EP4268162A1

EP4268162A1 - Method and system for transforming a start object situation into a target object situation (intuitive tacit solution finding)

Info

Publication number: EP4268162A1
Application number: EP21847502.8A
Authority: EP
Inventors: André STUTH
Original assignee: Individual
Current assignee: Individual
Priority date: 2020-12-24
Filing date: 2021-12-24
Publication date: 2023-11-01
Also published as: WO2022136703A1; US20240053753A1; LU102362B1

Abstract

Automatedly finding the solution to a problem by gathering and utilizing procedural event knowledge, comprising: defining a task statement, consisting in transforming a start object situation into a target object situation by means of an event sequence; searching a database for a suitable solution in the form of an event sequence, in accordance with an association of the at least the following variables with each other: identifiers of possible start and target object situation and of the object types involved and information regarding an event sequence, the event sequence being suitable for transforming the possible start object situation into the possible target object situation; selecting a suitable procedural event sequence in the database or forming a new event sequence, as an n-concatentation from the available event sequences.

Description

Method and system for converting a start object situation into a target object situation (Intuitive Tacit Solution Finding)

technical field

The present invention relates to a control method for controlling an actuator in order to convert a start object situation into a target object situation, with this process being able to be represented by event sequences which can be generated and which can be retrieved from a database in order to optimize it. Furthermore, the invention relates to a control system, comprising a computer program product for controlling an actuator in order to implement the control method, an industrial robot system comprising the control system, a vehicle guidance system comprising the control system, a traffic control system, a device and a method for robot-controlled process optimization, a method for normalization of object types and a method for normalizing object situations, with all of the listed systems or devices implementing the control method according to the invention.

State of the art

Control systems, in particular also so-called adaptive control systems, are known from the prior art. These are colloquially referred to as controls. In the following, the term control system is replaced by "control" for reasons of readability. Controls and regulations can monitor a wide variety of processes and, if necessary, adapt them.

Adaptive control is known, which is set up to constantly adapt certain operational parameters to changing conditions in order to achieve the best performance of the process. For this adjustment, a finite chain of actions in a time period is necessary, which produces the desired or required TARGET object situation from the received sensor data of the analyzed ACTUAL object situation in a certain spatial section with the help of the variables to be controlled by the adaptive control.

Every nth action converts the nth start object state S _n into the nth target object state Z _n . By stringing together (concatenation) m actions, the first start object state Si becomes transferred to the m-th target object state Z _m . Since the target-object situation is not necessarily changed exclusively via the control, i.e. action is not necessarily required, the term event is used below instead of the term action. So an nth event transforms the nth start object state S _n into the nth target object state Z _n .

A target variable is measured continuously for the adjustment, ie the current object status (the ACTUAL object situation) is recorded, while a target status (TARGET object situation) is specified for the control. In this way, the controller observes the changes in the target variable caused by a change in the input signals and then adapts its own behavior accordingly.

However, such controls often consume a lot of energy, do not work as efficiently as possible and valuable process time is not fully utilized or wasted on extensive and complex calculations.

This is particularly the case when sometimes very complex, non-linear systems are controlled, so that in the event of an instability of the complex system, even the smallest changes in the input parameters can be accompanied by significant changes in the target variable. In addition, complex systems sometimes have so-called limit cycles, bifurcations or chaotic behavior instead of achievable fixed points.

Therefore, some adaptive controllers have reference models of the system controlled by the adaptive controller, which they use to make predictions about the controlled system. Correspondingly, it is difficult for these controllers to react to states outside of the reference model. Furthermore, often only local minima are subsequently found during an optimization. It should be noted that there are different types of controls. Adaptive controls, supported by machine learning (ML) or statistical methods, are increasingly being developed, particularly for complex systems. These methods of control are also referred to as "intelligent control".

Adaptive controls based on statistical methods such as Bayesian probabilities require little background information about the controlled system, but are inefficient, which means that they do not necessarily find optimal solutions, and in doing so they require a lot of time and energy.

Adaptive controls, which are supported by ML, on the other hand, are quick and efficient in execution. This can be well illustrated using the example of a neural network: A neural network uses a number of input parameters (input) and one or more target parameters (output). Between the input and output layers are interconnected layers of "neurons". These represent a non-linear function in which the input values on the preceding layers are processed and combined to produce an output value. The system is then operated with a variety of possible input data trained to generate the correct output This requires a corresponding amount of initial training data, which must be sufficient to cover the relevant part of the phase space (n-dimensional space of all possible states of the system, where n corresponds to the number of measured parameters). and should also have as little statistical distortion as possible. During online or offline training, the so-called weightings, i.e. the strengths of the connections between the individual neurons, are varied until the neural network delivers the expected results. The "knowledge" about the domain, in which the neural network is applied thus implicit in the structure of the neural network.

While the training is very computationally intensive and therefore requires a large amount of training data and a lot of energy and time as well as powerful hardware, the use of the generated neural network for the assignment of a set of input values to a number of output values is comparatively less computationally intensive and can therefore also be done on less powerful hardware with a manageable expenditure of energy and time.

In both of the latter cases, the known adaptive controls lack the ability to adapt when the boundary conditions of the controlled system change, so that the phase space or the stabilities in the system also change.

What both options also have in common is that after the system has been set up for the first time, it is only possible with great effort to make changes to the controller, especially if it is in ongoing operation.

In the case of adaptive control, it is known that parameters can be adapted to the process. However, the complexity of the structure does not increase. The disadvantage is that new, more complex problems can therefore often not be solved or not optimally or only insufficiently. Even after longer waiting times (e.g. calculation and search times) with a high expenditure of time and energy, there is in many cases no new and optimal solution to the problem.

Frequently recurring problems are constantly repeated - regardless of possible complexity - only under the previously known constant effort and energy consumption. A disadvantage of the prior art is that the controls have a poor ability to learn. Classic controls, such as adaptive controls, are very common. A complete

Replacement of these is very expensive in many cases. In addition, classic controls in their specific design are often only adapted to their respective application.

task

The invention has set itself the task of providing an optimized, i.e. more efficient, more energy-saving, faster and/or better structured control method and to design this in such a way that the control method can also be “retrofitted” or “added” to existing control systems.

In particular, the invention recognizes the need for a high degree of abstraction, so that the methods and devices according to the invention are suitable for all purposes of use, in particular for interaction with all classic controls, i.e. independent of the purpose of use.

solution

The task is essentially solved by adding a knowledge management system to a control method, which includes event sequences in order to convert a start object situation into a target object situation.

The present object is achieved in particular by a control method for controlling an actuator for transfer, in particular for transferring the actuator from a start object situation to a target object situation (i.e. reaching a target object situation based on a start object situation) by means of a controller, preferably an adaptive control, comprising a) determining (SOI) a starting object situation by means of a sensor, b) defining (S02) a target object situation, c) determining (S03) an event sequence which is suitable to convert the starting object situation into the To transfer the target object situation from a set of known (partial) event sequences, in particular partial event sequences (i.e. partial procedures; individual process steps).

Iterative search (S03a) of known (partial) event sequences, in particular

Partial event sequences, comprising the start and/or target object situation and/or object situations from partial event sequences of previous iteration steps,

Selecting (S03bl) at least one event sequence for reaching the target object situation based on the starting object situation or forming (S03b2) a new event sequence for converting a starting object situation into a target object situation based on the partial event sequences found in the iterative search of the method and their concatenations, d) controlling (S04) the actuator based on the determined sequence of events by the controller.

It is clear to the person skilled in the art that in order to define (SOI) a task, which consists of converting a specific start-object situation into a specific target-object situation using an event sequence, a start-object situation must first be determined using a sensor and then a (desired) target object situation is specified and based on this a task for the transfer of the object from the specific start object situation to the specific target object situation is derived.

According to the invention, the determination then takes place from a set of known (partial) event sequences of an event sequence that is suitable for reaching the target object situation based on the start object situation. For this purpose, a database is preferably used, which for the purpose of searching for a suitable solution in form of an event sequence, which is suitable for solving the task, is read. The database is preferably suitable here for allocating at least the following variables to one another, in particular also for dealing with a weighted combination of these variables: an identifier for a possible starting object situation,

identifiers of the object types involved,

Identifier of a possible target object situation,

Identifier of the object types involved, information on an event sequence, the event sequence being suitable for converting the possible start object situation into the possible target object situation.

The selection (S03bl) of the at least one event sequence for reaching the target object situation based on the start object situation is preferably one that suits the task Solution. The solutions stored in a database are preferably quantified in the form of an event sequence with regard to the aforementioned variables and a solution suitable for the task is selected in the database if a corresponding solution for the task has been read (S02).

Alternatively, a new event sequence is formed (S03b2), in particular a chain of procedures for forming a more complex procedure for reaching a target object situation, starting from a start object situation on the basis of the partial event sequences found in step S03a and their concatenations. For this purpose, a new event sequence is formed (S03b) if a solution that matches the task was not read (S02), as an n-concatenation from the existing event sequences, preferably comprising the following steps:

Reading (S03b-01) the database for the purpose of searching for at least a first and an nth event sequence, in particular searching for event sequences from a first to an nth event sequence, where n denotes a natural number and the first event sequence is suitable for the possible starting object situation into a first intermediate object situation and for all natural numbers for which 1 < k < n, the k-th event sequence is suitable for transforming a (k-l)-th intermediate object situation into a k -th intermediate object situation, and the nth event sequence is suitable for converting an (n-l)th intermediate object situation into the possible target object situation

According to the invention, a step of iterating (B02a) over possible intermediate object situations and chains of intermediate object situations, in particular a step of recursive iteration, can be provided, with these intermediate object situations preferably being stored in a database, so that a step of reading ( B02b) a database for the purpose of searching for suitable event procedures.

A number n denotes the length of the respective chaining of the procedures for each possible chaining of the intermediate object situations and is given by a natural number greater than or equal to 2. In the step of selecting and chaining procedures to form a more complex procedure, each event procedure is preferably capable of either transforming the possible starting object situation into the first intermediate object situation, or for a k with 1 < k < n, to transform the (kl)-th intermediate object situation into the k-th intermediate object situation, or

(n-l)-th intermediate object situation to convert into the possible target object situation

For this purpose, in a first step, at least one quantitative suitability criterion for the chaining of the procedures ((partial) event sequences) is preferably calculated (B03) for each chaining of procedures caused by the iteration. Such that the step of selecting one or more concatenations occurs based on at least one quantitative suitability criterion/trait.

Analogously to this, it can be provided that the step of selecting an event sequence from a set of suitable event sequences in step S03b1 takes place on the basis of at least one quantitative suitability criterion/feature (as defined herein).

It goes without saying that the event sequences between two object situations, for example within a database, can be initially collected and stored in any programming or description language, in particular in any uniform programming or description language.

The method according to the invention, in particular the control method, can preferably also include a step for storing (SOS) the determined event sequence. If the determined event sequence is a newly created event sequence, it can be stored in a database as an n-concatenation, with this newly concatenated event sequence being suitable for converting the possible start object situation into the possible target object situation . This has the advantage that when carrying out a later event sequence that is identical or at least similar, it can be used in step (c) of the method according to the invention, namely the determination (S03) of an event sequence.

According to one development, the method according to the invention, in particular the control method, includes a first partial step for determining an event sequence from a set of known (partial) event sequences in a first memory in which successes are stored, ie event sequences that are suitable for starting To convert the object situation into the target object situation, a (basically) suitable event sequence being determined by iteratively searching for known (partial) event sequences and selecting a known event sequence or forming a new event sequence, in particular forming a new event sequence, and one second partial step for determining an event sequence from a set of known (partial) event sequences in a second memory in which failures are stored, ie event sequences that are not or only insufficiently suitable for converting the start object situation into the target object situation , by searching/determining the event sequence determined in the first partial step in the second memory or by comparing the event sequences determined in the first partial step with the event sequences determined in the second partial step, with the (partial) event sequences determined in the second partial step being discarded (cross-checking ).

Due to the fact that failures are not discarded, they can be used in addition to the variables defined herein, in particular quantitative suitability criteria for finding a suitable event sequence. The method according to the invention, in particular the control method or the system for controlling an actuator, is therefore at least quasi-adaptable, preferably capable of learning, so that at least in a first upstream step, very particularly preferably, it is completely based on a comparison of the (entire) determined quantity of potentially suitable Event sequences based on at least one size or one quantitative suitability criterion (as defined herein) can be dispensed with. This saves time and reduces the energy required to determine a suitable (partial) event sequence.

A first memory in which successes are stored is, for example, a so-called success database, i.e. a memory or a database in which (partial) event sequences are stored which occur with the same or a similar combination of starting object situation and target Object situation have led to success. A second memory in which failures are stored is, for example, a so-called

Failure database, i.e. a memory or database in which (partial) event sequences are stored which have not led to success with the same or similar combination of start object situation and target object situation.

According to one development, the method according to the invention, in particular the control method, also includes a step for determining prohibited or impermissible event properties (constraints) within the event sequence to be determined, which is (in principle) suitable for converting the start object situation into the target object situation from a set to convert from known (partial) event sequences, in particular by controlling an actuator based on the determined event sequence by the controller. Examples of such prohibited or prohibited Event properties include an undesired operation such as an object panning, rotating, or jerking due to the object's inertia, air movement, or vibration. Such forbidden or impermissible event properties can be determined separately for individual or within individual (partial) event sequences. In this way, (partial) event sequences can already be excluded in advance, which appear potentially suitable for converting a start object situation into a target object situation, but which are unsuitable due to the presence of a determined forbidden or impermissible event property. A comparison of the determined number of suitable event sequences on the basis of at least one quantitative suitability criterion (as defined herein) can thus advantageously be dispensed with.

According to one development, the method according to the invention, in particular the control method, also includes a step for monitoring an ongoing or a recurring process or a (partial) event sequence, in particular a step for monitoring the recurring start-object situation. An example of this is the action of a robot on an assembly line that repeatedly performs the same operation). In this way, despite the (original) determination of a (partial) event sequence that is/was (basically) suitable for converting a repetitive start object situation into the target object situation, and this (partial) event sequence (e.g. connected several times in a row) is carried out repeatedly, spontaneous changes within an ongoing process, in particular to recurring start-object situations, are determined and the control method is thus (independently) adapted. This has the advantage that the occurrence of errors caused, for example, by a (slightly) changed start object situation, can be prevented and that external intervention, for example by a user or another method, can be dispensed with. If, for example, such a spontaneous change is determined within an ongoing process, in particular in a recurring starting object situation, the system for controlling an actuator can decide in a step of evaluating this change whether the method according to the invention, in particular the control method, must be run through again to determine a new (alternative or adapted) event sequence that is suitable for converting the changed start object situation into the desired target object situation.

According to one development, the method according to the invention, in particular the control method, also includes at least one step of repeatedly determining an event sequence or a step of repeatedly running through the method according to the invention, in particular the control method (according to the steps defined herein, in particular excluding the sequence of events that was not suitable for converting the start-object situation into the target-object situation) if the previously determined and/or or the sequence of events that was carried out was not suitable for converting the start object situation into the target object situation. This has the advantage that the system, in particular the system for controlling an actuator, does not abort the process of transferring the start object situation into the target object situation (and e.g. goes into an error mode), so that external intervention, e.g a user or other process can be dispensed with. For example, the step of repeatedly determining an event sequence includes a preceding partial step in which the actuator returns to its starting position before the method according to the invention, in particular the control method, is run through again.

According to a development, the determination of an event sequence according to step (c) of the method according to the invention, in particular the control method, is only partially completed before the transfer of a start object situation into the ultimate target object situation begins. This means that with the beginning of the transfer of a start object situation into the final target object situation, the determination of the entire event sequence is not yet complete, but the individual partial event sequences are determined successively and step by step to the (entire) event sequence from individual partial Event sequences (procedural event knowledge, a memory, a database) are concatenated (linked together). This has the advantage that after selecting at least one, preferably at least two, particularly preferably at least 30%, particularly at least 50% of the (partial) event sequence(s) or forming one, preferably at least two, particularly preferably at least 30%, in particular at least 50% of the new (partial) event sequence(s) already with the transfer of a start object situation into the final target object situation via intermediate target object situations (i.e. the target object situations that do not yet end the entire required event sequence represent) can be started, which results in time and/or resource savings (e.g. required computing power).

In addition, it can be provided that after the completion of each part-event sequence, preferably at least after the completion of every second part-event sequence, very particularly preferably after Completion of at least 30%, in particular at least 50% of the partial event sequences of the entire event sequence, the (accordingly) expected intermediate target object situation is compared with the actual intermediate target object situation. For this purpose, the method according to the invention, in particular the control method, preferably includes a step of determining an actual intermediate target object situation (ie that of an object situation which describes a state between a start object situation and a target object situation and which does not yet mark the end of the entire required event sequence represents), which in turn can serve as a starting object situation for a further or independent (partial) event sequence, by means of a sensor and the comparison of this determined actual intermediate target object situation with the determined intermediate target object situation to be expected of a selected ( Partial) event sequence or newly formed (partial) event sequence. This has the advantage that when concatenating a large number of partial event sequences to achieve the ultimately (desired/desired) target object situation, deviations from the desired event sequence can be determined and compensated for by making suitable adjustments. When a corresponding deviation is determined by the controller, the step of determining an event sequence can be adapted on the basis of the determined deviation (e.g. with successive/partial determination of the entire event sequence, as defined in advance) or the actuation step is aborted of the actuator, which is controlled based on the originally determined sequence of events. Such deviations of the actual intermediate target object situation from the determined expected intermediate target object situation based on the desired event sequence can be caused, for example, by disruptive factors such as friction losses, wear and tear, external environmental factors, but also after the maintenance/repair of a device and a device with it e.g. associated ease of movement of components. Such disruptive factors can then be taken into account as an event criterion in the step of determining further event sequences.

General Benefits

The control method according to the invention enables an optimized transfer of a start object situation into a target object situation by accessing event sequences stored in a memory, in particular a database. In addition, errors that can occur due to an essentially manual specification of an event sequence are minimized.

The control method according to the invention also makes it possible to react to spontaneous changes within a continuous or repetitive process and to provide a new (alternative or modified) event sequence.

Ultimately, the control method according to the invention is designed to be capable of learning.

Further advantages can be found in the description and the exemplary embodiments.

Description of the invention

ITSF (Intuitive Tacit Solution Finding) is a new procedure for the collection, structured storage, easy retrieval and concatenation of building blocks of procedural event knowledge. This type of knowledge is the most sought-after "know-how" for a variety of processes. The advantage is that the event sequence does not have to be made explicit. This means that the event procedure can be left in the existing programming or description language in which it is collected by a technical facility . The method according to the invention comprises the following (ag): a) Collection and storage of event sequences: Method for the collection and structured storage of procedural event sequences as the connection of a start and a target object situation. Each of these stored event sequences is a small building block of the procedural knowledge of events b) Rapid retrieval of event sequences: The initially recorded event sequences are stored in a structured manner in the list of event procedures and can be retrieved very quickly and easily via their start and target object situations c) Normalization of object situations ions: Concrete objects are first assigned to more general object types (normalization of object types). These object types are then combined into an object situation within a limited spatial section and normalized using the list of object situations. The specific objects are preferably assigned to object types on the basis of information collected by means of sensors. In a particularly preferred manner, the specific objects are assigned to object types on the basis of the sensors and the processing of the data in an adaptive controller. d) Procedural event sequences: The event sequence sought as a connection between these two object situations always lies between the currently analyzed starting object situation and the required target object situation and is stored in the list of procedural event sequences. e) Concatenation of event sequences: If there is still no entry in the list of known event procedures for the request for a start and target object situation, a search is made for a concatenation of event sequences that has the start and at the very end the one being sought Contains target object situation. Each link in this concatenation chain is connected via its normalized target object situation equal to the start object situation of the next sequence that fits into this connection. If this concatenation is not found within the safety period that depends on the intended use of the control, the old adaptive control or a manual procedure must be called in order to collect the previously unknown event sequence for the first time and initially. As already described, an event represents a transition from a start object situation to a target object situation in a specific space section. These states can be interpreted as nodes in a graph, the events can thus be interpreted as edges between the states. Accordingly, all states that can be converted into one another can be summarized as a graph. Depending on whether events are reversible, they can be represented by directed or undirected edges. If an event sequence is now sought that is suitable for converting an initial state into a target state, simple routing algorithms such as A* or Dijkstra can be used. The graph can be regularly optimized for faster routing. Various graph databases that store information in graphs are already known from the prior art. These are optimized according to technical criteria. Further optimizations can be made, for example, by forming so-called contraction hierarchies. To do this, the entire graph is analyzed and virtual edges are generated, which store information about the fastest paths between two points. Thus, the search for an event sequence from a start state to a target state can take place extremely quickly. Herein, the terms "procedure" and "event" are used interchangeably. f) Ongoing ITSF operation in a database: Intuitive Tacit Solution Finding is a useful supplement and effective relief of known adaptive controls. The old controller runs in the background and makes fine adjustments to ITSF procedures if necessary. It is also used in completely unknown object situations and to collect completely new event sequences. g) Constant fully automatic further development: Due to the ongoing concatenation of event sequences, even in the idle or sleeping state, increasingly complex event procedures automatically arise in the list of procedural event sequences. This increases the complexity and larger, more strategic tasks can be taken on. However, the first procedures with their finer granularity are still retained. Both serve the extensive relief and acceleration of the old adaptive control. In addition, when ITSF is in the idle or sleeping state, all the concatenations previously possible in the list of procedural event sequences are generated.

ITSF - Intuitive Tacit Solution Finding is a new method for the collection, structured storage and easy retrieval of procedural event knowledge. This knowledge is not so easy to describe explicitly, mostly unconsciously (tacit) and therefore requires special treatment when processing it. With this new way of dealing with it, the old problems of dealing with the most important part of tacit knowledge can be solved thoroughly. But what is the general difference between knowledge and information or data?

First of all, data are only purely syntactic recordings of facts or events. They are able, so to speak, to flip new switches at another point in the processing chain. However, this data with its semantic aspects only becomes information when we relate it to our already existing semantic knowledge structure. Only then do data make sense and become correct information for us. You can verify, falsify or expand the already existing semantic knowledge. However, this information only becomes knowledge when we then correctly classify it in our semantic knowledge structure and often link it semantically.

The two most basic categories of physical reality around us are objects in space and events in time. This probably the most important distinction is always the cause of the two actual basic types of our knowledge according to their direct relation to the physical reality around us and is therefore the basic subdivision into: declarative object knowledge and procedural event knowledge.

Described in detail by Theo Mulder 2006 (Mulder, Theo (2006)): The adaptive brain: about movement, consciousness and behavior. Thieme, publisher C.H. Beck). Declarative object knowledge refers to local objects in space, which can be explicitly and extensively described with all their properties. You are always in local object situations with multiple, sometimes very extensive relationships between these objects in a section of space. Procedural event knowledge, on the other hand, refers to sequential or parallel events in a period of time, which are much more difficult to describe explicitly in their many sequential or parallel steps and are very often stored unconsciously (tacit), i.e. implicitly in their type and sequence in the brain. This unconsciousness also relates in particular to their duration and energy expenditure.

Nevertheless, the clear unconsciousness often goes hand in hand with this implicit knowledge that we take for granted, which we only become fully aware of when a process or an employee in a corresponding position fails and then no longer performs the necessary activity correctly due to the lack of procedural knowledge can be. Basic examples of implicit knowledge are activities that are actually very simple and yet difficult to describe, such as maintaining the correct balance when riding a bicycle, the very special handling of tools in arts and crafts, or even entire sequential work processes. In particular, the explication, structuring and subsequent retrieval of procedural event knowledge is still a really serious problem for many companies and technical institutions.

In order not only to react, but also to be able to act sensibly in a changing reality, every intelligent, adaptive controller today needs a well-structured knowledge management system with which it can learn from the findings and experiences of the past. It's no longer just about collecting data, it's about the analysis, classification and easy findability of knowledge modules. This means that control technology and knowledge management must work together ever more closely. To do this, it is necessary to collect the two most important types of knowledge separately, to store them in a structured manner and thus to be able to find both again easily, quickly and correctly be able. The main goal of Intuitive Tacit Solution Finding - ITSF is the fast and correct retrieval of knowledge modules of the procedural event knowledge.

For example, those skilled in the art know from "The role of tacit knowledge in Group Innovation", California Management Review, Vol. 40, No. 3, pp. 112-132, that "the management of subconscious information is poorly explored, especially when compared to the works in terms of explicit knowledge (see Figure 13).

On the surface, we actually take it for granted that we can articulate whatever knowledge we think we have. Only when we think about it more closely do we then realize that for all the knowledge and its articulation, at least a whole lot of unconsciously used basic knowledge, common thought models, abilities and skills such as those of speaking are necessary, which we unfortunately only rarely become fully aware of and which we can articulate only with difficulty or not at all.

Polanyi put it this way as early as 1966: "(...) that we know more than we know how to say." (Polanyi, Michael (1966): Implicit knowledge. (The tacit dimension.). Deutsch Suhrkamp 1985).

There is a very large area of knowledge that is difficult to articulate, which sometimes refers to very complex processes such as our movements or to highly complex external events. In logical contrast to local objects in space, there are always sequential or parallel events in time. This distinction is the most essential articulation in our physical reality and, for that reason, also the most essential articulation of our ways of knowing about that physical reality. Nevertheless, no one talks seriously about it today, this fact just seems so natural to us. However, only the investigation of the inner relationship between these two basic types of knowledge and their meaningful connection becomes interesting.

Many local objects are always involved in all sequential events and the two can only be separated in our purely theoretical knowledge and experience of these processes. This structure is one of our greatest intellectual achievements, which unfortunately is hardly appreciated today. In any case, the fact remains that sequential events in time are much more difficult to describe than local objects in space. Events are also always associated with a certain use of force or energy over time, which we find very difficult to describe and convey quantitatively in words. It all starts with the recognition and description of local objects in a section of space, their diverse properties, their location and diverse spatial relationships to one another. The specific objects are first assigned to more general object types that depend on the purpose of our spatial observation. Such a purpose can be, for example, autonomous driving in road traffic or, in contrast, the recognition of workpieces and their position in relation to each other when using industrial robots. Defined object situations then arise from the assignment of detected objects to object types and the position of these object types in relation to one another in a spatial section.

In road traffic, for example, object types can occur such as pedestrians, cyclists, cars, trucks, traffic signs, road boundaries, walls or even your own position in a section of space. When industrial robots work, object types such as cylinders, cuboids, cubes, blanks or the finished product are more likely to occur. After recognizing the local objects, assigning them to object types and their position, the entire object situation in a section of space is recognized. Both the detected object type and the detected object situation are compared with all object types and object situations already stored and can verify or expand these two lists. If known object types or object situations occur, these are simply assigned to the already known entries in both lists. New entries are made in the two lists for previously unknown object types or object situations (normalization of object knowledge). This has been collected, declaratively described and stored for a long time with the technologies available today. Most of the time, the collected data is then pumped into large data lakes, from which the unstructured data has to be analyzed very laboriously.

It is much more difficult to collect procedural event knowledge. Here we can only create an adequate description and good retrieval through the relationship to the declarative object knowledge. In reality, events always connect an actual object situation with a target object situation, ie we store procedural event knowledge as a connection between a start object situation and a target object situation as an event sequence. For the connecting event, the necessary description must always be recorded with the necessary change of position and the energy and time required for this. This is mostly unconscious to us quantitatively (tacit). In addition to the detected actual object situation, the controller must always have a required target object situation. In autonomous driving, for example, this consists of parameters such as those required Minimum distances to other object types or a target speed, direction and destination on the road.

The two main categories of physical reality are shown in Figure 14.

When using industrial robots, the target object situation consists, for example, of the correct position of a workpiece that has previously been assigned to an object type and an actual object situation in a specific spatial section. Simplifications can be achieved here, for example, if each object type is given a specific color beforehand. The industrial robot then creates the required target object situation from the recognized actual object situation. This can be done through resorting as well as through proper editing. There is always a specific procedural sequence of events with a specific time duration and a specific energy consumption between the actual object situation and the target object situation. After the object situations have been stored, these event sequences can then be collected for the first time by performing a one-off pre-exercise and storing them. The list of sequential event procedures must therefore always be initially filled at the beginning. The start ID is the detected actual object situation and the target ID is the required target object situation. The procedural event sequence is then recorded correctly and flawlessly structured in the new list of procedural events as content via both and can therefore also be found again. This content can be in any programming or description language. In addition to the event sequences, a lot of other useful information can also be collected and stored.

In addition to the start and target ID of object situations, the content of the event sequence and the energy required to drive a car or access the tools of an industrial robot must also be stored for the list of procedural events. In addition to these basic data of a procedural event, we are of course also interested in the duration and especially the degree of success of a procedure in achieving a target object situation. The degree of success should always be broken down into grades of evaluation and clearly reflect both success and failure. Thus, for future applications of this event procedure to achieve a desired object situation, this sequence can be particularly preferred, or it can also be avoided in principle. The combination of procedural event knowledge and declarative object knowledge is shown in FIG.

Structured in the way the procedural event knowledge is recorded here, it can also be retrieved easily, quickly and correctly. They are not unstructured data lakes and none Big data scientists are necessary for the evaluation, which does not exist anyway with autonomous driving or the use of industrial robots in the production process on site. In the simplest case, there is already an entry in the list of procedural event sequences that contains the correct actual and target ID of the object situations with a high assessment of the probability of success and the necessary time and energy. This is then selected and carried out based on the comparison of these parameters. If this is not the case, a chain of several event procedures must be searched for in the event list. The recognized actual ID(a) of an object situation a becomes the starting point and further procedures with target ID D(a to xn) = start ID(xl to z) are linked for as long as xl to xn until the desired target I D(z) of the required target object situation z is reached. For the selection of the event sequences, in addition to the important rating of success, the shortest duration and the lowest energy consumption are used as further criteria. Of course, this only happens until a specific safety period, which is very important for the respective application, is exceeded at some point. This depends on the purpose of the application.

Of course, the available safety period can be very different for autonomous driving than for industrial robots. If a suitable concatenation is found within this period of time, which contains the actual ID(a) of a recognized start object situation at the beginning and the target ID D(z) of the target or target object situation at the very end, this is concatenated Event procedure selected and executed. Of course, all sub-procedures should have the highest possible probability of success. In addition, of course, the comparison of the stored time duration and its energy consumption is always included in the evaluation and selection of similar suitable procedures. If such a chained procedure is found, it is executed and then a new entry for this procedure is created in the list of event sequences. As a result of the chaining, an ever increasing complexity and further development of new event procedures arises over time, while the fine granularity of the first procedures is retained.

If no such chained event procedure is found, a new procedure must be raised. Of course, this must first be practiced manually or using the old adaptive control, ie both in the case of autonomous driving and the use of industrial robots, the manual or the old mechanical control must be taken over for a short time. This then supplements the list of existing event procedures in a meaningful way. The fine granularity of the old procedures is retained, but the complexity of new concatenated procedures, on the other hand, is constantly increasing fully automatically. I would like to be able to quickly and correctly retrieve procedural event knowledge in the future Applications like to call it "Intuitive Tacit Solution Finding - ITSF". Intuitive because the solution is not found by calculation, but intuitively by searching in the memory of existing sequential event procedures and Tacit because we are not aware of this search. It doesn't work here more just about the declarative object knowledge related to the properties of objects, object types and object situations than the "know that". It's about the hard-to-explain, mostly completely unconscious procedural event knowledge as the often sought-after "know-how" for many types of technical and organizational applications.

With this type of processing of procedural event knowledge, it is possible for us to intuitively and unconsciously (tacit) collect procedural events, find them again and use them in the processing of a control without having to describe these events explicitly in a complicated way. By calling up a recognized actual object situation and a required target object situation, we can get back a defined procedural event of a certain granularity from the device, which in the past has already been carried out as successfully as possible and in the required time as a connection between these two object situations. Within the known world of this control device, it is therefore also possible to make predictions about the success, duration and energy consumption of a procedure to be used in the future.

These can be compared with the available safety time span until an action is necessary and the available energy reserve of a technical facility. This is precisely what is missing in detail from all of the adaptive control devices known today. Of course, it can also be used in a company's completely new knowledge management systems, which, for example, are about the correct storage and retrieval of organizational procedural event knowledge as "know-how". The advantage is clearly that explication of procedural event knowledge is no longer necessary (tacit) and the extremely short reaction time due to the new intuitive retrieval, which makes lengthy calculations of a new procedure unnecessary. Instead of the event sequences, various algorithms can of course also be collected, stored and concatenated for further processing in modern computer facilities. The advantage clearly lies in this that any different programming or description language for the procedures or for the algorithms can be used as the content of the event procedure list It must of course be consistent in each separate device.

Intuitive Tacit Solution Finding (ITSF, see FIG. 16) is thus a new development that replaces previously known controls, for example in autonomous driving in the automotive industry or when using industrial robots usefully supplemented in many branches of industry. It can be placed directly on known control processes and expands them over time, especially in more strategic decision-making processes, without human intervention. Through the correct combination of local object knowledge and procedural event knowledge, we enable us to process the two most important types of knowledge very fluently in order to better control a wide range of technical facilities. In addition, the intuitive and non-explicit handling of procedural event knowledge results in a high speed advantage compared to all the existing processing, such as previous calculations, calculations and the old downright desperate attempts at explication of this type of knowledge.

Procedural event knowledge is the most important type of knowledge used for controls, but it has so far not been able to be treated correctly enough. Through their meaningful and correct connection with the already known declarative object knowledge, we are given the opportunity for the first time to collect both correctly, to process them and to find them again. In particular, the newly developed chaining of many already known event procedures allows us to find completely new and strategic solutions for suddenly occurring requirements or problems to be solved in a flash that have not existed before. This processing can therefore rightly be described as creative in generating new solutions. In addition, intuitive retrieval always requires significantly less time than the complicated, detailed calculation and calculation of a required solution. The treatment of procedural event knowledge, which is hidden from us (tacit), also saves its complicated explication. With this essential addition by ITSF, adaptive controls become significantly smarter, faster and more efficient.

For the processing of procedural event knowledge, at least the following information must be recorded, stored and linked as knowledge building blocks. Only then will a fully automated treatment of the two most important types of knowledge and rapid retrieval be possible.

List of procedural event sequences in a time slice

List of normalized local object situations in a section of space

List of previously assigned object types (normalization of object types)

The search for the object types involved is carried out via the object child content. The object types involved are linked to the object situations using the object child IDs. The search for the correct object situation is carried out first via the object child IDs involved and then via the object situation content. From there, the correct event sequence is called using the Start and Target Object Situation IDs. The old adaptive control used up to now runs in the background and makes fine corrections if necessary. While the old control generated solutions in a cycle of 1 - 12 milliseconds, for the more strategic or completely new decisions of ITSF it is sufficient to make a decision for one or more event sequences once every tenth of a second, which then run with a certain duration. These called known event sequences in turn relieve the old adaptive control significantly. The safety margin available for finding a solution is an example here and always depends on the purpose of using ITSF. In the idle or sleeping state of ITSF, all possible concatenations of procedural event sequences that are still missing are generated and added to the list of procedural event sequences.

Possible applications arise in all fields where adaptive controls are already in use, such as industrial robots, autonomous driving or robotic process automation for computer equipment. The person skilled in the art also recognizes from this that at least one means is provided which is set up to execute at least one event of the determined event sequence.

An industrial robot is described below as a possible application example. Industrial robots are defined "as universally usable (flexible) moving machines with several axes, the movements of which are freely programmable in terms of movement sequence and paths or angles." Fixed programming without any sensors is used in older industrial robots. For adaptive controls, on the other hand, it is important that, in addition to the programmable manipulator, good sensors and adaptive controls are used to determine the current actual object situation(a) and to check the action result in a new target object situation(z).

So far, this ability to learn has only been established at the level of the collected position data. Increasingly complex tasks in the production process today require completely new procedural knowledge modules to be collected, processed and then put to practical use at a higher level of abstraction and processing instead of simple position and movement data. The most important difference lies in the semantics of the procedures, which no longer need to be explained, and in the complexity of processing ITSF.

When programming industrial robots, we differentiate between online and offline programming. While offline programming takes place independently of the real robot in a suitable development environment, online programming is directly connected to the real robot environment. ITSF, in particular an ITSF event sequence list, can be whole particularly useful for online programming of an adaptive robot setup. There are 3 variants for online programming: teach-in, master-slave and play-back. What all 3 variants have in common is that the programming is carried out by pre-exercising the desired movements and manipulations. In this way, the event procedure list of ITSF can also be initially filled in parallel to the adaptive control with a suitable sensor system. This then expands completely independently over the course of the use of this adaptive industrial robot. In addition, the complexity of new event sequences in this list increases fully automatically over time due to the concatenation of existing event procedures. In a preferred embodiment of the invention. If this is not stored locally but in the cloud, all the machines involved have access to extremely extensive information about the entire system integrated in ITSF. Depending on the degree of standardization, all controllers of a type, manufacturer or branch can then access this database of procedural event knowledge. The result is a search engine for procedural event knowledge.

Intuitive Tacit Solution Finding can be integrated as an additional function in any adaptive control of, for example, industrial robots and subsequently implemented. Through the constantly running concatenation of already existing sequences of events, it takes over increasingly complex, later more strategic or completely new creative functions in finding solutions for current and sometimes completely new tasks. While the old adaptive control runs in the range of 1 - 12 milliseconds, it is sufficient for ITSF to make a decision about once every tenth of a second. The selected process then runs for the entire known duration of the sequence of events in question. ITSF thus relieves the old adaptive control device in particular more and more over time. Of course, the adaptive control continues to run in the background and is used for completely new and unknown processes for the one-time pre-exercise of completely new procedures or, if necessary, also for important fine corrections during the entire work and runtime of ITSF event procedures.

Autonomous driving is to be described as a further application example. According to the specifications of SAE (Society of Automotive Engineers) and those of the market leader Bosch, autonomous driving in road traffic is divided into 5 different levels of automation of this autonomous driving process: SAE-L1 driver assistance systems, SAE-L2 partially automated driving, SAE-L3 partially automated driving, SAE-L4 Highly Automated Driving, SAE-L5 Fully Automated Driving. In addition to the use of extensive sensors in the driving process (e.g. multifunction cameras and radar), all these 5 levels require very fast adaptive control, which determines the necessary correct event sequence as a connection between the current object situation (a) and the required target object situation (z). The difference lies in the amount of event procedures that lie between the two new object situations. The three action components Sense, Think, Act are therefore required for the smooth automatic process of this adaptive control. "Today's assistance and partially automated systems support the driver, but they do not replace him. These include, for example, the Stop&Go Pilot or the Active Lane Change Assist. Autonomous systems, on the other hand, will go one step further in future cars: the driver will become a pure passenger in the future. The difference between automated and autonomous driving is also of legal importance." By using ITSF procedures in the cloud and using them for all road users, a huge new field can be opened up here.

In autonomous driving, ITSF is programmed online, or the ITSF event sequence list is collected, by the manual or adaptive control of the vehicle, i.e. event sequences are collected, processed and stored that have already been successfully used in the past Autonomous driving process were used. These can be concatenated to increasingly complex event sequences on all 5 levels of autonomous driving and thus develop fully automatically. This will then result in a higher level of autonomous driving. This information is fully available to all road users via the cloud and there is a huge collection of important knowledge modules about all road users.

In contrast to the calculation and calculation of the necessary event procedures by the adaptive control, ITSF determines the next event sequence by intuitively searching in the procedural event list resulting from the previous work of the old control. In addition to the constantly increasing relief of the adaptive control, more and more complex event sequences are created here and, through their concatenation over time, enable more strategic or even completely new and creative solutions for suddenly occurring problems. Any redundancies must be eliminated. The complexity is constantly increasing and must be linked to geodetic data for autonomous driving right from the start. At the end it is sufficient to enter the starting point and destination of a journey that is already known, and a largely autonomous journey to the target object situation is carried out. The cloud opens up huge opportunities for all road users. Robotic process automation is to be described here as a further application example. From taking over monotonous data entry to automating customer service request responses, Robotic Process Automation (RPA) will enable e.g. finance professionals to save significant time on repetitive, labor-intensive tasks and enable value creation within banks on an industrial scale. RPA revolutionized the banking sector by enabling banks to perform back-end tasks more accurately, faster, and efficiently without completely overhauling existing operating systems and processes. Here, too, all RPA participants can be connected through the cloud.

In the simplest case, the object situation(a) is the desk of a desktop PC that is available after the start. The event sequences are generated by the human-machine interfaces with mouse and keyboard. RPA records this and checks the resulting object situation by comparing it with the known target object situation (z) at the end of the RPA procedure. With ITSF, these can then be concatenated in an ITSF event sequence list and thus fed into larger automation procedures. ITSF can be easily and unobtrusively installed on the same PC environment without having to replace old applications. ITSF learns from the stored processes of RPA and constantly forms new larger process chains. In the end, it is enough to call up a target or target object situation to let a process chain run fully automatically. The sensor technology here lies in checking the degree of achievement of the target or target object situation. Without this check, RPA is blind and must be controlled manually. These process chains can be stored in the cloud for all participants in a company.

The ITSF can be used for robotic process automation. Unlike other traditional IT solutions, RPA allows organizations to automate at a fraction of the cost and time previously involved. RPA is also non-intrusive and leverages existing infrastructure without disrupting underlying systems, which would be difficult and costly to replace. With RPA, cost efficiency and compliance are no longer operational costs, but a by-product of automation."

If RPA is supplemented by ITSF, new and increasingly complex solutions can be created fully automatically from a large number of RPA procedures. ITSF does not require any other environment, but can also run on the PC environment without any problems. In addition to PC use, use in the cloud, on smartphones and tablets is also planned for the future. RPA automates processes on the PC. ITSF enables increasingly complex processes on all end devices for all users of a business. The current start or actual object situation(a) is analyzed automatically. Only the desired target or target object situation (z) has to be requested within the known world of control. As a result, the associated and sometimes very complex ITSF event procedure is called and executed. At the end, the achieved target object situation is compared with the desired result and the degree of success, duration and energy expenditure for the next call of this event procedure are stored, ideally in a company's own cloud. Through the concatenation of procedures, the degree of automation increases continuously and fully automatically.

According to one development, a step of the method according to the invention, in particular of the control method, includes the selection of an event sequence from a set of suitable event sequences based on at least one quantitative suitability criterion. In an additional partial step, a (partial) event sequence can advantageously be linked to a quantitative suitability criterion, while a further (partial) event sequence can be linked to the same or a further quantitative suitability criterion. Particular preference is given to (partial) event sequences which were already suitable in the past for converting an identical or at least similar starting object situation into a target object situation, in a memory, in particular a first memory, for example in a (success )Database deposited. Thus, in the concatenation of (partial) event sequences to determine a suitable event sequence, (partial) event sequences, in particular partial event sequences, which were already suitable for converting an identical or similar starting object situation into an identical or similar target transfer object situation. The quantitative suitability criteria can be weighted differently. Depending on the desired suitability criterion to be optimized, optimized (partial) event sequences, in particular optimized in terms of time, energy, effort and/or success, can advantageously be selected as a result.

In one development, an effort and/or cost indicator is used as a quantitative suitability criterion. In this way, a process or a (partial) event sequence can advantageously be optimized with regard to the effort to be used.

In a further embodiment, the energy expenditure is used as a quantitative suitability criterion. Thus, the control method or the control can advantageously be optimized to the effect that so that it can be carried out as energy-efficiently as possible. This is particularly relevant for processes that are characterized by the occurrence of a so-called peak power (a maximum power required within a process), since a (partial) event sequence can be determined in which the peak power is reduced and/or a even distribution of power can be adjusted.

In further training, a success indicator is used as a quantitative suitability criterion.

In a preferred embodiment, the determined sensor data are normalized before they are used to determine the starting object situation. A key goal of normalization is to eliminate redundant information. Advantageously, there are no longer any data duplications after the normalization and each piece of information is only stored in one place in the database. Eliminating the duplication increases database consistency.

In a further embodiment, the method includes a step for storing the determined event sequence. This allows determined event sequences to be pooled in a database. The more determined event sequences are in the pool, the easier it is to convert a (new) start object situation into a (new) target object situation, since a suitable event sequence can already be found that is either identical to or similar to an optimal new one (Partial) event sequence is. As a result, the computing effort can be minimized, as a result of which the method can run more efficiently.

In a preferred embodiment, the method according to the invention, in particular the control method, comprises a step for terminating the determination of an event sequence if no suitable event sequence for reaching the goal/for reaching the target-object situation can be determined and/or if the control includes the step for determining a Event sequence cannot perform. The controller cannot carry out the step of determining an event sequence, for example, if there is no identical or similar (partial) event sequence or at least two (partial) event sequences are present with regard to the consideration and weighting of at least one quantitative suitability criterion that is suitable to convert an object situation into an (intermediate) target object situation. This has the advantage, for example, that the method does not remain “trapped” in the iterative search step and/or a signaling step indicates to a user or another method or system for control that external intervention is required. In one development, the step of breaking off is automatically broken off after a predefined period of time, which can be freely selected by a person skilled in the art, has elapsed. This advantageously prevents the downtime or idle time from being too long. However, in order to convert a start object situation into a target object situation, an event sequence that is at least similar to a sought-after optimal event sequence can be selected with the termination. It can thus be advantageously achieved that there is not too long a standstill or idle time.

A development includes a step of adding a (partial) event sequence newly determined by the controller to the set of known (partial) event sequences, in particular in a memory. As a result, the above memory (pool) of available (partial) event sequences, in particular in a success database, is expanded, which advantageously gives a higher probability that a suitable (partial) event sequence will be found that has a start object situation in a target -Object situation transferred.

In a development, the method also includes a step of automatically supplementing the known event sequences with possible further event sequences, while the method is not used to achieve a target object situation.

The invention also includes a system for controlling an actuator for converting a start object situation into a target object situation by means of a controller, in particular for converting the actuator from a start object situation into a target object situation by means of a controller, preferably an adaptive controller : a) input means for receiving (SOI) a starting object situation from a sensor, b) input means for receiving (S02) a target object situation, c) computing means for determining (S03) an event sequence that is suitable for the starting object situation to transfer into the target object situation from a set of known (partial) event sequences

Iterative searching (S03a), preferably in a database, of known (partial) event sequences comprising the start and/or target object situation and/or object situations from partial event sequences of previous iteration steps, Selecting (S03bl) at least one event sequence for converting the start object situation into the target object situation or creating (S03b2) an optimized event sequence for converting the start object situation into the target object situation on the basis of the (partial) found in the partial step of the iterative search Event sequences and their concatenations, d) output means for outputting a control signal for driving (S04) an actuator based on the event sequence determined by the controller.

In one development, the system includes a means for detecting known event sequences and/or known object situations and/or assignments between them.

In an alternative embodiment, the system comprises a means for storing at least one success indicator and/or at least one duration indicator and/or an effort indicator and/or at least one cost indicator and/or at least one relevant point in time for event sequences.

According to one development, the system for controlling an actuator also includes a means for monitoring an ongoing or a recurring process or a (partial) event sequence. As a result, despite the (originally) successful determination of a (partial) event sequence that is/was (basically) suitable for converting a repetitive start object situation into the target object situation, and this (partial) event sequence (e.g . Several times connected in series) is carried out repeatedly, spontaneous changes can be determined within an ongoing process and the control method can thus be adapted. This has the advantage that the occurrence of errors that can occur due to a (slightly) changed start object situation can be prevented and external intervention, for example by a user or another method, can be dispensed with. If, for example, such a spontaneous change is determined within an ongoing process, a system for controlling an actuator can decide in a step of evaluating this change whether the method according to the invention, in particular the control method, is run through again in order to determine a new or adapted event sequence , which is suitable for converting the changed start object situation into the desired target object situation.

In a preferred embodiment, the invention comprises a computer program product which comprises instructions which, when the program is executed by a computer, cause it to Determining an event sequence to convert a start object situation into a target object situation. The computer program product can perform an iterative search and/or select an event sequence and/or combine a (partial) event sequence to form concatenations. A computer program product advantageously enables these steps to be implemented quickly, preferably in the range from milliseconds to seconds. Furthermore, the computer program product can advantageously be quickly adapted to different circumstances (different start/destination object situations). The computer program product is preferably stored on a computer-readable medium, which is also preferably either comprised by a computer and/or can be exchanged between computers.

The invention further relates to an adaptive controller comprising a computer as described above.

In a further development, a computer program product includes instructions which, when the program is executed by a computer, cause the computer to display the results of a computer program in human-readable form, or to convert the results into another (data) format which can be displayed in human-readable form by another computer program product and/or to cause the computer program product according to one of the four preceding claims to implement the method.

Furthermore, the invention relates to at least one industrial robot system, which includes a control system as defined above, a sensor for determining a starting object situation and an actuator for converting a starting object situation into a target object situation.

In addition, the invention relates to a vehicle guidance system which is preferably adapted to a motor vehicle, in particular a driver assistance system or system for partially automated or autonomous driving, comprising a control system as described above, a sensor for determining a starting object situation and an actuator for converting a starting object situation into a target-object situation.

The invention also includes a traffic control system which is suitable for implementing the control method according to the invention, the traffic control system comprising: a large number of motor vehicles, and for each motor vehicle: o a first communication interface, in particular wireless interface, which is set up to communicate with other motor vehicles in a first immediate vicinity of the motor vehicle, o a second communication interface, in particular wireless interface, in particular by means of a mobile radio connection, in particular 5G, for all vehicles to communicate with a server.

The invention also includes a device for robot-controlled process optimization, comprising: a human-machine interface, preferably a desktop environment, preferably a desktop environment of a workstation PC comprising a mouse and/or keyboard, a sensor or a sensor system set up for this purpose is to record an object situation of the man-machine interface, a comparison unit which is set up to compare at least two object situations, a memory and a CPU which are set up to execute the control method according to the invention, the memory in particular also being in a cloud can be provided and comprises a database set up for the methods mentioned, wherein a robot can communicate with the cloud via a data interface, in particular wireless data interface, in particular by means of a mobile radio connection, in particular 5G. Communication via a mobile radio connection or a radio standard such as WLAN, Bluetooth advantageously allows data, in particular event sequences, to be transmitted without a cable connection being necessary.

In addition, the invention also relates to a method for robot-controlled process optimization, which includes the control method described above and a system described above, wherein in particular the start and target object situations can designate virtual situations, a system being provided, at least comprising a man-machine interface, in particular a desktop environment, in particular a desktop environment of a workstation PC comprising a mouse and/or keyboard, a sensor system that is set up to record an object situation from the human-machine interface, a memory and a CPU for processing, and the method also includes at least one step of comparing, in which two necessary object situations are compared with one another. wherein the method further comprises at least one step of a comparison, in which two necessary object situations are compared with one another.

The invention also relates to a method for normalizing object types to support the control method defined above and a system defined above, in particular by collecting and/or using declarative object knowledge, further comprising the following steps:

initiating a spatial consideration,

Determination of a purpose of the spatial consideration in the form of at least one statement of the purpose of the spatial consideration,

Detection of at least one specific object in a section of space,

Assigning a specific object to an object type, in particular a more general object type, depending on at least one purpose of the spatial consideration, and

Saving the assignment of the specific object and the object type using a database.

Furthermore, the invention relates to a method for normalizing object situations

Supporting the control method defined above, a system defined above and/or below Use of a method for normalizing object types, in particular by collecting and/or using declarative object knowledge, also comprising the following steps:

Initiation (D01) of a spatial consideration,

Determination (D02) of a purpose of the spatial consideration in the form of at least one statement of the purpose of the spatial consideration,

Detection (D03) of at least one specific object in a section of space,

Assigning (D04) the specific object to an object type to which the specific object belongs, in particular assignment by reading the object type from a database,

detecting (D05) first information about a location/position, in particular relative location/position, of the at least one specific object in space, and

Determination (D06) of a normalized object situation for the spatial section using the object types and the first information.

The task is also solved specifically by the method for the adaptive control of a process or a control system, also referred to herein as a control method. Accordingly, there is a method for the adaptive control of a process, a system or a control system, in particular for controlling an actuator for transferring the actuator from a start object situation to a target object situation by means of a control, preferably an adaptive control or for automatically finding the solution a problem or a task, in particular by collecting and/or using procedural event knowledge, including:

Definition (SOI) of a task, which consists of converting a concrete start object situation into a concrete required target object situation with the help of an event sequence (since the start object situation is given, this corresponds to the definition of the desired target object situation),

Reading (S02) of a database, possibly on a server (of the company and/or a service provider) to search for a suitable solution in the form of an event sequence that is suitable for solving the task, with the database being suitable for assigning at least the following variables to each other : ■ an identifier of a possible start object situation,

■ identifiers of the object types involved,

■ Identifier of a possible target object situation,

■ identifiers of the object types involved,

■ information about an event sequence, with the event sequence being suitable for converting the possible start object situation into the possible target object situation,

Selecting (S03a) an event sequence as one that fits the task

Solution in the database if a solution that matches the task is read

(S02) became, or

Creation (S03b) of a new sequence of events, if a solution suitable for the task was not read, as an n-concatenation from the existing ones

Event sequences, comprising the following steps:

Reading (S03b-01) the database to search for at least a first and an nth event sequence, in particular a search for event sequences from a first to an nth event sequence, where n denotes a natural number and

* the first event sequence is suitable for converting the possible starting object situation into a first intermediate object situation, and

* for all natural numbers for which 1 < k < n, the k-th event sequence is suitable for converting a (k-l)-th intermediate object situation into a k-th intermediate object situation, and

* the nth event sequence is suitable for converting an (n-l)th intermediate object situation into the possible target object situation,

If necessary, storing a newly created event sequence as an n-concatenation in the database, with the new concatenated event sequence being suitable for converting the possible start object situation into the possible target object situation, with the event sequences between two object situations being in any programmer - or description language are initially collected and stored in particular in any programming or description language that is always uniform.

The advantages of the solution according to the invention result from the above statements. In particular, a method is provided which can dynamically provide complex procedures for solving highly complex problems from simpler building blocks. Due to the use and maintenance of the database according to the invention, possibly in the cloud, the procedural event knowledge is becoming more and more extensive. The controller learns and becomes more and more efficient with increasing complexity.

The abstraction provided allows the invention to be used in all industrial application areas, for example in industrial robots, the autonomous or semi-autonomous control of vehicles, or robot-controlled process automation. What all areas of application of the method according to the invention have in common, however, is that at least one means is included which can be brought to implement at least one event from an event chain.

A high level of abstraction is also provided on the part of the event sequences. These can be in any programming or description language. This again increases and supports the provided compatibility and the integration of the invention into already existing systems. For example, partial sequences can be available in completely different descriptions, including any programming and description languages, but must always be uniform for the respective application in a technical facility.

It should be understood that a start object situation herein describes the state (e.g. location/position) of an object or subject at a start time in a space or space section. For example, the gripper arm of a robot can be in a certain position at the start time. In contrast, the target-object situation describes a state of an object or subject at an end point in time. For example, a robot's gripper arm moves from the start object situation until the target object situation is reached.

The expert understands a sensor to be a technical component that has certain physical or chemical properties (physically, e.g. heat quantity, temperature, humidity, pressure, sound field sizes, brightness, acceleration or chemically, e.g. pH value, ionic strength, electrochemical potential) and/or the material quality of its surroundings qualitatively or can be recorded quantitatively as a measured variable. These variables are recorded using physical, chemical or biological effects and converted into an electrical signal (data) that can be further processed.

An iteration (lat. repetition) is generally understood as the multiple execution of one or more instructions. The iteration is realized by loops. The loop is terminated by a termination condition. In the mathematical sense, an iteration describes a process of gradually approaching the solution of an equation using a repetitive calculation process. An iterative search works like depth-first search, but avoids the disadvantages of completeness by limiting the search depth. In iterative search, a bounded depth-first search is performed iteratively, increasing the level to which the bounded depth-first search explores the graph by one with each iteration.

An input means is a means of transmitting data, which data may come directly from a sensor, a communications interface, a computer, or from memory. An input means can include a current-carrying element (e.g. an electrical line), a wireless interface establishing a radio connection (e.g. LTE, WLAN, Bluetooth, LoRaWAN), an optoelectronic transmission (e.g. laser), or another means which is suitable for data transmit and/or receive.

A computing device according to the invention preferably comprises an electronic computing device (e.g. a computer) which is set up to carry out a calculation using a command sequence. According to the invention, a computing device serves in particular to either determine an event sequence or to generate a (partial) event sequence and concatenations from (partial) event sequences.

A cloud or cloud computing is understood here as the Internet-based provision of storage space, computing power or application software as a service. These infrastructures are used primarily via programs (computer program products) on the accessing devices (clients) and via the web browser.

A server includes a program that waits for a client to make contact in order to fulfill a specific information technology service (service) for the client. The server's service is specific to the server, so there is a separate server for each service. The data exchange between client and server is defined by a service-specific protocol. The object is also achieved by the method, in particular the control method for selecting and linking procedures to form a more complex procedure (event sequence). Accordingly, a method, in particular a control method, for selecting and concatenating procedures or (partial) event sequences to form a more complex procedure is provided, comprising at least the following steps:

Defining (BOI) a task, which consists of converting a concrete start object situation into a concrete target object situation,

Iterating (B02a) over possible intermediary object situations and concatenations of intermediary object situations, in particular recursive iteration, and reading (B02b) a database to search for suitable event procedures, with a number n for each possible concatenation of the intermediary object situations being the length of the respective concatenation of the procedures and is given by a natural number greater than or equal to 2 and the first event sequence is suitable for converting the possible starting object situation into a first intermediate object situation and for all natural numbers for which 1 < k < n applies, the k-th event sequence is suitable for converting a (k-l)-th intermediate object situation into a k-th intermediate object situation, and the n-th event sequence is suitable for converting a (n-l)-th intermediate object situation into the possible To transfer target object situation, where any event procedure is suitable, either

0 to convert the possible starting object situation into the first intermediate object situation, or

0 for a k with 1 < k < n, to transform the (k-l)-th intermediate object situation into the k-th intermediate object situation, or

0 to convert the (n-l)-th intermediate object situation into the possible target object situation and the concatenation of the event procedures is suitable as a result, the

To solve the task, whereby the database is suitable for assigning at least the following variables to each other: 0 a first identifier of a possible first object situation,

0 a second identifier of a possible second object situation,

0 information about an event sequence, the event sequence being particularly suitable for converting the possible first object situation into the possible second object situation,

calculating (B03) at least one quantitative suitability criterion of the chaining of the procedures for each chaining caused by the iteration, selecting (B04) one or more chainings on the basis of at least one quantitative suitability criterion.

The advantages of the solution according to the invention result from the above statements. In particular, a method is provided which can dynamically provide complex procedures for solving complex problems from simpler building blocks, possibly in the cloud for a number of participants. As a result of the use, updating and maintenance of the database according to the invention, the procedural event knowledge is becoming more and more extensive. The controller learns and becomes increasingly efficient.

The abstraction of procedural event knowledge that is provided allows the invention to be used in all industrial application areas, for example in industrial robots, the autonomous or semi-autonomous control of vehicles, or robot-controlled process automation.

A high level of abstraction is also provided on the part of the event sequences. These can be in any programming or description language. This again increases and supports the provided compatibility and the possible integration of the invention into already existing systems. For example, partial sequences can have completely different descriptions, including completely different programming languages, but they must be uniform.

The task is also solved by the industrial robot system. Accordingly, an industrial robot system is provided which is set up to execute the method according to the invention, in particular the control method and/or the method for selecting and linking procedures to form a more complex procedure, the industrial robot system also comprising: at least one industrial robot, in particular an industrial robot suitable for online programming, in particular an industrial robot suitable for teach-in and/or master-slave and/or playback, a robot controller for controlling the industrial robot, in particular an adaptive controller for controlling the industrial robot, a memory and a CPU, which are set up to provide the method according to the invention, in particular the control method and/or the method for selecting and concatenating procedures for forming a more complex procedure for the industrial robot, wherein the memory can also be provided in particular in a cloud and a comprises a database set up for the methods mentioned, the database comprising event sequences of such high complexity, in particular after a run-in period, that only less than two complex event sequences per tenth of a second are required for full-load operation of the industrial robot en, more particularly less than the decision for one event sequence per tenth of a second, more particularly less than three event sequences per second. The cloud could thus collect information from several institutions. The person skilled in the art knows that the event sequences per unit of time depend on the area of application and are variable.

The adaptive control is relieved so much, which protects it and saves time and energy. The overall system becomes more efficient as increasingly complex event sequences are used as procedures for all participants. Part of the solution is also to generate and store all previously possible concatenations of procedural event sequences while the device is in the idle or sleeping state. This constantly adds meaningful value to the list of complex procedural event sequences.

The task is also solved by the system for a vehicle. Accordingly, a system for a vehicle, in particular a motor vehicle, in particular a driver assistance system or a system for partially automated or autonomous driving, is provided, which is set up to implement the method according to the invention, in particular the control method and/or the method for selecting and linking procedures to form a more complex to perform and/or benefit from such a procedure, for example in a cloud, the system further comprising: a classic control of the vehicle, in particular a manual and/or adaptive control of the vehicle or a combination of such, a memory and a CPU which are set up to implement the method according to the invention, in particular the control method and/or the method for selecting and linking of procedures for forming a more complex procedure for the motor vehicle, the memory being able to be provided in particular in a cloud and comprising a database set up for the methods mentioned, the motor vehicle having a data interface, in particular a wireless data interface, in particular by means of a mobile radio connection, in particular 5G or other new standards that can communicate with the cloud, in particular after an initial run-in period, the database includes event sequences of such high complexity that only less than two event sequences per tenth of a second for the traffic bed drives of the motor vehicle are required, more particularly less than one event sequence per tenth of a second, more particularly less than ten event sequences per second.

The vehicle's control is relieved, which protects it and saves energy and time. The overall system becomes more efficient as increasingly complex event sequences are used as procedures. Outsourcing to the cloud ("connected cars") provides fast communication and high computing power or distributed computing. In addition, information about dangers in traffic can be taken into account and vehicles can be coordinated.

The task is also solved by the traffic control system. Accordingly, a traffic control system is provided, comprising a large number of motor vehicles, and for each motor vehicle:

0 a first communication interface, in particular a wireless interface, which is set up to communicate with other motor vehicles in a first immediate vicinity of the motor vehicle,

0 a second communication interface, in particular wireless interface, in particular by means of a mobile radio connection, in particular 5G, for communication with a cloud, wherein the traffic control system is set up to provide the method according to the invention, in particular the control method and/or the method for selecting and linking procedures to form a more complex procedure for at least one of the motor vehicles, in particular for two or more motor vehicles in a coordinated manner between the motor vehicles involved Way.

The load on the vehicle controls is relieved, which protects them and saves time and energy. The overall system becomes more efficient as increasingly complex event sequences are used as procedures. The outsourcing to the cloud ("connected cars") provides fast communication and high computing power or distributed computing. In addition, indications of dangers in traffic can be taken into account and it can be coordinated between vehicles. A direct, even faster coordination can be between the vehicles take place directly, which again promotes and ensures the smooth and safe overall dynamics of the control and flow of traffic.

The task is also solved by the device for robot-controlled process optimization. Accordingly, a device for robot-controlled process optimization, comprising a human-machine interface, in particular a desktop environment, in particular a desktop environment of a workstation PC comprising a mouse and/or keyboard, is a sensor system which is set up to detect an object situation of the human being -machine interface, a comparison unit which is set up to compare at least two object situations, a memory and a CPU which are set up to implement the method according to the invention, in particular the control method and/or the method for selecting and chaining procedures for Forming a more complex procedure to be carried out, in which case the memory can in particular also be provided in a cloud and comprises a database set up for the methods mentioned, in which case the motor vehicle has a data interface, in particular a wireless data interface, in particular by means of a mobile radio connection, especially 5G, can communicate with the cloud.

This makes RPA even more automated and efficient. The increasingly complex procedures save energy and time and further increase efficiency and reliability. In particular, ITSF can run on a commercially available PC or processor solution.

In another example, customer service requests are automatically answered. With ITSF, this becomes even faster, more precise and more accurate. Due to the increasing complexity, the "virtual customer service representative" improves continuously and in a structured way during his "lifetime".

The task is also solved by the method for robot-controlled process optimization.

Accordingly, a method for robot-controlled process optimization, comprising a method according to the invention, in particular a control method according to the invention and/or the method for selecting and linking procedures to form a more complex procedure, in which case the starting and target object situations can designate virtual situations, in which a system is provided, at least comprising a human-machine interface, in particular a desktop environment, in particular a desktop environment of a workstation PC comprising a mouse and/or keyboard, a sensor system which is set up to detect an object situation from the human machine interface, a memory and a CPU for the processing, and the method further comprises at least one step of a comparison, in which the existing actual and the necessary target object situation are compared with each other.

This makes RPA even more automated and efficient. The increasingly complex procedures save time and energy and further increase efficiency and reliability.

The task is also solved by the method for normalizing object types. Accordingly, a method for normalizing object types to support adaptive control of a process, a system or a control system (each as defined herein), in particular by collecting and/or using declarative object knowledge, is provided, further comprising the following steps:

Initiation (C01) of a spatial consideration,

Determining (C02) a purpose of the spatial consideration in the form of at least one statement of the purpose of the spatial consideration,

Detection (C03) of at least one specific object in a section of space, Assigning (C04) a specific object to an object type, in particular a more general object type, depending on at least one purpose of the spatial consideration,

Saving (C05) the assignment of the specific object and the object type using a database.

By normalizing the object types, additional procedural event knowledge is made available for a specific problem solution. As a result, a problem solution can be created for otherwise unsolvable problems or problems that can only be solved inefficiently and laboriously, in particular due to the abstraction additionally obtained through the normalization.

The task is also solved by the method for normalizing object situations. Accordingly, a method for normalizing object situations to support adaptive control of a process, a system or a control system (each as defined herein) and/or using a method for normalizing object types as defined herein, in particular with collection and/or use is declarative Object knowledge, provided, further comprising the following steps:

Initiation (D01) of a spatial consideration,

Detection (D03) of at least one specific object in a section of space, assignment (D04) of the specific object to an object type to which the specific object belongs, in particular assignment by reading out the object type from a database, detection (DOS) of initial information about a location/ Position, in particular relative location/position, of at least one specific object in space,

By normalizing the object situations, additional procedural event knowledge is made available for a specific problem solution, based on more complex situations, usually comprising a number of objects and/or object types. As a result, a problem solution can be created for otherwise unsolvable problems or problems that can only be solved inefficiently and laboriously, in particular due to the abstraction additionally obtained through the normalization. According to a further development, the method according to the invention, in particular the control method, further comprises a step of carrying out a complete or partial handover (S03c-l) of control over the process and/or the system to a control system, in particular a conventional adaptive control system, when a suitable new event -Sequence could not be formed (S03b).

So ITSF just complements the adaptive control and there is no undue hesitation in cases where the classic/adaptive control is able to solve the problem but unfortunately ITSF is not (yet).

According to one development, the method according to the invention, in particular the control method, also includes a step of supplementing (S03c-2) the database with one or more entries which are suitable for assigning at least the following variables to one another: an identifier for the possible starting object situation,

Identifier of the object types involved, an identifier of the possible target object situation,

Identifier of the object types involved, information on an event sequence, the event sequence being particularly suitable for converting the possible start object situation into the possible target object situation, and the information comprising an identifier which allows the event sequence to be identified and/or or one or more information based on an observation of the events of a control by the control system, in particular the conventional adaptive control system.

This is how ITSF learns. ITSF not only learns from new, more complex event sequences that it has formed itself, but also learns directly from the adaptive and/or classic control. The procedures learned in this way are stored in the database, and ITSF can later independently form more complex event sequences by considering and using the newly recorded sequence as a building block for this.

ITSF thus makes it possible to obtain new basic building blocks of event sequences in parallel and to further complex them through continuous concatenation. According to a preferred embodiment, an event sequence is selected from a set of suitable event sequences in step S03b1 on the basis of at least one quantitative feature or suitability criterion, in particular as a suitability criterion that is dynamically calculated and/or stored in the database. Examples of such quantitative features are, as set out hereinafter, a success indicator, a duration indicator, an effort indicator and/or a cost indicator, a point in time indicator and/or a timestamp (also referred to herein as a timestamp).

According to one development, the database is therefore also suitable for also including the following variables and/or assigning them to the other variables in the database in particular: a success indicator (also referred to herein as success rating) which quantifies a degree of successful execution of an event sequence, in particular Degree of success in previously attempted and/or completed execution of the sequence of events.

Success is a key deciding factor for or against an event sequence. This is especially the case when there are multiple possible ways to solve a problem. In this way, the probability of success and, as a result, the efficiency with ITSF can be maximized.

According to a further development, the database is also suitable for also including the following variables and/or for assigning them in particular to the other variables in the database: a duration indicator (also referred to herein as duration), which indicates a duration and/or a variable corresponding to a duration, a Execution of an event sequence, in particular a successful execution of an event sequence, quantifies, in particular a degree of success in the previously attempted and/or executed execution of the event sequence. An example of a time indicator is classification based on the shortest route and the least number of partial event sequences (process steps).

The duration/execution speed is another important deciding factor for or against an event sequence. This is especially the case when there are multiple possible ways to solve a problem. In this way, the overall time and, as a result, the efficiency with ITSF can be maximized. According to a further development, the database is also suitable for also including the following variables and/or for allocating them to the other variables in the database in particular: an effort indicator and/or cost indicator which shows an effort, in particular energy expenditure or total effort, for carrying out an event sequence, in particular a successful implementation, quantifies, in particular a calculated variable from one or more of energy expenditure, computing effort, organizational effort, time expenditure, costs, coordination effort, risk costs.

A specified effort, in particular but not necessarily an energy effort or cost effort, is another important decision factor for or against an event sequence. This is especially the case when there are multiple possible ways to solve a problem. For example, the total costs or the total energy required can be minimized and, as a result, the efficiency can be maximized by ITSF. Therefore, an effort and/or cost indicator is preferably used as a quantitative feature.

According to one development, the database is also suitable for also including the following variables and/or for assigning them in particular to the other variables in the database: a point in time indicator and a time stamp/timestamp, which indicates a point in time when a sequence was recorded, in particular a point in time when a first recording was made Sequence, and/or a point in time of a first successful implementation/reproduction of the sequence of events.

This makes it possible to assess in particular how solid and/or long-term established a procedure already is. The reference to the first successful execution/reproduction also ensures that unsuccessful attempts do not falsify the result.

According to a development, the database is also suitable for using the energy expenditure (herein also referred to as effort) as a quantitative feature.

According to a further development, the database is also suitable for also including the following variables and/or for allocating them in particular to the other variables in the database:

A correlator of at least two, but especially three or more, event sequences, in particular a correlator which quantifies a correlation for one or more of: probability of success and/or success indicator, time expenditure, energy expenditure, in particular a success correlator and/or an energy expenditure correlator.

According to one development, the method according to the invention, in particular the control method, particularly preferably within step (c) of the control method, comprises the comparison of the determined event sequences with and/or the database also includes a success correlation table and/or event sequences/sub-procedure synergy table , which is suitable for recording and/or managing correlators, in particular those which quantify a correlation for one or more of: probability of success and/or success indicator, time expenditure, energy expenditure, in particular success correlators and/or an energy expenditure correlator of event sequences/sub-procedures .

According to one development, the method also includes a step of prematurely terminating (S03c) a search (S03b-01) for matching event sequences, which is carried out if at least a predetermined safety time period has elapsed since the start of the search (S03b-01) and up to now no suitable result was found, in particular no result which also meets a predetermined suitability criterion, was found, in particular the predetermined safety time period depends on the purpose of the application and/or the type of system to be controlled and/or the adaptive control, the safety time period In particular, it is also suitable for quantifying a period of time up to which an action is required for a smooth running of a process and/or the event sequence.

This avoids hesitation, which can be damaging in dynamic systems (e.g. in an automotive traffic control system). By choosing the desired application-specific safety period, the special needs of the respective application can be addressed in a targeted manner (see also the respective specific applications in particular).

According to a development, the method according to the invention, in particular the control method, particularly preferably within step (c) of the control method also includes a step of comparing an effort indicator and/or cost indicator, in particular one Energy expenditure, the sequence of events with an energy reserve of a technical facility, in particular with an energy reserve allocated to the process, the system or the control system and/or present in the system or the control system.

According to one development, the method also includes a step of carrying out a complete or partial handover (S03c-l) of control over the process and/or the system to a control system, in particular a conventional adaptive control system, which is carried out after the step of premature termination has been carried out becomes.

In this way, the problem is solved and it is ensured that there is no detrimental delay. In addition, ITSF is given a one-time demonstration of how to solve the problem, which means that ITSF continues to learn.

According to one development, the method also includes the following steps: a step of recording the processes (event sequences) of the classic control system, in particular the conventional adaptive control system, which transfer the system from a start object situation to a target object situation, and/or a step of a Filing at least one new entry in the database for the recorded processes, specifying at least one indicator of the start object situation and one indicator of the target object situation.

In this way, ITSF saves the problem solution that has already been practiced and saves it in the database ready to be called up. By storing it in the database, ITSF will be able to use this sequence of events in the future as a solution to a problem or as a solution to a sub-problem. In this way, ITSF can solve the problem in the future, which will relieve the adaptive control in the future. In addition, ITSF can use problem solving as a building block for solving even more complex problems.

Thus, by storing a valuable individual building block, large areas of new, complex ITSF procedures can be opened up under certain circumstances.

According to one development, the method also includes a step of comparing an effort indicator and/or cost indicator, in particular the energy expenditure, of the event sequence with an energy reserve of a technical device, in particular with a Energy reserve allocated to the process, the system or the control system and/or present in the system or the control system.

This allows efficiency to be optimized. In particular, user requests/existing resources/external constraints and/or a combination thereof, in particular also a weighted combination of these factors, can be fully addressed. In this way, ITSF brings the result that is optimally tailored to the specific situation and the specific circumstances.

According to a further development, the memory of the industrial robot also includes instructions for filling, in particular initial filling, of the database by using the classic control, in particular an adaptive control, with the industrial robot being set up in particular after initial filling so that only fewer than two event sequences per Tenths of a second are needed for full-load operation of the industrial robot, more particularly less than one event sequence per tenth of a second, more particularly less than three event sequences per second, and/or the classic control, in particular the adaptive control, is relieved.

The ITSF is already activated by the initial filling and the industrial robot is put into a state in which it already works extremely efficiently with the ITSF and can hand over a large part of the control to the ITSF. Initial training, for example at the customer's site, is no longer required and a direct, efficient and fully functional start is ensured.

exemplary embodiments

The present invention is explained in more detail below with reference to the exemplary embodiments given in the schematic figures of the drawings. It shows as follows:

1a is a schematic diagram for purposes of illustration of the present invention, showing schematically a system suitable for use with the present invention.

Fig. 1b is a schematic diagram for purposes of illustration of the present invention, showing schematically another system suitable for use with the present invention. 2 shows a schematic representation of a table (list of object types) for the purpose of illustrating the present invention within the scope of an exemplary embodiment which uses an entity relationship model,

3 shows a schematic representation of a table (list of object situations) for the purpose of illustrating the present invention within the framework of an exemplary embodiment which uses an entity relationship model,

4 shows a schematic representation of a table (list of procedural event sequences) for the purpose of illustrating the present invention within the framework of an exemplary embodiment which uses an entity-relationship model,

5 shows a schematic representation of a database entry for the purpose of illustrating the present invention,

6 shows a schematic representation of a database entry for the purpose of illustrating the present invention,

7 shows a schematic representation of a database entry for the purpose of illustrating the present invention,

8 shows a schematic representation of a database entry for the purpose of illustrating the present invention,

9 shows a schematic representation of a database entry for the purpose of illustrating the present invention,

10 shows a schematic representation of a database entry for the purpose of illustrating the present invention,

11 shows a schematic representation of a knowledge process flow for the purpose of illustrating the present invention.

The figures merely illustrate examples of possible embodiments and aspects of the present invention.

Figure 1a shows a schematic diagram for purposes of illustration of the present invention, showing schematically a system suitable for use with the present invention. A computer 100 with a memory 101 for a database is in principle suitable for using the method of the invention.

Figure 1b shows another schematic diagram for purposes of illustration of the present invention, showing schematically another system suitable for use with the present invention. In this example, the database is in a cloud or a distributed system, such as a distributed computer network. Calculation steps of the methods can also be carried out in this distributed system.

A computer program according to the invention can also be transmitted wired or wirelessly via such a computer network.

FIG. 2 shows a schematic representation of a table (list of object types) for the purpose of illustrating the present invention within the framework of an exemplary embodiment which uses an entity relationship model.

In one example, a list of object types, for example in the form of a database table, is used. For example, an object type 110 is assigned to a content 111 of the object type, and an identifier 112 is also assigned. In this example, the identifiers 112 are assigned consecutively numerically, but this does not have to be the case.

FIG. 3 shows a schematic representation of a table (list of object situations) for the purpose of illustrating the present invention within the framework of an exemplary embodiment which uses an entity-relationship model.

In an example, a list of the object situations, for example in the form of a database table, is used. For example, the IDs 113 of the object types of the objects that are contained in the respective object situation 114 and/or were recognized are assigned to the object situations 114 . A content 115 of the object situation can also be assigned. In addition, identifiers 112 can be assigned. In this example, the identifiers 112 are assigned consecutively numerically, but this does not have to be the case.

As will become more clear in the following discussion, such a database table facilitates the handling and administration of relevant procedural event knowledge. In addition, a further profitable level of abstraction is created by the object types compared to the concrete objects. This further increases the reusability of procedural event knowledge. For example, such an existing procedure can be applied to another concrete object, but of the same object type.

This increases the efficiency of the overall system. A greater proportion of tasks can be taken over directly by ITSF and the learning curve, especially the initial learning curve, is steeper.

FIG. 4 shows a schematic representation of a table (list of procedural event sequences) for the purpose of illustrating the present invention within the framework of an exemplary embodiment which uses an entity relationship model.

In one example, a list of procedural event sequences, for example in the form of a database table, is used. For example, an event sequence 119 is assigned an ID 118, information about a start object situation, for example a corresponding ID 117, and information about a target object situation, for example a corresponding ID 124, and corresponding content 120.

The event sequence can use its content 120 to transfer the start object situation (ID 117) to the target object situation (ID 124). For example, the content includes instructions in a programming and/or description language. In this or another example, the content contains further references and references to other content.

Additional sizes and identifiers can be assigned. For example, a success indicator 121, a duration indicator 122 and an effort indicator 123 are assigned in this way.

FIG. 5 shows a schematic representation of a database entry for the purpose of illustrating the present invention. (Target OS ID a = Start OS ID z).

The concatenation of two event sequences shown here creates a new event sequence. Only an example is shown. The sequence F, which converts the object situation "7" into the object situation "6", is connected to the sequence G, which converts the object situation "6" into the object situation "8". This creates a new sequence of events which is suitable for converting object situation "7" into object situation "8".

Such an event sequence can also be stored in the list (cf. FIG. 4). FIG. 6 shows a schematic representation of a database entry for the purpose of illustrating the present invention. FIG. 7 shows a schematic representation of a further database entry for the purpose of illustrating the present invention. FIG. 8 shows yet another schematic representation of a database entry for the purpose of illustrating the present invention. FIG. 9 again shows a further schematic representation of a database entry for the purpose of illustrating the present invention.

The sequences in FIGS. 6-9 can be concatenated according to the invention, since target OS ID a=start OS ID xl, target OS ID xl=start OS ID xn, target OS ID xn=start OS ID z.

FIG. 10 shows a schematic representation of a database entry for the purpose of illustrating the present invention. (Target OS ID a = Start OS ID xl, Target OS ID xl = Start OS ID xn, Target OS ID xn = Start OS ID z)

The concatenation creates a new sequence of events, which is suitable for converting the object situation "5" into the object situation "32". For this purpose, the sequences of FIGS. 6-9 are concatenated according to the invention.

Such an event sequence can also be stored in the list (cf. FIG. 4).

FIG. 11 shows a schematic representation of a knowledge process flow for the purpose of illustrating the present invention.

The top five steps or units 1001-1005 relate to the starting object situation. The three steps or units 1006-1008 relate to the target object situation. The steps or units 1010 - 1015 relate to the event sequences, i.e. the procedural event knowledge.

New procedures are added here as required, and new concatenations are generated and saved for further direct use.

It is particularly important to write back the found concatenations of event sequences to the list of collected event sequences (1014). With this, an ever increasing complexity in the list of event sequences develops fully automatically.

FIG. 12 shows an example of a concatenation of event sequences. The sequences a, xl, xn and z are stored in a database. Each event sequence comprises an identifying feature (Start OS ID) of a starting object situation and an identifying feature of a target object situation (Target OSID). The sequences are each represented by a puzzle piece, the left and right sides of which represent different shapes, each representing a specific starting and target object situation. For example, the event sequence xl has a target object situation with an ID of 24, while the event sequence has a start object situation with an ID of 24. Therefore, these two object situations can follow one another. Finally, a concatenation of the event sequences is shown, as a result of which an event sequence was formed that is suitable for reaching the target object situation with ID 32 starting from the start object situation of event sequence a with ID 5. Below the puzzle pieces of the event sequences are shown, whose start and end object situations fit together due to the same start and object situations.

Reference List

100 computer system

101 storage with database

110 Object Kind (OK, object type)

111 Object Kind Content

112 Object Child ID

113 object child IDs

114 Object Situation (OS, object situation)

115 Object Situation Content

116 Object Situation ID

117 boot OS ID

118 Event Sequence ID

119 Event Sequence (ES, Event Sequence)

120 Event Sequence Content

121 success rating

122 Duration

123 Effort

124 Target OSID

1001 Object Analysis and Identification

1002 Object Kind Identification for all Objects in Space Section

1003 Analysis Object Situation in Space Section

1004 Object Situation already in List or add a new Object Situation 1005 Defined Start Object Situation in Space Section, moment of now

1006 Defined Parameters for Target Object Situation

1007 Calculated Target Object Situation for the next moment

1008 Defined Target Object Situation in space section, next moment

1010 Start Transcripting Procedural Event Sequences

1011 Storing new Procedural Event Sequence only for the first one time

1012 Calling stored Event Sequence by using Start and Target Object Situations

1013 Concatenation of more than one Procedural Event Sequences with Target-OSa = Start-OSz

1014 Storing a new concatenation of more than one Procedural Event Sequences

1015 Well-defined Procedural Event Sequences from Start-OS to Target-OS

Claims

patent claims

1. Control method for controlling an actuator for converting a start-object situation into a target-object situation by means of a controller, preferably an adaptive controller, comprising: a) determining (SOI) a start-object situation by means of a sensor, b) defining (S02) a Target object situation, c) determining (S03) an event sequence that is suitable for converting the start object situation into the target object situation from a set of known (partial) event sequences, by o Iterative searching (S03a) of known ( Partial) event sequences comprising the

Start and/or target object situation and/or object situations from parts

Event sequences of previous iteration steps, o selecting (S03bl) at least one event sequence for transferring the

start object situation into the target object situation or formation (S03b2) of a new event sequence for transferring the start object situation into the target

Object situation based on the sub-step of the iterative search found

Partial event sequences and their concatenations, d) driving (S04) of the actuator based on the determined event sequence by the

Steering.

2. The method as claimed in claim 1, wherein the step of selecting (S03bl) the event sequence from a set of suitable event sequences on the basis of at least one quantitative

suitability criterion.

3. The method according to claim 2, wherein an effort and / or cost indicator as a quantitative

eligibility criterion is used.

4. The method according to claim 2 or 3, wherein the energy expenditure as a quantitative

eligibility criterion is used.

5. The method according to any one of claims 2 to 4, wherein a success indicator as a quantitative

eligibility criterion is used.

6. The method according to any one of the preceding claims, wherein the determined in step (a).

Sensor data are normalized before they are used to determine the starting object situation.

7. The method according to any one of the preceding claims, further comprising a step for

Saving (S05) the determined event sequence.

8. The method according to any one of the preceding claims, further comprising a step for

Aborting step (c) of the method if no suitable sequence of events for achieving the goal can be determined and/or the controller completes the step of determining a

Event sequence to achieve the target object situation can not perform.

9. The method of claim 8, wherein step (c) of the method automatically after

expiry of a predefined period of time is aborted.

10. The method according to claim 8 or 9, further comprising a step of adding a (partial) event sequence newly determined by the controller to the set of known events

(Partial) event sequences.

11. The method according to any one of the preceding claims, further comprising a step of automatically supplementing the known event sequences with possible further ones

Event sequences while the method is not used to reach a target object situation.

12. System for controlling an actuator for transferring a start object situation into a target

Object situation by means of a controller, preferably an adaptive controller, comprising: a) input means for receiving (SOI) a starting object situation from a sensor, b) input means for receiving (S02) a target object situation, c) computing means for determining (S03) an event sequence that is suitable for the starting

Object situation into the target object situation from a set of known

To convert (partial) event sequences by o Iterative search (S03a) of known (partial) event sequences comprising the

Start and/or target object situation and/or object situations from partial

starting object situation into the target object situation or forming (S03b2) an optimized event sequence for transferring the starting object situation into the

Target object situation based on the sub-step of the iterative search found

(Partial) event sequences and their concatenations, d) output means for outputting a control signal for driving (S04) the actuator based on the event sequence determined by the controller.

13. The system of claim 12, further comprising means for detecting known

Event sequences and/or known object situations and/or assignments between them.

14. The system of claim 12 or 13, further comprising means for storing at least one

Success indicator and/or at least one duration indicator and/or one

Effort indicator and / or at least one cost indicator and / or at least one relevant time for event sequences.

15. Computer program product, comprising instructions which, when the program is executed by a computer, cause this step (c) of the method according to any one of claims 1 to

11 to implement.

16. Computer-readable medium on which the computer program product according to claim 15 is stored.

17. A computer comprising at least one computer-readable medium according to claim 16.

18. Control, preferably an adaptive control, comprising a computer according to the preceding claim.

19. A computer program product, comprising instructions which, when the program is executed by a computer, cause the latter to correspond to results of a computer program

Claim 15 to display in human-readable form, or to convert the results into another (data) format that can be displayed in human-readable form by another computer program product and/or to cause the computer program product to implement the method in accordance with one of the four preceding claims.

20. Industrial robot system comprising at least one industrial robot, at least one

Control system according to one of claims 12 to 14, a sensor for determining a start

Object situation and an actor for transferring a start-object situation into a target

object situation.

21. Vehicle guidance system, preferably for a motor vehicle, in particular driver assistance system or system for partially automated or autonomous driving, comprising a control system according to one of claims 12 to 14, a sensor for determining a start-object situation and an actuator for converting a start-object situation into a target -Object situation.

22. Traffic control system suitable for implementing the method according to one of

Claims 1 to 11, comprising: a large number of motor vehicles, and for each motor vehicle: a first communication interface, in particular wireless interface, which is set up to communicate with other motor vehicles in a first immediate

To communicate around the motor vehicle, a second communication interface, in particular wireless interface, in particular by means of a cellular connection, in particular 5G, for

Communication of all vehicles with one server.

23. Device for robot-controlled process optimization, comprising: a human-machine interface, preferably a desktop environment, preferably one

Desktop environment of a workstation PC comprising a mouse and/or keyboard, a sensor which is set up to detect an object situation of the human-machine

Record interface, a comparison unit, which is set up to compare at least two object situations, a memory and a CPU, which are set up to carry out the control method according to one of claims 1 - 11, wherein the memory can in particular also be provided in a cloud and a database set up for the methods mentioned, wherein a robot can communicate with the cloud via a data interface, in particular wireless data interface, in particular by means of a mobile radio connection, in particular 5G.

24. Method for robot-controlled process optimization, comprising a method according to any one of claims 1 - 11 and/or the system according to any one of claims 12 - 14, wherein in particular the start and target object situations can designate virtual situations, a system being provided at least comprising a man-machine interface, in particular a desktop environment, in particular a desktop environment of a workstation PC comprising a mouse and/or keyboard, a sensor system which is set up to detect an object situation from the man-machine

interface, a memory and a CPU for processing, and wherein the method further comprises at least one step of comparing, in which two necessary object situations are compared with each other.

25. A method for normalizing object types to support the control method according to any one of claims 1 to 11, of a system according to any one of claims 12 to 14, in particular by collecting and/or using declarative object knowledge, further comprising the following

Steps: Initiation (C01) of a spatial consideration,

Determining (C02) a purpose of the spatial consideration in the form of at least one

Purpose of the spatial consideration,

Detection (C03) of at least one specific object in a section of space,

Assigning (C04) a specific object to an object type, in particular a more general object type, depending on at least one purpose of the spatial consideration, and

26. A method for normalizing object situations to support the control method according to any one of claims 1 to 11, a system according to any one of claims 12 to 14 and/or using a method for normalizing object types according to claim 1

25, in particular by collecting and/or using declarative object knowledge, further comprising the following steps:

Initiation (D01) of a spatial consideration,

Determining (D02) a purpose of spatial consideration in the form of at least one

Purpose of the spatial consideration,

Detection (D03) of at least one specific object in a section of space,

Detection (D05) of first information about a location/position, in particular relative

location/position of at least one concrete object in space, and

Determination (D06) of a normalized object situation for the space section under

Use of the object types and the first information.