in a storage device 3 or permanently stored. The functions A, B,..., N can communicate with one another here, for example via an on-board power supply system 4, here in particular in the form of a DDS 5. The functions A, B,..., N can send requests and / or states to a hyper-agent stack 6 - HAS for short, which has a function agent stack 7 - FAS for short, to which the functions A, B,..., N transmit their requests and / or states, wherein the functions A, B,..., N can also send an evaluation function in order to be able to carry out an evaluation, based thereon, of the requests and / or of the states of the functions A, B,..., N in the sub-agents 13 aof the FAS 7, In particular, with respect to tasks to be performed or handled that can be mapped by the sub-agents 13 aof the FAS 7.The motor vehicle 100 has in particular a series of signal generators 16, such as actuators 17, manipulated variables 18 and / or control devices 19, so-called ECUs. These may in turn transmit requests and / or states to the HAS 6, where the requests and / or states in a signaling agent stack 14-SAS for short-may be evaluated via sub-agents 13b. The sub-agents 13 bof the SAS 14 represent the signal generators 16 and the tasks to be fulfilled by them. With respect to these tasks, an assessment of the states and / or the requirements of the signal generators 16 may take place.In summary and in other words, this means that: In the computing device 1, in particular embodied as a central computer 2, the individual functions A, B,..., N are implemented in particular. The functions can communicate all or partially among one another, as can in particular also communicate all or partially with the HAS 6, in particular configured as a hyper-agent stack function (e.g. via DDS data distribution service). Each or a selection of requests for a function runs through in particular the individual sub-agents 13 aof the FAS 7. The task of each of the sub-agents can be defined as desired, including their arrangement. In an embodiment shown here, the sub-agents 13 amay be arranged in a row to be passed through correspondingly consecutively. In other embodiments not shown here, it may be provided that random and / or dynamic arrangements are provided, as described elsewhere. The same applies in particular to the sub-agents 13 bof the SAS 14. The sub-agents 13 a, 13 bof the SAS 14 and FAS 7 may correspond to one another in the task to be mapped.Conceivable as partial agents are, for example, performance agents 9 a, 9 bthat attempt to generate the maximum performance / sportiness of a driving behavior of the motor vehicle 100 as far as possible, efficiency agents 10 a, 10 bthat attempt to enable the best efficiency / the lowest consumption as far as possible, legal agents 11 a, 11 bthat can implement set permissions / laws / homologation, safety agents 12 a, 12 bmay implement safety-relevant aspects or X agents, 8 y, 8 bthat can implement engine settings.In the individual sub-agents 13a, the requirements and states of the functions A, B,..., N are evaluated in particular. The underlying evaluation function is sent along in particular by the function, as described elsewhere herein.The signal transmitters 16, in particular designed as actuators 17, also have an agent stack, the SAS 14, which can pass its requirements. This has in particular also different layers, analogous to FAS 7.If, for example, two different functions attempt to control the same signal transmitter 16, in particular the same actuator 17, with different values, a hyper-agent 15 of the HAS 6 decides, together with the information of the FAS 7 and the SAS 14, how to resolve the conflict. By the hyper-agent 15, in particular function requirements and actuator states can be evaluated centrally, whereby conflicts can be resolved there, in particular centrally, in particular securely and in particular unambiguously.FIG. 2 shows an example conflict 50 of requirements 21, 22 of two functions A, B. Here, as a simple example, a conflict is described with respect to an actuator 17 for a rear spoiler (not shown). The function A here means, for example, "Aero Spoiler Function" with requirement 21 to arrange a spoiler position in the ECO mode, for example at a speed above 90 km / h (that is to say "kilometers per hour"). The function B here means, for example, "curve-down force function" with requirement 22 of a spoiler position in a performance operation, since lateral acceleration above 1 g (g as gravitational acceleration). This results in particular in a conflict 50 between the two requirements 21, 22.Each of the functions A, B can transmit a function value 28, in particular as a numerical value, at least one evaluation function parameter 29 and an ID 30 of the evaluation function to the HAS 6, for example in a string or another transmission form 26, 27.Each function value 28 may be individually selected for each function A, B. In this example, values may be normalized, where 0 stands for "spoiler retracted", 0.5 stands for "spoiler ECO position", and 1 stands for "spoiler performance position". This example is provided for explanation and illustration. Configurations differing therefrom are possible in this case.The evaluation function parameters 29 can represent information which is required by the evaluation functions in order to apply the signal values standardized to the evaluation functions. In this example, polynomial factors can be transmitted, for example, a1x+a2being able to become [a1, a2]. The latter can be transmitted in the form of evaluation function parameters 29.The ID 30 of the evaluation function can make it possible in the HAS 6 to identify from a wide variety of evaluation functions stored there which evaluation function is to be used in the respective partial agent 7, wherein the decision about which evaluation function is actually used, in particular itself, makes the function A, B.In particular, it can be provided that each piece of information which is transmitted to the HAS 6, starting from the functions A, B,..., N, has a function value 28, in particular as a numerical value, at least one evaluation function parameter 29 and an ID 30 of the evaluation function.FIG. 3A shows an exemplary resolution of a conflict 50 of requirements 21, 22 of two functions A, B, here again for explanation purposes the two functions A, B as shown in FIG. 2 and described in this respect, in an HAS 6. In this case, FIG. 3B shows an exemplary catalog 51 of evaluation functions 40. These may be more or less in embodiments. They can also deviate in their type and / or in their values and / or in their curves. Here, a linear-increasing evaluation function 52, a linear-decreasing evaluation function 53, a stagnant constant evaluation function 54, and a logarithmically increasing evaluation function 55 are shown.The numerical values of FIG. 3A are intended to be understood, but should not be construed or understood as being the only feasible way to implement an embodiment. As already described with reference to FIG. 2, transmission forms 26, 27 are formed which can contain the information relevant for an evaluation. This can be transmitted by the various sub-agents 13 a, wherein the various types of sub-agents 13 ain the FAS 7 can be in particular a performance agent 9 a, an efficiency agent 10 a, a legal agent 11 aand a safety agent 12 ain this case. Here, the respective values of the evaluation functions from all steps are summed in the FAS in order to be able to perform a corresponding overall evaluation in order to be able to resolve a conflict.In this case, it may be that the (aero) function A is not interested in performance, wherein the (down force) function B may have a maximum down force as target (i.e. task). Both functions A, B can correspond in particular to the effect that an efficiency target can be the minimum resistance of the moving motor vehicle. An evaluation of the legal agent 11 acan also be irrelevant for both positions. Both requirements of the functions can correspond in a safety agent 12a in that the greatest safety provides the position with the greatest down force. In this case, there is thus in particular a conflict between the two positions and the given values I, II for these positions, which conflict can be evaluated via the respective evaluation function 40 which determines the respective function A, B. Here, for example, the function A can have decided for a constant, stagnant evaluation function 54 for a performance agent 9 a, for which reason a constant value 0.5 can result as an example. The function B, on the other hand, may have decided on a linearly increasing function 52 for a performance agent 9 a, for which reason the value 1 is given here as an example. Accordingly, the further sub-agents 13a can be traversed, each issuing an assessment. These can then be summed, for example in the hyper-agent 6, which correspondingly makes the decision that the position 1 of the function B, i.e. the performance position, can be sent to the actuator 17.FIG. 4A shows an example in which a function C is intended to implement a corresponding request, i.e. to improve the view during parking, for example "extend spoilers for better camera vision during parking". By failure or due to a particular implementation, it is conceivable here as an extreme example that it may happen that this function requests that the spoiler go into a break position even though the motor vehicle is driving at 250 km / h. In this case, the spoiler is intended to drive only into a break position in order to enable better vision when the rear-view camera is parked.Here, in the actuator agent 44, it can also be implemented as a partial agent 13 bof an SAS 14 that a break position of the spoiler can be requested (or allowed) only when a braking intervention is detected.Requirements in this regard can be defined by the supplier of the actor 17 and / or the component. These can define the conditions, in particular regardless of which function A, B,..., N controls the actuator 17 (or another type of signal transmitter 16) later. This can relate in particular to limits of the component or hazardous states.Here, the actuator agent 44 can engage a Veto with the hyper agent 15 in such a case, and the function C cannot re-or enforce its requirements here.FIG. 4B shows an example with actuator 17 error 45. At speeds above 160 km / h, this function D can require the spoiler to be driven into a high-down force position. However, the actuator 16 has a fault 45, which is why a corresponding embodiment may not be possible.The actuator agent 44 receives in particular the information about the fault 45.The actuator agent 44 therefore requests in particular a safety position of the spoiler and the function N, which sets a speed limit to 90 km / h, in order to enable safe continued travel.A speed limit of 90 km / h is set by the safety-relevant error and the corresponding function N is loaded. As a result, a Veto can be placed in the hyper-agent 15 by the actuator agent 44 in this case, and the function D does not implement its request. The fault can also be reported, for example to a driver.The foregoing explanation of the embodiments describes the present invention solely by way of examples. Of course, individual features of the embodiments can be freely combined with one another, insofar as technically expedient, without departing from the scope of the present invention.

Claims

Computing device (1), in particular central computing device (2), of a motor vehicle (100), designed and configured in such a way as to - form at least one communication between functions (A, B,..., N) and at least one hyper-agent stack (HAS) (6) in such a way as to supply requests (21, 22) for the functions (A, B,..., N) to at least one function agent in order to carry out an evaluation of at least one of the requests (21, 22) for the functions (A, B,..., N); and in order - to configure at least one communication of at least one signal transmitter (16) and the at least one hyper-agent stack (HAS) (6) in such a way as to supply requests (21, 22) and / or states of the signal transmitter (16) to at least one signal transmitter agent in order to perform an evaluation of at least one of the requests (21, 22) of the signal transmitter (16) and / or in order to perform an evaluation of at least one of the states of the signal transmitter (16); wherein, in the event of a conflict (50) of requests (21, 22) for acting on a signal transmitter (16), a hyper-agent (15) of the HAS (6) decides, on the basis of the evaluations of the requests (21, 22) and / or of the states, how the conflict (50) is to be resolved.The computing device (1) according to claim 1, characterized in that the at least one function agent is a sub-agent (13a) of a function agent stack (FAS) (7), wherein the FAS (7) is configured to map a plurality of tasks; and / or wherein the at least one signaling agent is a sub-agent (13b) of a signaling agent stack (SAS) (14), wherein the SAS (14) is configured to map a plurality of tasks.The computing device (1) according to any one of claims 1 or 2, characterized in that at least one of the following agents is formed and configured in at least one of the FAS (7) or the SAS (14): - a performance agent (9a, 9b) formed and configured to adapt, in particular maximize, a sportiness or performance of driving; - an efficiency agent (10a, 10b) formed and configured to adapt, in particular optimize, and / or adapt, in particular minimize, a consumption; - a legal agent (11a, 11b) formed and configured to observe at least one selected from a set admission, a legal regulation or a homologation; or - a safety agent (12a, 12b) formed and configured to implement a safety-relevant aspect.Computing device (1) according to one of the preceding claims, characterized in that the functions (A, B,..., N) also send an evaluation function to be used for the evaluation of at least one request in communication with the FAS (7).Computing device (1) according to one of the preceding claims, characterized in that the signal generators (16) have at least one selected from one of an actuator (17), a control variable (18) or a control device (19).The computing device (1) according to any one of the preceding claims, characterized in that the sub-agents (13a, 13b) are arranged in at least one of a FAS (7) or a SAS (14) to be traversed in an ordered communication pattern, in particular in a row; or wherein the sub-agents (13a, 13b) are arranged in at least one of a FAS (7) or a SAS (14) to be traversed in any communication pattern.Computing device (1) according to one of the preceding claims, characterized in that the computing device (1) has the functions (A, B,..., N).Computing device (1) according to one of the preceding claims, in particular according to Claim 7, characterized in that the functions (A, B,..., N) communicate with one another.Motor vehicle (100) having an on-board power supply system for establishing a communication between at least one signal transmitter (16) and at least one computing device (1) according to one of the preceding Claims 1 to 6, in order - to form at least one communication between functions (A, B,..., N) and at least one hyper-agent stack (HAS) (6) in such a way as to supply each request for the functions (A, B,..., N) to at least one function agent in order to carry out an evaluation of at least one of the requests for the functions (A, B,..., N); and for - establishing at least one communication of at least one signal transmitter (16) and the at least one HAS (6) in such a way as to supply each request (21, 22) and / or each state of the signal transmitter (16) to at least one signal transmitter agent, to perform an evaluation of at least one of the requests (21, 22) and / or one of the states of the signal transmitters (16); and to initialize an output of a control signal to a signal transmitter (16) in order to resolve a conflict (50) of requests (21, 22) and / or states.A method designed and arranged to resolve conflict between requests (21, 22), comprising the steps of: - communicating between functions (A, B,..., N) and at least one hyper-agent stack (HAS) (6) in such a way as to supply requests of the functions (A, B,..., N) to at least one function agent in order to perform an assessment of at least one of the requests (21, 22) of the functions (A, B,..., N); communicating between at least one signal transmitter (16) and the at least one HAS (6) in such a way as to supply requests (21, 22) of the signal transmitter (16) and / or states of the signal transmitter (16) to at least one signal transmitter agent, in order to perform an assessment of at least one of the requests (21, 22) of the signal transmitters (16) and / or in order to perform an assessment of at least one of the states of the signal transmitters (16); - a resolution of a conflict (50) of requests for acting on a signal transmitter (16) by the hyper-agent (15) of the HAS (6), based on the assessment of the requests (21, 22) and / or the assessment of the states.Computer program product, designed and configured to perform a method according to Claim 10, when the computer program product is executed on a computing device (1) according to one of Claims 1 to 8.

Images