DE102021100426A1 - Method, device and a computer program for operating a vehicle - Google Patents

Abstract

One aspect of the invention relates to a method for operating a vehicle or a fleet of vehicles, comprising determining a probability measure for a driver error and/or a technical error in a plurality of first vehicles in connection with one or more specific first situations and/or one or several first specific roadway sections, determining a parameter of a distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first roadway sections, each of which corresponds to a specific pattern, and operating a second vehicle or a vehicle fleet depending on the determined parameter the distribution function.

Description

The invention relates to methods, devices, systems and computer programs for operating a vehicle.

In the context of the document, the term "automated driving" can be understood as driving with automated longitudinal or lateral guidance or autonomous driving with automated longitudinal and lateral guidance. The term "automated driving" includes automated driving with any degree of automation. Exemplary degrees of automation are assisted, partially automated, highly automated or fully automated driving. These degrees of automation were defined by the Federal Highway Research Institute (BASt) (see BASt publication "Research compact", issue 11/2012). With assisted driving, the driver constantly performs longitudinal or lateral guidance, while the system takes over the other function within certain limits. With semi-automated driving (TAF), the system takes over longitudinal and lateral guidance for a certain period of time and/or in specific situations, whereby the driver has to constantly monitor the system, as with assisted driving. With highly automated driving (HAD), the system takes over longitudinal and lateral guidance for a certain period of time without the driver having to constantly monitor the system; however, the driver must be able to take control of the vehicle within a certain period of time. With fully automated driving (VAF), the system can automatically handle driving in all situations for a specific application; a driver is no longer required for this application. The four levels of automation mentioned above, as defined by BASt, correspond to SAE levels 1 to 4 of the SAE J3016 standard (SAE - Society of Automotive Engineering). For example, according to BASt, highly automated driving (HAF) corresponds to level 3 of the SAE J3016 standard. Furthermore, SAE J3016 also provides SAE Level 5 as the highest degree of automation, which is not included in the BASt definition. SAE Level 5 corresponds to driverless driving, in which the system can automatically handle all situations like a human driver throughout the journey; a driver is generally no longer required.

In manual driving and in the variants of automated driving that will be marketable in the medium term, the driver must actively participate in driving the vehicle and bear responsibility for driving the vehicle. A driver error made or caused by a human driver of a vehicle is usually not predictable or explainable on the basis of logical relationships. In contrast to a technical device, at least the numerous subjective causes that can lead to an error or so-called human failure in a human being can hardly be clearly or sufficiently reliably mathematically mapped or calculated.

The object of the invention is to provide an improved method, device, system and computer product for operating a vehicle or a fleet of vehicles. In particular, the object of the invention is to positively influence the operation of a vehicle or a vehicle fleet by means of information or to protect it from the consequences of various driver errors, which is not necessarily based or does not have to be based on specific earlier consequences. In particular, the object of the invention is to enable the vehicles or the vehicle fleet to be operated in a way that is resistant or tolerant to a greater extent with regard to driver errors caused by human drivers.

The object is solved by the features of the independent patent claims. Advantageous embodiments are described in the dependent claims. It is pointed out that additional features of a patent claim dependent on an independent patent claim without the features of the independent patent claim or only in combination with a subset of the features of the independent patent claim can form a separate invention independent of the combination of all features of the independent patent claim, which can be made the subject of an independent claim, a divisional application or a subsequent application. This applies equally to the technical teachings described in the description, which can form an invention independent of the features of the independent patent claims.

A first aspect of the invention relates to a method for operating a vehicle or a fleet of vehicles, comprising determining a probability measure for a driver error and/or a technical error in a plurality of first vehicles in connection with one or more specific first situations and/or a or a plurality of first specific roadway sections, determining a parameter of a distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first roadway sections, each of which corresponds to a specific pattern. Furthermore, the method can provide the information on the basis of the determined parameter of the distribution function and/or the operation of a second vehicle or a vehicle fleet depending on the determined parameter of the distribution function or on the information provided on the basis of the parameter.

In particular, a situation is or includes a driving situation. The driving situation can, for example, be understood as a specific one caused by an arrangement, action, interaction, driving parameters of road users. In particular, a driving situation is characterized by a specific pattern, e.g. a pattern of the arrangement, speed of the road users and/or pattern of the parameters that characterize a situation. The driving situation can also be characterized by a spatial pattern of the so-called free spaces in the vicinity of the vehicle. Furthermore, the driving situation can take into account one or more parameters of the traffic rules (relevant in connection with this), e.g. traffic signs, right of way, traffic light phases. For example, the driving situation parameter(s) can be taken into account relating to or based on surroundings sensor data and/or information transmitted to the vehicle (e.g. from another road user, using a car-2-car or car-to-X). The surroundings sensor data can be data from a surroundings-sensing sensor, in particular a sensor system of at least one vehicle, that has been processed in a specific way.

• • eine (bestimmte) räumliche Verteilung der Verkehrsteilnehmer und/oder der Bewegungsparameter der Verkehrsteilnehmer, insbesondere ein Verteilungsmuster und/oder Bewegungsmuster der Verkehrsteilnehmer in der Umgebung des Fahrzeugs,
• • eine (bestimmte) räumliche Verteilung unbeweglicher Objekte in der Umgebung des Fahrzeugs,
• • eine relative Position und/oder Bewegungsparameter zu bestimmten Arten von Spurenmarkierungen, Verkehrszeichen, Ampeln (nicht notwendigerweise zu bestimmten Ampeln, etc.),
• • eine Information über die Vorfahrt des Fahrzeugs, insbesondere gegenüber bestimmten Verkehrsteilnehmern und/oder Verkehrsteilnehmern, die tatsächlich oder zumindest potentiell aus bestimmten Richtungen, z.B. einer querenden Straße rechts oder von links kommen oder kommen können,
• • eine Information zu einer, z.B. einen Grenzwert überschreitenden, Handlung eines Verkehrsteilnehmers in der Umgebung des Fahrzeugs, z.B. Hupen, Lichthupe, Drängeln, Überholen des Fahrzeugs, ein Überholversuch und dergleichen.

The driving situation and/or a parameter or a pattern of the driving situation can preferably be characteristic of:

• • a (specific) spatial distribution of the road users and/or the movement parameters of the road users, in particular a distribution pattern and/or movement pattern of the road users in the vicinity of the vehicle,
• • a (certain) spatial distribution of immovable objects in the vicinity of the vehicle,
• • a relative position and/or movement parameters to certain types of lane markings, traffic signs, traffic lights (not necessarily to certain traffic lights, etc.),
• • information about the right of way of the vehicle, in particular in relation to certain road users and/or road users who are actually or at least potentially coming from certain directions, e.g. a crossing road to the right or left,
• • Information on an action by a road user in the vicinity of the vehicle, for example one that exceeds a limit value, for example honking the horn, flashing the headlights, jostling, overtaking the vehicle, an attempt to overtake and the like.

The at least one driving situation can preferably be a driving situation that exceeds specific limit values or a driving situation that is characterized by parameters that exceed specific limit values. For example, the at least one driving situation can be a driving situation with an unwanted or dangerous approach to an object or a road user, an acceleration value that exceeds a limit value, an undesired arrangement in relation to other road users, etc. The driving situation can be a special situation (e.g. one that occurs comparatively rarely) or a dangerous driving situation, e.g. a driving situation for which an increased risk is determined or assumed.

In particular, the meaning of the term “driving situation” differs from a frequently used colloquial meaning of the term “traffic situation”. Alternatively or additionally, however, the present traffic situation can also be determined and appropriately taken into account. An identifier of the traffic situation can be read from data from a navigation system, for example, and taken into account when driving on a corresponding section of the road. In a simplified case, a should identifier, e.g. correspond to certain categories "free traffic", "heavy traffic", "slow traffic", "traffic jam", "end of traffic jam" etc.

The patterns can be specific, e.g. predefined on the basis of theory and/or patterns determined from history. Alternatively or additionally, the patterns can be patterns determined by means of a simulation. For example, patterns can be determined as a function of (local and/or a global) maxima of a probability of occurrence or a consequence of a driver error and/or a technical error.

The roadway section can be understood to mean a specific roadway section, global position of a roadway section, corresponding data of a digital map, and/or a type or pattern of the roadway section. In particular, the at least one roadway section can also be understood within the scope of the present document as a parking lot or an area on which vehicles (can) park or maneuver.

In particular, no specific driver error and/or technical error has to be included in the method or no error of a certain type can be predicted or its parameter can be predicted. Nevertheless, the information for operating the second vehicles can be determined and made available in such a way that the number of particularly probable errors is reduced.

The technical error can be an error in (or relating to) the detection and/or interpretation of the surroundings of the vehicle using means of the vehicle, implementation of an action initiated by the driver, support for a driver action, in the planning and/or execution of an action, e.g. a controlling or regulating intervention in the guidance of the vehicle. Furthermore, the technical error can relate to the vehicle signaling to the outside, e.g. by means of audible or visually perceptible signals and/or communication. The technical error does not necessarily correspond to a technical defect. Rather, this can, for example, correspond to a specific weakness and/or insufficiently precise functionality and/or a specification gap of a vehicle system or be caused by these, e.g. under certain circumstances.

Particularly preferably, probability measures for the driver error and/or for the technical error, and consequently also the corresponding distribution functions, can also be determined for driver errors that have only partially (e.g. to an uncritical extent) not been made, that have not been made and/or none have drawn consequences. In other words, probability measures for (specific) driver errors (e.g. in the respective first vehicle or determined on the basis of the data from the vehicle) can also be taken into account, even if the respective driver error only almost happened, did not happen and/or had no consequence.

The probability measures can (according to various methods) be determined from data which, for example, also identify or indicate near misses. A very rough or greatly enlarged database can thus be created in the method, on the basis of which the second vehicles can be operated. A large number of data that can be made usable as a result can result in the advantage that significantly more precise and/or more reliable information can be provided for operating a second vehicle or the vehicle fleet.

For example, depending, in particular on the basis of the parameters of the distribution functions, measures for the second vehicle or for the second driver and/or temporary changes in a property of a second vehicle, e.g driver errors or consequences that have occurred would not be possible. The method per se can be carried out more precisely and with greater tolerance for errors as a result of a greatly increased quantity of usable data.

Operating a second vehicle may (in a simplified case) include effecting certain automatic actions or measures in the at least one second vehicle.

Operating the at least one second vehicle or the vehicle fleet comprising one or more second vehicles is preferably to be understood as a, in particular dynamic, determination or adjustment of the information used during operation of the vehicles (e.g. driving operation, parking operation of the vehicles). For example, the operation can include a dynamic, specific to certain roadway sections and/or situations or roadway sections and/or situations of a certain type or pattern, determining or adjusting the parameters and/or functional logic of a system or functionality of the second vehicle.

One or more parameters of the distribution function which characterizes or represents the distribution function is preferably determined or selected for further processing. The one or more parameters of the distribution function of the probability measures for the first driver error and/or the technical error can particularly preferably characterize the so-called density, density function, shape, pattern of the distribution function and/or a measure of a deviation of the distribution function from a typical, statistical distribution , or mark normal distribution. The information based on the parameters can be provided and/or further processed in the method. Preferably, only or predominantly only selected and/or relevant parameter values of the restricted distribution function or parameters of the distribution function can be determined. Not all parameters of the distribution function have to be determined.

The method can have the advantage that explicit information about the errors made by (specific) drivers can no longer be calculated from the distribution function or from the determined parameters. As a result, the personal rights or the rights of the drivers associated with data protection cannot be endangered or protected more effectively. Furthermore, there is the advantage that very extensive data from one Many of the first vehicles can be reduced to a very compact amount of data which characterizes the distribution function or one or more parameters of the distribution function (and can therefore be processed and taken into account quickly and economically).

On the basis of one or more distribution functions for the first error, information characterizing the probability distribution function for a (e.g. not yet made) second driver error (in a second vehicle) can be determined. This can (particularly effectively) be taken into account for operating the second vehicle or the vehicle fleet.

In particular, an appropriate change in the operation, e.g. the driving operation for the vehicle or vehicle fleet, can be based on the information mentioned, e.g. on the basis of processing (also to be understood as an analysis) of the information determined, in particular one or more distribution functions of the probability measures in a backend , take place.

This can also take place if it is determined (only by means of a complex subsequent analysis in a data center) that a driver error and/or a technical error could have happened with a certain degree of probability.

Operating a vehicle can in particular include determining or adjusting an operating mode, in particular the driving mode of the vehicle and/or a mode of a system of the vehicle, or determining or adjusting a behavior of the (second) vehicle.

In particular, the operation of the second vehicle may involve putting the vehicle into a mode that does not require a high degree of cooperation or observation by the driver of the vehicle and/or a short-term lowering of the degree of automation to be carried out of the vehicle if this falls below a first predetermined threshold, degree of automation was in particular 2, and/or a brief increase in the degree of automation to be carried out for the vehicle if this was above a second predetermined threshold, in particular degree of automation 3, and/or avoiding a specific situation and/or section of the road.

The operation of a second vehicle or a fleet of vehicles depending on the determined distribution function can take place according to a specific, in particular at least partially predetermined and/or at least partially dynamically variable mathematical dependency.

This can include a mathematical dependency, according to which one or more specific actions, reactions, and/or behavior of the second vehicle is to be determined or adjusted. This can be carried out independently, in particular without controlling individual, specific actions or reactions, e.g. driving actions of the vehicle. The operation of a vehicle can include or be: weighing, deciding, executing, or refraining from an at least partially automated action of the vehicle. The behavior of the vehicle can particularly preferably be determined or adjusted for a comparatively short time interval and/or roadway location at which a driver error is predicted.

For example, depending on the at least one distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first sections of the road, adjustment information for operating a second vehicle or a vehicle fleet can be determined and (e.g. given priority to certain or potentially affected by the risk of a Vehicles affected by driver error) are provided. Such adjustment information can be information characterizing a functional parameter to be effected and/or changed in the second vehicle and/or in the vehicle fleet and/or functional logic to be effected and/or changed.

Such adjustment information can relate to specific situations and/or road sections or specific patterns corresponding to situations and/or road sections. For example, different adjustment information can be determined for different (e.g. typical, expected) patterns of the situations and/or road sections. A pattern is to be understood in particular as a data pattern formed by corresponding parameters. For example, the method can include the execution of a correspondingly configured pattern recognition.

The distribution function of the probability measures of driver errors and/or technical errors from a large number of the first vehicles and/or the adjustment information for a second vehicle or for the vehicle fleet on the basis of the corresponding parameters can be determined in advance, at least in part by means of a backend, and/or or be stored. This can result in the advantage that comparatively computationally intensive data processing can be afforded using cheap computing power that does not have to be provided in the control units of the vehicles, or at least partially offline, time-equalized and/or in reserve.

The at least one second vehicle can be designed to read in the said adjustment information provided (also to be understood as data dependent on the determined adjustment information), in particular during driving operation or parking operation, and then at least one operating parameter, in particular relating to an interaction with the driver and/or or to change at least partially automated driving or parking depending on the adjustment information, in particular according to the adjustment information. The adjustment information can explicitly or implicitly include information identifying an expected driver error or one that is to be avoided.

The operation of one or more of the second vehicles can be dependent on the one or more determined parameters, in particular according to a specific (static or dynamic changeable) mathematical relationship (e.g. parameters, logic, formula, coefficients of a formula, etc.). take place at least one distribution function. Such a mathematical relationship can, for example, be used in corresponding data processing in the backend.

For example, the adjustment information for operating the one or more second vehicles can be based on a qualitatively and/or quantitatively determined parameter (e.g. on several, finely graded intermediate values of the respective parameters from a large number of the first vehicles). A qualitative and/or quantitative (e.g. fine-grained) parameter of the adjustment information for operating the one or more second vehicles is preferably determined or adjusted.

The driving operation of a second vehicle or vehicle fleet can be changed in a qualitative and/or quantitative manner depending on the adjustment information determined or provided. For example, one of more than two or three operating modes of a system of the (second) vehicle (preferably dynamic) can be set or switched over. For example, parameter values of one or more parameters and/or logic for operating the one or more second vehicles (also to be understood as a system or functionality of the respective vehicle) are determined or adjusted in fine steps.

The operation of one or more (second) vehicles can include that the vehicles (even without actively controlling or having to control the vehicles at the moment, e.g. from the backend) with regard to their behavior, in particular driving behavior, interaction behavior, to certain situations and/or or road sections are prepared. There may be road sections and/or situations in which a driver error (a specific driver error or driver error with a specific parameter) with a distribution function based on a large number of the first vehicles with a specific parameter, in particular density, density function, shape, pattern, occurred or almost occurred has occurred and/or is at least to be expected with a certain probability in the second vehicle. This can also take place before specific situations and/or roadway sections are known in the second vehicle or can be detected by means of the vehicle.

The present method can be carried out in particular without (explicitly) predicting or having to predict a specific driver error or a technical error in a second vehicle. In particular, however, a (second) driver error can also be predicted on the basis of the determined data.

In an advantageous embodiment, the operation of the second vehicle or a vehicle fleet comprising one or more second vehicles includes determining or adapting the behavior of the second vehicle. For example, depending on the one or more parameters of the distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first roadway sections in one or more first vehicles, an adjustment information item for adjusting the behavior of the second vehicle or a vehicle fleet comprising one or several second vehicles identified and provided.

In particular, the vehicle's behavior may relate to the execution of certain automatic actions, vehicle decisions, or a proportion of vehicle actions or decisions.

For example, depending on the one or more parameters of the distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first roadway sections in the first vehicles, a parameter indicating an activation threshold, deactivation threshold, decision threshold or intensity of an automated action can be varied.

For example, the (second) vehicle can be made to act more independently and/or to take somewhat greater risks and/or that an (eg helping) action by the user of the vehicle will be necessary. For example, can the behavior of the (second) vehicle in relation to other road users can be determined or adjusted.

The behavior of a vehicle can be independent of the initiation and/or control of specific or concrete actions by the vehicle. In particular, the behavior is determined or adapted, which is independent of and/or abstracted from specific actions of the vehicle.

The behavior of the second vehicle (or a vehicle from the vehicle fleet) can be achieved by changing one or more specific thresholds (e.g. activation thresholds, deactivation thresholds, and/or decision thresholds) that apply to specific actions and/or features. This can be done with or without triggering an activation or deactivation of a specific action or a performance feature of a second vehicle. For example, the at least one second vehicle or the vehicle fleet comprising one or more second vehicles is parameterized or conditioned (e.g. using the adjustment information provided). Such a parameterization preferably relates to specific road sections and/or driving situations. The operation of the second vehicle can thus be adapted in a very targeted and precise manner.

In a further example, the adjustment information can be used to cause or change a specific behavior of one or more second vehicles, e.g. according to a behavior parameter and/or behavior pattern determined or adapted in the method. Such behavior can in particular be or include a specific behavior of the vehicle, in particular a behavior that is at least partially abstracted from specific parameters. This can also be carried out when a situation to come for the second vehicle is not known or only partially known to a small extent.

A change in the behavior of a second vehicle, in particular with regard to the driving behavior, other road users, control of the degree of automation, the driver of the vehicle can preferably be initiated. This can be a specific first change in behavior, in particular a measure of a difference (difference, factor, proportion, etc., e.g. according to a predefined condition, scale or formula). In particular, the term “behaviour” can also be understood to mean a way of acting, a way of interacting and/or a way of behaving and/or an automated driving style of the respective vehicle.

In a further advantageous embodiment, information determined on the basis of said distribution function of the probability measures of driver errors and/or technical errors is linked to the digital map and/or updated.

In particular, road section-specific and/or situation-specific and/or location-dependent (also to be understood as locally different) information is determined, assigned to the corresponding sections of the digital map (also: correspondingly linked or stored under a correspondingly assigned link or memory address), and /or read by a second vehicle when driving on the section of the digital map. In particular, the information can correspond to, include or refer to the adjustment information described in this document.

For example, the information determined, in particular adjustment information based on the parameters of the distribution function of the probability measures of driver errors and/or technical errors, can be stored, managed and/or updated in a database that is linked to the digital map, e.g. to a so-called layer of the digital map and/or linked (also to be understood as associated, e.g. linked) with a plurality of the position information. The information can also be stored within a layer of the digital map as respective locally different values.

In a further advantageous embodiment, the second vehicle is operated when driving on a section of the digital map as a function of the linked or stored information. For example, the second vehicle is operated when driving on a section of the digital map depending on the information linked to the respective roadway sections of the map based on the respective parameters of the distribution function relating to the corresponding roadway sections.

For example, the behavior of the second vehicle is influenced, preferably depending on the position and/or continuously, depending on the information provided and preferably associated with the map.

For example, one or more specific parameters of the vehicle's behavior are influenced in several (fine) stages or in an almost continuous dependence on the information.

This can take place as an alternative or in addition to initiating specific measures in connection with the specific sections of the digital map and/or with specific situations.

In a further advantageous embodiment, the one or more first driver errors and/or technical errors (in the first vehicles) are errors that could (potentially or only potentially) have led to a specific and/or significant consequence, but in particular have not led.

A probability measure for a second driver error and/or second technical errors and thus for a second consequence of a second driver error and/or technical error can also be determined as a function of the distribution function. This can result in an objective technical effect that a second vehicle can be operated, e.g. a specific measure and/or behavior of the second vehicle, depending on information that is not necessarily based on specific negative events, e.g. serious consequences of a driver error or a technical error must be based. In other words, a second vehicle can be operated in the invention based on the information about the driver errors and/or the technical errors that have remained without a significant consequence, and a probability for a second driver error, a technical error and/or a second Consequence of one or more wide errors can be significantly reduced.

In a further advantageous embodiment, the at least one probability measure for a driver error and at least one probability measure for a technical error are determined from different first vehicles or from different journeys by one or more first vehicles.

There is an objective technical effect that if a driver error has happened in some, e.g. 2% of the first vehicles and a technical error has happened in another 1% of the second vehicles, one parameter can (already) be varied when operating a second vehicle . This can prevent a (possibly extremely serious) consequence that a (human) driver error and a technical error occur in the second vehicle.

The (next) second vehicles can be operated, in particular change their behavior appropriately, depending on the (individually) harmless errors or errors that (again by a happy coincidence, fortunately) have not led to any consequence in the other (first) vehicles.

In other words, it is also possible to avoid extremely rare things or events that have rarely or never happened up to now (e.g. in the second vehicles and/or in the first vehicles).

In a further advantageous embodiment, the at least one probability measure and/or a distribution function of the probability measures relates to a combination of a driver error and/or a technical error.

Extremely rare events, which include a combination of a human and/or technical error or a chain of events, which are based, for example, on a combination of at least one human error and at least one technical error (also to be understood as a so-called " unfortunate chain of events”) can be avoided or its consequences reduced.

In a further advantageous embodiment, a technical error relates to a driver assistance system or a system for at least partially automated driving and/or parking and includes a missing, exaggerated, understated and/or incorrect action of the system.

In a further advantageous embodiment, the one or more driver errors, one or more technical errors and/or a combination of a driver error and a technical error correspond to a specific criterion or pattern. In other words, a pattern (e.g. by means of machine learning) can be learned and/or (re)recognized, which includes both data on one or more driver errors and data on one or more technical errors. The second vehicle can then be operated depending on the pattern determined (using the data from the first vehicle).

Alternatively or additionally, a pattern from a driver error, a technical error, a situation and/or a road section can be recognized and taken into account.

Data identifying driving errors and/or technical errors is particularly preferably determined (also to be understood: collected) and takes into account a specific consequence (also to be understood as a consequence of a specific type) or a would entail consequences exceeding a certain level.

For example, data relating to the driver of several first vehicles overlooking an entrance is collected. These are only or also collected or relate only or also to cases when no road user has come out of the entrance at the given time or when a driver assistance system has averted the consequences of the oversight.

When operating a second vehicle, these can still be taken into account to advantage.

The driver error can in particular also be a driver error which (at least in these cases) has not led to a (serious) consequence. For example, the driver's error may involve overlooking another road user and/or reacting too late or inappropriately to a situation that could have led to a specific (e.g. essential, serious) consequence, but did not. In particular, the driver error relates to a near miss or a potentially possible near miss.

In a further advantageous embodiment, the operation of the second vehicle is determined or adapted depending on a second situation and/or a second road section, with in particular a degree of similarity between one or more first situations and second situations and/or a degree of similarity between a second road section and the first Road section is determined and taken into account.

For example, the data described above relating to a driver error and/or a technical error, for example at an entrance, can also be used (correspondingly) for similar entrances (e.g. identical in construction, having similar dimensions and/or proportions) at other locations.

In a further advantageous embodiment, one or more parameters of the boundary conditions and/or context are determined and taken into account in connection with the occurrence of the at least one driver error in one or more first driving actions.

One or more parameters of the boundary conditions that precede the occurrence of the driver error and/or parameters of the boundary conditions are present essentially during the occurrence of the driver error are determined and taken into account.

In particular, parameters of a driver action, an operating parameter or an action of the driver assistance system or the vehicle system for automated driving and/or the corresponding parameters of the boundary conditions and/or the context can be stored in a memory, e.g. a ring memory or a comparable memory structure and provided and/or further processed depending on the occurrence or detection of a driver error and/or technical error. The data can also be determined afterwards. Thus, the corresponding data does not have to be recorded constantly or frequently. The method can be made very economical to carry out.

In particular, one or more parameters of the boundary conditions, preferably a pattern of parameters of the boundary conditions, are determined and taken into account in connection with the occurrence of the driver error in one or more first operator actions.

The data can preferably be determined, compared and taken into account on the basis of the parameters of the boundary conditions in a large number of the first usage processes (e.g. journeys with the first vehicles).

• • Bewegung, insbesondere ein Bewegungsmuster des jeweiligen Fahrzeugs.
• • Eine Bediensituation, z.B. Aktionen die vorher kamen und/oder Aktionen die (zumindest wahrscheinlich) folgen würde.

One or more parameters of the boundary conditions can be representative:

• • Movement, in particular a movement pattern of the respective vehicle.
• • An operating situation, eg actions that came before and/or actions that would (at least probably) follow.

Furthermore, a temporal and/or spatial change characteristic, in particular a gradient, for example a temporal and/or spatial gradient of a parameter or parameter value, can also be a parameter or be taken into account as a parameter.

In a further advantageous embodiment, one or more parameters of the consequence of at least one driver error in relation to the one or more second driving actions, in particular several possible variants of an imminent driving action by the driver and/or the driver error, are determined and taken into account.

The at least one change in the operation of the second vehicle is preferred, in particular a specific measure, change in behavior, reaction, etc. in connection with a road section and/or a situation, depending on one or more parameters a corresponding consequence, which is in particular predicted in the procedure.

Depending on a predicted consequence, it is possible, for example, to react earlier and/or with other measures. For example, if the driving action is carried out in good weather, good road condition, good visibility or audibility, a different measure is taken or a measure is taken at a different time than if the driving action was performed in bad weather, poor road condition, poor visibility or .Audibility pending or running. For example, one of at least two different measures or different types of measures in connection with the at least one second driving action is selected depending on the data based on one or more parameters of the consequences of one or more first driving actions. In other words, a consequence of a specific driver error can be predicted as a function of the data from one or more first operating processes.

The data can preferably be determined and taken into account on the basis of the parameters of the consequence of a large number of the first driver errors, in particular by means of an aggregation.

For example, statistical accumulations of occurrences of certain consequences or consequences of driver errors characterized by certain parameters of a certain type, in particular in the case of one or more first operating processes, situations, contexts, etc. corresponding to a certain criterion, e.g. patterns, are determined and taken into account.

Information on the number or frequency of the occurrence and/or non-occurrence of at least one (in particular a specific) driver error is determined after and/or approximately at the same time as the driver error occurs from a large number of the first operating processes. The at least one measure can then be determined during a second trip of the at least one second vehicle depending on the information on a number or frequency of occurrence of the driver error (in the multiplicity of first vehicles). For example, information is determined that characterizes the number or frequency of the occurrence of a driver error, a specific driver error or a driver error of a specific type in a large number of first uses (e.g. trips or parking or maneuvering operations) in connection with certain parameters of the boundary conditions, for example certain patterns of the parameters of the boundary conditions.

Alternatively or additionally, the number or frequency of the non-occurrence of a driver error, a specific driver error or a driver error of a specific type in a large number of first uses in connection with specific parameters, in particular the same parameters of the boundary conditions, for example specific patterns of the parameters of the boundary conditions.

In a further advantageous embodiment, one or more parameters of a second driving action are determined, and at least one similarity indicator is determined with at least one or more first parameters of one or more first driving actions and taken into account in the method.

One or more patterns of the parameters of a second driving action are particularly preferably determined and at least one similarity indicator is determined with at least one or more parameters of a first driving action and taken into account in the method. In an example that can be combined with all features of the invention, a combination and/or a pattern of the parameters that characterize a position and/or movement profile of the driver can be determined and taken into account.

• • Ermitteln oder Prädizieren einer zweiten Situation eines zweiten Fahrzeugs und/oder einer zweiten Fahrbahnabschnitts der von dem zweiten Fahrzeug befahren wird,
• • Ermitteln, insbesondere aus einem Backend, einer Information kennzeichnend eine Verteilungsfunktion der Wahrscheinlichkeitsmaße der Fahrerfehler und/oder der technischen Fehler auf erste Situationen und/oder erste Fahrbahnabschnitte die jeweils einem Muster entsprechen, der zumindest ein gewisses Ähnlichkeitsmaß mit der zweiten Situation und/oder mit dem zweiten Fahrbahnabschnitt (auch zu verstehen: mit dem Muster der zweiten Situation und/oder des zweiten Fahrbahnabschnitts) aufweisen, und daraufhin,
• • Betreiben des zweiten Fahrzeugs abhängig von der ermittelten Information.

A second aspect of the invention relates to a method for operating a vehicle or a vehicle fleet, comprising:

• • determining or predicting a second situation of a second vehicle and/or a second section of road on which the second vehicle is driving,
• • Determination, in particular from a backend, of information characterizing a distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first road sections, each of which corresponds to a pattern that has at least a certain degree of similarity with the second situation and/or with the second lane section (also to be understood: with the pattern of the second situation and/or the second lane section), and thereupon,
• • Operating the second vehicle depending on the determined information.

This can (also from the visual perspective of the first aspect of the invention) be the second vehicle or the vehicle fleet comprising second vehicles. The method preferably comprises ren the features according to the first aspect and according to the second aspect of the invention.

A third aspect of the invention relates to a device for operating a vehicle or a fleet of vehicles, comprising means for carrying out the method, in particular according to one of claims 1 to 14.

This can be a device of the first vehicle, the second vehicle or a device arranged at a distance (e.g. stationary) from the first or second vehicles. This latter device may, for example, comprise a backend for operating the first and second vehicles, respectively.

A fourth aspect of the invention relates to a system for operating a vehicle or a fleet of vehicles, comprising means for carrying out the method, in particular according to one of claims 1 to 14.

The system preferably includes one or more devices in the plurality of first vehicles, a backend for carrying out data transmission and/or data processing, and one or more second devices in the one or more second vehicles. In particular, the system is set up to operate a second vehicle or a fleet of vehicles comprising one or more second vehicles according to one or more features of the method.

A fifth aspect of the invention relates to a computer program, in particular a computer program product comprising a computer program comprising instructions which, when the program is executed by a computer, cause the latter to execute the method, in particular according to one of claims 1 to 14.

The computer program product can be embodied as an update of a previous computer program, which includes the parts of the computer program or the corresponding program code, for example as part of a functional expansion, for example as part of a so-called “remote software update” and/or a campaign to collect data from a vehicle . In particular, the computer program product includes at least one encrypted file. For example, the computer program includes an authorized and/or encrypted right of access to certain data (e.g. from internal interfaces, vehicle electrical system communication, etc.) or computing devices of a vehicle. The invention is described in more detail below with reference to exemplary embodiments and figures, without restricting the generality. All features of the exemplary embodiments can be combined with one another and/or with the features of the claims.

the 1 illustrates several (first) vehicles 10, 11, 12, ... 1x, at least one second vehicle 20, 21, 22, 2x, and a backend BE. The preferred data connections are shown as arrows.

The vehicles each include a (first) device V10 or the (second) device V20, and (not shown in the figure) each V11, V12, ... V1x or V21, V22, ... V2x. The first device V10, V11, V12, first device V10, V11, V12, ... V1x, or as the second device V20, V21, V22, ... V2x. Alternatively, the first device V10, . . . and the second device V20, . . . can be installed in the same vehicle 10, . This can be or include the device according to the applicable device claim.

The first device V10, . . . is designed to ascertain one of the items of information described in this document (from the respective first vehicle 10, . These first vehicles 10, 11, 12, ... 1x or the devices V10, V11, V12, ... Vx of the first vehicles are set up, e.g. by means of an exterior sensor system and/or interior sensor system (any sensor system operated in the interior of the vehicle ) to recognize the first errors of the respective first driver and/or to determine parameters of the corresponding situations, sections of road, boundary conditions, contexts that applied shortly before, during or shortly after the driver error.

• - der Fahrer zu spät (wesentlich später als eine typische bzw. gebotene Reaktionszeit) auf ein (ursächliches, eine Reaktion erforderndes) Ereignis reagiert hat; und/oder
• - der Fahrer nicht situationsadäquat (vereinfacht: übertrieben oder untertrieben bzw. in einer für die Situation nicht zweckmäßiger Art und Weise) gehandelt und/oder reagiert hat; und/oder
• - der Fahrer seine eigene fahrerische Handlung (z.B. nach einer kurzen Zeit) korrigiert hat; und/oder
• - einem bestimmten Muster entsprechende Situation und/oder eine riskante im Zusammenhang mit einer fahrerischen Handlung (auch zu verstehen: weil eine entsprechende fahrerische Handlung gefehlt hat) entstanden ist; und/oder
• - eine ungewöhnliche (z.B. keinem der vorbekannten und/oder vorausbestimmten Mustern entsprechende) Situation entstanden ist; und/oder
• - eine Reaktion eines anderen Verkehrsteilnehmers (z.B. eine Reaktion einer bestimmten Art, mit einem bestimmten Parameter, und/oder entsprechend eines bestimmten Musters) ermittelt wurde; und/oder
• - Signalisierung bzw. Kommunikation von einem anderen Verkehrsteilnehmer. Beispielsweise wird erkannt, dass das Fahrzeug von einem anderen Verkehrsteilnehmer angehuppt und/oder angeleuchtet wird;

A measure of the probability of a first driver error can be determined dependent on the detection that:

• - the driver reacted too late (significantly later than a typical or required reaction time) to a (causal, reaction-requiring) event; and or
• - the driver has not acted and/or reacted appropriately to the situation (simplified: exaggerated or understated or in a way that is not appropriate for the situation); and or
• - the driver has corrected his own driving action (e.g. after a short time); and or
• - a situation corresponding to a certain pattern and/or a risky situation in connection with a driving action (also to be understood: because a corresponding driving action was missing) has arisen; and or
• - an unusual situation (e.g. not corresponding to any of the previously known and/or predetermined patterns) has arisen; and or
• - a reaction of another road user (eg a reaction of a certain type, with a certain parameter, and/or according to a certain pattern) was determined; and or
• - Signaling or communication from another road user. For example, it is recognized that the vehicle is hitched up and/or illuminated by another road user;

• - ein Wahrnehmung, Einschätzung und/und Interpretation der Situation, insbesondere der Fahrsituation durch den Fahrer;
• - eine Fahrzeugführung, insbesondere eine sogenannte Bahnführung des Fahrzeugs, z.B. Lenken, Bremsen, Beschleunigen;
• - eine steuernde Aktion oder Eingriff z.B. durch ein Pedal, Lenkrad des Fahrzeugs,
• - eine Steuerung eines Fahrerassistenzsystems, z.B. ein Veranlassen, Verändern, Überstimmen oder Beenden einer durch das Fahrerassistenzsystem zumindest teilweise automatisiert auszuführenden Aktion;
• - Veranlassen einer Signalisierung, z.B. Hupen, Lichtsignalisierung, an die Umgebung des Fahrzeugs, z.B. an andere Verkehrsteilnehmer;

For example, driver error concerns:

• - A perception, assessment and/and interpretation of the situation, in particular the driving situation, by the driver;
• - A vehicle guidance, in particular a so-called path guidance of the vehicle, eg steering, braking, acceleration;
• - a controlling action or intervention, e.g. by a pedal, steering wheel of the vehicle,
• - a control of a driver assistance system, for example initiating, changing, overruling or ending an action to be carried out at least partially automatically by the driver assistance system;
• - Initiating a signaling, eg honking, light signaling, to the environment of the vehicle, eg to other road users;

A driver's action can be understood as a specific action pattern or sequence pattern of actions (simplified also an action sequence) of the driver and/or execution or initiation of a part of a maneuver. A driver error can, for example, also be recognized by a parameter that deviates from the norm, a typical or statistical value, e.g. in the driver's action of the type at hand, that is unexpected and/or unusual.

The degree of probability can, for example, be determined, encoded or processed as 2, 4, 8, 16, 32... value ranges or gradations.

A driver error can also be recognized by a further (e.g. corrective or compensatory or apologetic) driver's action or its consequences or by a subsequent action and/or communication by the driver. For example, it can be recognized that the result of the driver's action did not correspond to the expectations of the driver or another road user.

The backend BE is designed to read in or call up, process and expediently aggregate the information identifying the probability measures of the first driver errors and/or technical errors that are determined in a large number of the first vehicles V10, V11, V12, . . . V1x. In this case, parameters of one or more distribution functions of the probability measures are determined. The information to be provided for the at least one second vehicle or the vehicle fleet can be determined (as the result of the processing). This can be or include the adjustment information for operating one or more second vehicles V20, V21, V22, . . . V2x. The adjustment information can be provided as required.

The second device V20, .

The second device V20, to change the properties or the behavior of one or more systems of the vehicle 20, 21, 22, ... 2x.

The one or more first devices V10, ..., the at least one second device V20, ... and the backend BE are e.g or to form a fleet of vehicles and/or to function as part of the system for operating a vehicle or fleet of vehicles. A number (for example: a subset) of the vehicles in a customer fleet that are currently active or in operation can be viewed as a vehicle fleet.

Each of the described devices V10, V11, V12, V1x, ... BE, V20, V21, V22, V2x can be used to set up and/or operate a (based on data or Data exchange based, information technology) operative connection with one or more of the other devices V10, V11, V12, V1x, ... BE, V20, V21, V22, V2x described in this document. The method is made executable according to one or more features identified in this document.

Nevertheless, the invention also includes other design variants of the device or the system. For example, in a simplified variant of the invention, the method can be carried out with only one vehicle. In particular, this can act as the first vehicle 10 or the second vehicle 20 at different time intervals.

• • Ermitteln jeweils eines Wahrscheinlichkeitsmaßes für einen Fahrerfehler und/oder einen technischen Fehler in einer Vielzahl von ersten Fahrzeugen 10, 11, 12, ... 1x im Zusammenhang mit einer oder mehreren bestimmten ersten Situationen und/oder einem oder mehreren ersten bestimmten Fahrbahnabschnitten,
• • Ermitteln eines Parameters der Verteilungsfunktion der Wahrscheinlichkeitsmaße der Fahrerfehler und/oder der technischen Fehler auf erste Situationen und/oder erste Fahrbahnabschnitte die jeweils einem bestimmten Muster entsprechen, und ausgestaltet zum
• • Betreiben eines zweiten Fahrzeugs 20, 21, 22, ... 2x oder einer Fahrzeugflotte 10, 11, 12, ... 1x, ... 20, 21, 22, ... 2x abhängig von dem ermittelten Parameter der Verteilungsfunktion.

The system is designed for:

• • Determining a probability measure for a driver error and/or a technical error in a large number of first vehicles 10, 11, 12, ... 1x in connection with one or more specific first situations and/or one or more first specific road sections,
• • Determining a parameter of the distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first roadway sections, each of which corresponds to a specific pattern, and designed for
• • Operating a second vehicle 20, 21, 22, ... 2x or a fleet of vehicles 10, 11, 12, ... 1x, ... 20, 21, 22, ... 2x depending on the determined parameter of the distribution function.

• • Ermitteln oder Prädizieren einer zweiten Situation eines zweiten Fahrzeugs 20, 21, 22, ... 2x und/oder einer zweiten Fahrbahnabschnitts der von dem zweiten Fahrzeug 20, 21, 22, .. 2x befahren wird,
• • Ermitteln, insbesondere aus einem Backend BE, einer Information kennzeichnend eine Verteilungsfunktion der Wahrscheinlichkeitsmaße der Fahrerfehler und/oder der technischen Fehler auf erste Situationen und/oder erste Fahrbahnabschnitte wobei die ersten Situationen und/oder Fahrbahnabschnitte jeweils einem Muster entsprechen, und das Muster zumindest ein gewisses Ähnlichkeitsmaß mit der zweiten Situation und/oder mit dem zweiten Fahrbahnabschnitt oder einem Muster der zweiten Situation und/oder des zweiten Fahrbahnabschnitts aufweist, und daraufhin,
• • Betreiben des zweiten Fahrzeugs 20, 21, 22, ... 2x abhängig von der ermittelten Information.

Alternatively or additionally, the system is designed for:

• • determining or predicting a second situation of a second vehicle 20, 21, 22, .
• • Determining, in particular from a backend BE, information characterizing a distribution function of the probability measures of driver errors and/or technical errors on first situations and/or first roadway sections, the first situations and/or roadway sections each corresponding to a pattern, and the pattern at least one has a certain degree of similarity with the second situation and/or with the second section of the road or with a sample of the second situation and/or the second section of the road, and thereupon,
• • Operating the second vehicle 20, 21, 22, ... 2x depending on the information determined.

Said information (to be provided, made available) can, for example, correspond to the customization information described in this document, include it or be the basis for determining the customization information.

Furthermore, the system can include a digital map, in particular a so-called layer of a digital map, which is managed in particular in the backend BE and/or can be read in by the second vehicles V20, . For example, respective data sections of the adjustment information that apply to the sections of the map that represent certain roadway points are assigned (linked to) them. These can be read in and taken into account by means of the respective second vehicle 20, .

the 2 FIG. 12 illustrates a distribution function of driver error probability measures. In one example, the distribution of the probability measures of the driver errors can characterize a distribution of the probability measures of the driver errors as a function of at least two different parameters of the situation PS1, PS2, in particular the driving situation, alternatively or additionally to parameters characterizing features of the roadway points PF1, PF2. In this case, the respective values can also be reciprocal values, coded and/or pre-processed values. Alternatively or in addition to that in the 2 In the example shown, the distribution function can identify a dependency on the parameter for a feature of the roadway section and/or on a position of the roadway point or position of the respective vehicle within a roadway point. In the context of the present document, the term "identify" is to be understood in particular as representing or representing a respective characteristic property. The distribution function is preferably a distribution function of the probability measures with two, three or more than three dimensions.

In particular, a parameter of the distribution function characterizing the density, density function, form (also form factor) of the distribution function can be determined and processed in the method. In this case, the operation of the second vehicle can be dependent on the parameter characterizing the respective situation, in particular the current road section and/or the road section to be driven on and/or the detected and/or related situation Density, density function Form of the respective distribution function.

Furthermore, in the 2 symbolically different threshold values S1 and S2 marked. A change in the operation, for example a behavior parameter, of a second vehicle takes place depending on whether the at least one threshold value S1, S2 is reached or exceeded.

The data processing in the procedure is to be understood as fully or partially computer-aided. This can be fully automated, largely real-time capable and/or at least partially involve human involvement, in particular a developer or dispatcher.

For example, at least one step of the method, in particular relating to the steps that can be carried out in the backend BE, can be carried out or influenced by people by means of a development service, decision, adjustment, parameterization of the vehicle fleet. For example, an operating option for people (outside the vehicles) can be provided in the system or in the method to monitor, influence, evaluate and/or regulate at least one step of the method, e.g. the determination of the adjustment information. Furthermore, the data collected or provided in the method can be used to design, design and/or further develop a system or functionality relating to individual mobility.

As a result, vehicles or vehicle fleets can be made more resistant or more tolerant, so to speak, "vaccinated" against a possible or probable occurrence of certain driver errors, or driver errors that are characterized by certain parameters (values), in particular patterns, or combinations of driver errors and technical errors. will.

Claims (17)

A method of operating a vehicle or fleet of vehicles, comprising: • Determining a probability measure for a driver error and/or a technical error in a large number of first vehicles in connection with one or more specific first situations and/or one or more first specific road sections, • Determining a parameter of a distribution function of the probability measures of the driver error and/or the technical error in first situations and/or first roadway sections, each of which corresponds to a specific pattern, and • Operating a second vehicle or a fleet of vehicles depending on the determined parameter of the distribution function. procedure after claim 1 , wherein operating the second vehicle includes determining or adjusting the behavior of the second vehicle. Method according to one of the preceding claims, wherein information determined on the basis of the parameters of the distribution function of the probability measures of driver errors and/or technical errors is linked to a digital map and/or dynamically updated. Method according to one of the preceding claims, wherein the second vehicle is operated when driving on a section of the digital map depending on the information linked to the respective lane sections of the map on the basis of the respective parameters of the distribution function relating to the lane sections. Method according to one or more of the preceding claims, wherein the one or more driver errors and/or technical errors in the first vehicles are errors which could have led to a specific and/or significant consequence, but in particular did not lead. Method according to one or more of the preceding claims, wherein the at least one probability measure for a driver error and at least one probability measure for a technical error are determined from different first vehicles or from different journeys by one or more first vehicles. Method according to one or more of the preceding claims, wherein the at least one probability measure and/or a distribution function of the probability measures relates to a combination of a driver error and/or a technical error. Method according to one or more of the preceding claims, wherein a technical error relates to a driver assistance system or a system for at least partially automated driving and/or parking and includes a missing, exaggerated, understated and/or incorrect action of the system. Method according to one or more of the preceding claims, in which the one or more driver errors, one or more technical errors and/or a combination of a driver error and a technical error conforming to a specific criterion or pattern. Method according to one or more of the preceding claims, wherein the operation of the second vehicle is determined or adapted depending on a second situation and/or second road section, in particular a degree of similarity between one or more first situations and second situations and/or a degree of similarity between one second lane section and the first lane section is determined and taken into account. Method according to one or more of the preceding claims, in which one or more parameters of the boundary conditions and/or context are determined and taken into account in connection with the occurrence of the at least one driver error in one or more first driving actions. Method according to one or more of the preceding claims, wherein the one or more parameters of the consequence of at least one driver error in relation to the one or more second driving actions, in particular several possible variants of an impending driving action by the driver and/or the driver error, are determined and taken into account . Method according to one or more of the preceding claims, wherein one or more parameters of a second driver's action are determined, and at least one similarity indicator with at least one or more first parameters of one or more first driver's actions are determined and taken into account. A method of operating a vehicle or fleet of vehicles, comprising: • determining or predicting a second situation of a second vehicle and/or a second section of road on which the second vehicle is driving, • Determination, in particular from a backend, of information characterizing a distribution function of the probability measures of driver errors and/or technical errors in first situations and/or first road sections, each of which corresponds to a pattern that has at least a certain degree of similarity with the second situation and/or with have the second lane section, and then, • Operating the second vehicle depending on the determined information. Device for operating a vehicle or a fleet of vehicles, comprising means for carrying out the method according to one of Claims 1 until 14 . System for operating a vehicle or a fleet of vehicles comprising means for performing the method according to a Claims 1 until 14 . Computer program, comprising instructions which, when the program is executed by a computer, cause the latter to carry out the method according to one of Claims 1 until 14 to execute.

Images