EP4591227A1 - Sichere steuerung von technisch-physikalischen systemen - Google Patents
Sichere steuerung von technisch-physikalischen systemenInfo
- Publication number
- EP4591227A1 EP4591227A1 EP23768194.5A EP23768194A EP4591227A1 EP 4591227 A1 EP4591227 A1 EP 4591227A1 EP 23768194 A EP23768194 A EP 23768194A EP 4591227 A1 EP4591227 A1 EP 4591227A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- areas
- computer
- implemented method
- machine
- input vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/091—Active learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
Definitions
- Various examples concern techniques to enable control of technical-physical systems.
- Various examples concern techniques to estimate a hidden system state of such technical-physical systems and, if applicable, to determine an associated uncertainty.
- Machine-learned algorithms are used in various application areas because the corresponding tasks to be solved by the machine-learned algorithm can be defined by a description of requirements.
- machine-learned algorithms can be used to control technical-physical systems (i.e. technical systems).
- the machine-learned algorithm can predict a hidden system state of the technical-physical system (inference).
- machine-learned algorithms can be used to control a vehicle, such as a motor vehicle or a train.
- Machine-learned algorithms can be used to monitor and, if necessary, adjust maintenance intervals for technical-physical systems such as vehicles or infrastructure (e.g. rails, switches, means of passenger transport such as escalators or treadmills, production facilities, etc.).
- technical-physical systems such as vehicles or infrastructure (e.g. rails, switches, means of passenger transport such as escalators or treadmills, production facilities, etc.).
- the machine-learned algorithm therefore determines one or more output features (typically in the form of an output vector) given the presence of input features (which are typically passed to the machine-learned algorithm in the form of an input vector).
- the machine-learned algorithm is trained based on one or more training data sets.
- a training data set includes several training data pairs of an input vector and associated ground truths.
- the input vector comprises several entries, each of which encodes characteristics of status data of the technical-physical system.
- the ground truths describe a target position in the vector space of the output vector of the machine-learned algorithm. Training the machine-learned algorithm ensures that the output vector obtained from the machine-learned algorithm matches the target position in vector space as closely as possible.
- Examples include: B. autonomous vehicles that take part in traffic.
- a machine-learned algorithm that determines output vectors based on input vectors can be used.
- the input vectors can each encode status data of a technical-physical system.
- the output vectors can predict a respective hidden system state of the technical-physical system.
- the machine-learned algorithm can predict discrete system states (classes), i.e. implement a classifier. In some examples, regression could also be performed by the machine-learned algorithm.
- the input vectors can have two or more dimensions.
- the output vectors can have two or more dimensions.
- the position of an input vector in the corresponding vector space defines the position of the associated output vector in the corresponding vector space (output vector space).
- context data is used for the machine-learned algorithm in inference.
- the context data provides supplementary information to the machine-learned algorithm.
- the contextual data is determined in the context of training the machine-learned algorithm.
- the contextual data can help determine a reliability of an estimate of the hidden system state by the machine-learned algorithm.
- the information content of the context data may vary.
- the context data are, for example, indicative of areas in the input vector space of the machine-learned algorithm, i.e. define hypervolumes of the input vector space. It can be checked how the input vector is positioned with respect to these areas. From this it can be deduced whether there is a reliable or unreliable prediction by the machine-learned algorithm for this input vector.
- the areas can e.g. B. can be determined as a Delaunay triangulation of the input space, so that input vectors of training data pairs of a training data set form corners of the respective surfaces (to be precise, hypersurfaces of any dimensionality; but for the sake of simplicity reference is only made to surfaces). This means that the areas interpolate before between the input vectors for which ground truths are also available within the training data pairs.
- the context data may also include a partitioning of the regions into those associated with low uncertainty in estimating the hidden system state and those associated with high uncertainty in estimating the hidden system state. Alternatively, such a partitioning could still be created for inference.
- the context data can index one or more permitted results for the hidden system state for at least some areas. So in other words, this means that the ranges can indicate which output vectors of the machine-learned algorithm are allowed.
- Such contextual data can therefore be used in inference - when no ground truth is available - to check the output of the machine-learned algorithm for plausibility.
- the contextual data can be determined in the context of training the machine-learned algorithm. For example, it would be conceivable that the context data is determined based on input vectors of a training data set. For example, the above-mentioned areas indexed by the context data can be determined by interpolation between positions of the input vectors of the training data set.
- the context data cannot only be used in connection with inference. For example, it would also be possible to use the context data to determine for which positions in the input vector space additional training data pairs - that This means that combinations of an input vector and an output vector describing a ground truth are required.
- the technical-physical system can be controlled in a particularly safe manner. If, for example, it is recognized based on the control data that an output of the machine-learned algorithm is subject to a comparatively high level of uncertainty, the permissible control actions can be restricted and/or the technical-physical system can be transferred to a safe operating mode and/or a warning message can be issued. In addition, the training of the machine-learned algorithm can be particularly targeted, so that reliable estimates of the hidden system state can be obtained more frequently.
- a machine-learned algorithm is used for this estimation.
- the estimation of the machine-learned algorithm for the hidden system state is either allowed or blocked, each for a corresponding input vector. This is based on at least one predetermined criterion.
- the at least one predetermined criterion takes into account a position of the input vector in the corresponding input vector space. This means that depending on the position of the input vector in the input vector space, the output of the machine-learned algorithm can be prevented. This can result in unsafe expenses or Uncertain estimates for the hidden system state can be avoided.
- a computer-implemented method for estimating a hidden system state of a technical-physical system using a machine-learned algorithm includes obtaining an input vector.
- the input vector encodes state data of the engineering/physical system.
- the method also includes loading the machine-learned algorithm.
- the method further includes loading predetermined context data.
- the predetermined context data is associated with the machine-learned algorithm.
- the predetermined context data indexes a plurality of regions in the vector space of the input vector.
- the method also includes determining a relative position of the input vector with respect to the plurality of regions.
- the method further includes optionally processing the input vector in the machine-learned algorithm to thereby estimate the hidden system state or outputting an error indicator. The selection between processing and outputting the error indicator is based on the relative position of the input vector with respect to the plurality of regions.
- a device comprises at least one processor and a memory.
- the at least one processor is configured to load program code from the memory and to execute the program code.
- the at least one processor is configured to execute a method for estimating a hidden system state of a technical-physical system using a machine-learned algorithm based on the program code.
- the method comprises obtaining an input vector.
- the input vector encodes state data of the technical-physical system.
- the method also comprises loading the machine-learned algorithm.
- the method further comprises loading predetermined context data.
- the predetermined context data is associated with the machine-learned algorithm.
- the predetermined context data indexes a plurality of regions in the vector space of the input vector.
- the method also comprises determining a relative position. tion of the input vector with respect to the plurality of regions.
- the method further comprises optionally processing the input vector in the machine-learned algorithm so as to estimate the hidden system state or outputting an error indicator. The selection between processing and outputting the error indicator is based on the relative position of
- a computer program or a computer program product or a computer-readable storage medium includes program code.
- the program code can be loaded and executed by at least one processor.
- the processor executes the program code, this causes the processor to execute a method of estimating a hidden system state of a technical-physical system using a machine-learned algorithm.
- the method involves obtaining an input vector.
- the input vector encodes state data of the technical-physical system.
- the process also includes loading the machine-learned algorithm.
- the method further includes loading predetermined context data.
- the predetermined context data is associated with the machine-learned algorithm.
- the predetermined context data indexes a variety of regions in the vector space of the input vector.
- the method also includes determining a relative position of the input vector with respect to the plurality of regions.
- the method further includes selectively processing the input vector in the machine-learned algorithm to estimate the hidden system state or outputting an error indicator.
- the choice between processing and outputting the error indicator is based on the relative position of the input vector with respect to the plurality of regions.
- a computer-implemented method for parameterizing a machine-learned algorithm is disclosed. Using the machine-learned algorithm, a hidden system state of a technical-physical system can be estimated. The method includes obtaining a training data set. The training data set includes multiple training data pairs, which in turn include input vectors and ground truths. The input vectors encode state data of the technical-physical system. The basic truths concern the hidden system state of the technical-physical system. The method also includes carrying out training of the machine-learned algorithm.
- Training is performed based on the training data set. Performing training produces parameter values for the machine-learned algorithm.
- the method further includes determining a plurality of regions based on the input vectors. The variety of regions includes regions that are defined in the vector space of the input vectors.
- the method further includes generating context data for the machine-learned algorithm such that the context data is indicative of the plurality of areas.
- the method also includes storing the machine-learned algorithm and storing the context data associated with the machine-learned algorithm.
- a device includes at least a processor and a memory.
- the at least one processor is set up to load program code from memory and to execute the program code.
- the at least one processor is set up to execute a method for parameterizing a machine-learned algorithm based on the program code.
- a hidden system state of a technical-physical system can be estimated.
- the procedure includes obtaining a training data tensentence.
- the training data set includes multiple training data pairs, which in turn include input vectors and ground truths.
- the input vectors encode state data of the technical-physical system.
- the basic truths concern the hidden system state of the technical-physical system.
- the method also includes carrying out training of the machine-learned algorithm. Training is performed based on the training data set.
- the method further includes determining a plurality of regions based on the input vectors.
- the variety of regions includes regions that are defined in the vector space of the input vectors.
- the method further includes generating context data for the machine-learned algorithm such that the context data is indicative of the plurality of areas.
- the method also includes storing the machine-learned algorithm and storing the context data associated with the machine-learned algorithm.
- a computer program or a computer program product or a computer-readable storage medium includes program code.
- the program code can be loaded and executed by at least one processor.
- the processor executes the program code, this causes the processor to execute a method of parameterizing a machine-learned algorithm.
- a hidden system state of a technical-physical system can be estimated.
- the method includes obtaining a training data set.
- the training data set includes multiple training data pairs, which in turn include input vectors and ground truths.
- the input vectors encode state data of the technical-physical system.
- the basic truths concern the hidden system state of the technical physical system.
- the method also includes carrying out training of the machine-learned algorithm. Training is performed based on the training data set. Performing training produces parameter values for the machine-learned algorithm.
- the method further includes determining a plurality of regions based on the input vectors.
- the variety of regions includes regions that are defined in the vector space of the input vectors.
- the method further includes generating context data for the machine-learned algorithm such that the context data is indicative of the plurality of areas.
- the method also includes storing the machine-learned algorithm and storing the context data associated with the machine-learned algorithm.
- FIG. 1 schematically illustrates a system comprising a data processing system and a technical-physical system according to various examples.
- FIG. 2 schematically illustrates a data processing pipeline with a machine-learned algorithm and associated context data according to various examples.
- FIG. 3 illustrates details related to the context data indexing multiple regions in a vector space of input vectors for the machine learned algorithm.
- FIG. 4 is a flowchart of an example method.
- FIG. 5A is a flowchart of an example method.
- FIG. 5B shows aspects related to a Delaunay triangulation.
- FIG. 5C shows aspects related to a Delaunay triangulation.
- FIG. 5D shows aspects related to min-max triangulation.
- FIG. 5E shows aspects related to regions in the input vector space associated with different classes.
- FIG. 5F shows aspects related to regions in the input vector space associated with different classes.
- FIG. 5G shows aspects related to regions in the input vector space that are associated with different classes.
- FIG. 5H shows aspects related to regions in the input vector space that are associated with different classes.
- FIG. 51 shows aspects related to the plausibility check of context data.
- FIG. 6 illustrates input vectors of a training data set in a corresponding vector space.
- FIG. 7 illustrates a classification result of a machine-learned algorithm based on the training data set from FIG. 6 was trained according to various examples.
- FIG. 8 illustrates a classification result of a machine-learned algorithm based on the training data set from FIG. 6 was trained according to various examples.
- FIG. 9 illustrates a classification result of a machine-learned algorithm based on the training data set from FIG. 6 was trained according to various examples.
- FIG. 10A illustrates several regions determined in the vector space of the input vectors of FIG. 6 according to various examples.
- FIG. 10B illustrates details of one of the areas of FIG. 10A.
- FIG. IOC illustrates an input vector space according to various examples.
- FIG. 10D illustrates a region in the input vector space shown in FIG. IOC is determined.
- FIG. 10E illustrates another region located in the input vector space shown in FIG. IOC is determined.
- FIG. 11 is a flowchart of an example method.
- FIG. 12 illustrates context data according to various examples.
- FIG. 13 illustrates context data according to various examples.
- the estimation of the hidden system state could be used to control the engineering-physical system. This means that the operation of the technical-physical system can be adjusted accordingly based on the prediction of the ML algorithm.
- the techniques described herein can be used in various areas of application, i.e. for different technical-physical systems.
- the techniques described herein could be used to control vehicles, e.g. B. in rail-bound transport, public road traffic, in industrial plants or buildings as well as ships, aircraft or spacecraft.
- the techniques described herein could be used to provide maintenance information to adapt intervals for traffic routes such as rail sections or switches.
- the techniques described herein can be used to detect fault conditions of rail vehicles or track infrastructure or to allow the release of certain rail/track sections to road users or to control traffic in transport networks.
- the techniques described herein could be used to monitor the operation of production lines or industrial plants, such as power plants or turbines, to control the distribution of energy or material flows, or to control technical components (e.g. surge protection), people (e.g . in industrial production), to protect company assets (e.g. in economic transactions) or the environment (e.g. from fires).
- the various techniques described herein use a machine-learned algorithm.
- the machine-learned algorithm could be a classifier, that is, mapping continuous values of an input vector to discrete classes.
- the machine-learned algorithm could also provide regression, that is, have continuous values of an output vector.
- a deep neural network could be used, for example with convolutional layers with a trained convolution kernel.
- a "support vector machine” (SVM) could also be used.
- SVM support vector machine
- the input space is divided into classes in such a way that the widest possible area between the different classes remains free.
- Another machine-learned algorithm that can be used would be the “nearest neighbor” classifier. This can be done to For example, a parameter that describes the number of nearest neighbors to be taken into account is learned as part of the training of the machine-learned algorithm.
- Another type of machine-learned algorithm that may benefit from the techniques disclosed herein is the multilayer perceptron or radial basis function network (RBF network).
- RBF network multilayer perceptron or radial basis function network
- the context data it is possible to use the context data to check whether a prediction of the machine-learned algorithm for a particular input vector is reliable or unreliable.
- the input vector can either be processed with the machine-learned algorithm (provided its output is reliable), or an error indicator could be output.
- the technical-physical system can also be controlled depending on an output of the error indicator. For example, a safe operating mode could be activated. A user review could be triggered. A warning signal can be issued. An autonomously acting functionality such as an economic transaction could be aborted. In this way, the risk to the environment due to an unreliable prediction by the machine-learned algorithm can be avoided.
- Normal operation refers to the state of the system in which the system is not in a safe state, but rather fulfills a safety-relevant utility function; this could e.g. B. be a moving train that does not stop at a safe position at a station where passengers can board and disembark safely.
- this can be done during training - for example when collecting and recording examples, for example.
- the capture of further examples increases the reliability of the prediction by the machine-learned algorithm and the technical system.
- the at least one criterion for blocking or allowing can take into account the position of the corresponding input vector - which forms the basis of the estimate of the machine-learned algorithm.
- the prediction could be blocked if the position of the input vector corresponds to an extrapolation of the machine-learned algorithm based on state data of the technical-physical system that is included in a training data set for the machine-learned algorithm.
- This state data can mark corresponding positions (e.g. closest ones) in the input vector space; that can be indicated by the context data.
- the input vector can have a position in the input vector space that is spaced from areas that extend between positions in the input vector space corresponding to the state data.
- a prerequisite for allowing the estimation of the machine-learned algorithm can be that there is no extrapolation, i.e. there is an interpolation.
- the input vector has a position in the input vector space, which is arranged in a region that extends between the positions of the state data from the training data set (if the input vector space has the dimension D, then each D+l input vectors, which spans a space of dimension D and defines a hypervolume T of the input space.
- the boundary of the hypervolume T is then limited by hypersurfaces, each of which connects three neighbors of the state data. If the input vector lies within a hypervolume, which is represented by examples of the training set as above described is limited, then we speak of an interpolation based on the training data. If this is not the case, it is an extrapolation.
- the longest distance between the two or more state data of the technical-physical system, between which interpo- is not greater than a predetermined threshold value can also be required that an upper threshold value can be required for the side length of the hypersurfaces delimiting a region.
- Another exemplary criterion that can lead to a blocking of the prediction would be falling below a minimum distance to uncertain positions in the input vector space.
- These unsafe positions could e.g. B. based on a course of a class boundary between different predicted classes in the input vector space. The positions can therefore be labeled as unsafe.
- TAB . 1 Different levels of certainty that can be obtained in the context of prediction by a machine-learned algorithm. Such stages can be described in the context of specific regions in the input vector space defined by contextual data. The areas could be divided into different security levels using an evaluation metric. This can be done either for inference or during training. Determining the areas or The application of the evaluation metrics can be carried out in a quality-assured manner, which means that dangerous loss or falsification of the information can be reliably avoided.
- FIG. 1 illustrates aspects related to a system 90.
- the system 90 includes a data processing system 91 and a technical-physical system 95 (e.g. an industrial plant, an air, water, space or ground vehicle, a power plant, a means of passenger transport such as an elevator or an escalator, a manufacturing machine, a factory, a production line, an energy distribution facility or material flows or a system for controlling traffic, to give just a few examples).
- a technical-physical system 95 e.g. an industrial plant, an air, water, space or ground vehicle, a power plant, a means of passenger transport such as an elevator or an escalator, a manufacturing machine, a factory, a production line, an energy distribution facility or material flows or a system for controlling traffic, to give just a few examples.
- the data processing system 91 has a processor 92 and a non-volatile memory 93.
- the data processing system 91 also includes a communication interface 94.
- the processor 92 can communicate with the technical-physical system using the communication interface 94. For example, it would be possible for status data 81 of the technical-physical system 95 to be received via the communication interface 94.
- the processor 92 could, for example, be measured values from a sensor that monitors certain operating parameters of the technical-physical system 95.
- the status data 81 could also be operating values of the technical-physical system 95, for example certain control specifications for actuators of the technical-physical system 95. It is also possible for the processor 92 to transmit control data via the communication interface 94
- the operation of the technical-physical system can be adjusted using the control data 82.
- FIG. 1 While in FIG. 1 only one processor 92 is shown, the data processing system 91 can generally have several processors.
- the processor 92 can load and execute program code from the memory 93.
- the program code may define a machine-learned algorithm that is set up to process input vectors that encode the state data 81.
- the machine-learned algorithm can then provide an output vector that estimates a hidden system state of the engineering-physical system.
- a corresponding data processing pipeline is also shown in FIG. 2 shown.
- FIG. 2 illustrates aspects related to a machine learned algorithm 89.
- the machine-learned algorithm 89 processes an input vector 85 that encodes state data 81 of the technical-physical system 95.
- the machine-learned algorithm 89 then provides an output vector 86 that predicts the hidden system state of the engineering-physical system 95. This corresponds to an inference.
- the verification can be done based on the position of the input vector 85 in the input vector space and using corresponding context data 50.
- the context data 50 is associated with, but different from, the machine-learned algorithm 89. Different machine-learned algorithms 89 have different context data 50.
- the context data 50 can, for example, be determined in connection with the training of the machine-learned algorithm 89, which will be explained in detail later.
- FIG. 3 illustrates aspects related to the position of the input vector 85 in the corresponding input vector space 71 (for illustration, the input vector space 71 is shown two-dimensionally, but can be higher-dimensional; the input vector space is thus strictly speaking a hyperspace).
- the input vector space 71 is shown two-dimensionally, but can be higher-dimensional; the input vector space is thus strictly speaking a hyperspace.
- several regions 72 are also shown in the input vector space 71.
- the regions 72 are separated from one another by hypersurfaces (or lines in the two-dimensional case shown).
- the input vector 85 has a specific position in relation to these regions 72 (in the example in FIG. 3, the input vector 85 is arranged in one of the regions 72).
- the areas can be indexed by context data 50 associated with the machine-learned algorithm 89.
- the context data 50 could indicate the areas between the areas 72.
- the areas can have different security levels in accordance with. TAB . 1 be associated.
- the input vector machine-learned algorithm 89 can be indexed by the context data 50 of the machine-learned algorithm 89.
- the context data 50 can be loaded by the processor 92 from the memory 93, for example.
- the context data 50 can be created in connection with the training of the machine-learned algorithm, i.e. H . when parameterizing the machine-learned algorithm; see FIG. 4.
- FIG. 4 is a flowchart of an example method.
- FIG. 4 illustrates different phases of using one machine-learned algorithm such as the machine-learned algorithm 89.
- the training can be carried out based on a training data set .
- the training data set contains training data pairs , each of which comprises an input vector and an associated output vector .
- the output vector represents the ground truth for the output of the machine-learned algorithm for the respective input vector .
- a ground truth can, for example, be obtained based on manual annotation by an expert with domain knowledge . There are also techniques to automatically obtain ground truths, although this can vary depending on the area of application, e.g. by means of a simulation of the technical-physical system .
- parameter values for the machine learned algorithm are obtained. Different parameter values can be obtained for different types of machine-learned algorithms (for example, for a deep neural network with one convolutional layer, the convolution kernel can be determined).
- the training can be carried out using previously known techniques, which typically include an iterative optimization of the various parameter values to minimize a loss function.
- a gradient descent method for optimization with backward adjustment of the parameters can be carried out.
- the loss function describes a distance of the output vector of the machine-learned algorithm in the current separation state compared to the corresponding vector of the ground truth of a corresponding pair of training data.
- the parameterization 3001 may also include generating the context data 50 for the machine-learned algorithm. These can be determined based on the input vectors of the training data set. This means that the context data is specific to the respective machine-learned algorithm that has undergone a certain training. Different machine-learned algorithms then have different context data (see also vertical dashed arrow in FIG. 2).
- Box 3002 is where the inference takes place. This inference takes into account both the previously trained machine-learned algorithm 89 in order to estimate, if necessary, a hidden system state of the technical-physical system; as well as the context data 50 to determine whether an output of the machine-learned algorithm is secure or unsafe. No ground truth is available as part of the inference in Box 3002. That is, the machine-learned algorithm 89 outputs an output vector 86 that cannot be compared to a ground truth.
- FIG. 5A is a flowchart of an example method.
- FIG. 5A illustrates aspects related to parameterizing a machine-learned algorithm according to box 3001 of FIG. 4.
- the method from FIG. 5A is executed by processor 92 based on program code 12 loaded from memory 93 and then executed by processor 92.
- a training data set is obtained in Box 3505.
- the training data set includes several pairs of training data.
- Each training data pair has a corresponding input vector, corresponding to input vector 85 (see FIG. 2).
- the input vector encodes status data of a technical-physical system.
- a ground truth is assigned to each input vector, for a corresponding hidden system state of the technical-physical system.
- the basic truths are available in the form of output vectors (compare output vector 86 in FIG. 2).
- the training of the machine-learned algorithm is then carried out in box 3510 based on the training data set. This is done to obtain parameter values for the machine learned algorithm. Details about the training have already been described above in connection with Box 3001.
- At least some of the areas are determined in such a way that they are arranged between the positions of the input vectors in the corresponding input vector space. This means that the input vectors each define the outer boundaries of the areas. This means that the areas extend between the input vectors. So - to determine the areas - interpolation is carried out between the positions of the input vectors in the input vector space.
- the areas are based on triangular areas of a triangular network, whose nodes through the input vectors are determined, is formed.
- a triangulation e.g. B.
- a Delaunay triangulation can be used.
- the context data 50 can then be generated, which is indicative of the areas.
- the context data 50 could include the triangular areas of the triangular network.
- the areas can therefore be delimited by surfaces (more precisely: hypersurfaces) that extend between the positions of neighboring input vectors in the corresponding input vector space.
- KD-Trees algorithm can be used.
- Other classic tree structures would be conceivable, such as quadtrees or R-trees.
- Approximate nearest neighbor methods such as locality-sensitive hashing are suitable for particularly high-dimensional data.
- the areas are determined in such a way that a position in the input vector space that is recognized as unsafe is surrounded. If a position is recognized as unsafe, the extent of such an area can be determined based on a specified distance measure (parameter "b").
- the parameter b supports the control of problems at the class boundaries in applications with a strong imbalance with regard to the importance of certain Types of errors.
- An example is a safety-relevant function that regulates driving over a traffic light. Driving over a red light can endanger the life and health of people. However, if a vehicle stops at a wrong traffic light, then there is only an availability problem of the vehicle, since it doesn't always move then.
- the parameter "bk” then defines a "safety distance" from an example from a class "k”. All examples that are less than bk in the input space of an example of the Class “k” removed are then assigned to class “k”. Accordingly, for examples that are less than bk away from an uncertain position, the use of the machine-learned algorithm's prediction is blocked, i.e. H . e.g. B. no prediction is calculated or the prediction is discarded. It is not necessary in all variants that the areas around uncertainly recognized positions are also determined. Sometimes the areas that extend between positions of input vectors could also be evaluated accordingly via an evaluation metric (see Box 3520) (this will be explained later).
- the context data 50 are also indicative of permitted results of the machine-learned algorithm if a corresponding input vector is arranged in one of the areas. These permitted results can be determined in box 3516.
- a special algorithm could be used that determines the permitted results depending on the geometric properties of the respective area.
- a corresponding additional machine-learned algorithm could be used.
- a multilayer perceptron or a support vector machine could be used.
- a classic rule-based algorithm could also be used.
- the allowed results could be determined based on ground truths for the input vectors of the training data pairs used to determine the ranges.
- the corresponding region could be characterized in the context data such that the output vectors of the machine-learned algorithm are either "yellow” or "blue” may display.
- the allowed outcomes of the hidden system state can be determined based on the ground truths associated with the nodes of the triangular mesh.
- the areas could be partitioned into first areas and second areas in box 3520 based on a predetermined evaluation metric, for example according to. TAB . 1, security level I or security level I I.
- a predetermined evaluation metric for example according to. TAB . 1
- security level I or security level I I
- the first ranges in terms of the evaluation metric may be associated with a comparatively low uncertainty for the estimation of the hidden system state by the machine-learned algorithm; while the second areas are associated with a comparatively large uncertainty for estimating the hidden system state in terms of the evaluation metric.
- the context data that is then saved can also be indicative of the partitioning of the areas.
- the context data could e.g. B. the security level according to TAB . Show 1.
- the evaluation metric could include a component that evaluates the variety of areas based on their distance from positions recognized as unsafe. If there is an area z. B. within a distance bk from a position recognized as unsafe, the area can be classified as unsafe.
- the evaluation metric could include a component that partitions the multitude of areas based on an edge length of edges of the areas (see TAB. 1: Security level I vs. II). For example, the longer the individual edges of an area are, there could be a tendency for the corresponding area to be classified as unsafe. This is based on the knowledge that the probability of a significant change in the output of the machine-learned algorithm is greater the larger the areas are, i.e. the greater the edge length of the edges of the area. The longer the edge length, the greater the probability of a double class change along the respective edge, i.e. e.g. B. from a first class to a second class and back.
- a critical threshold a which determines a distance in the input vector space, below which the ML algorithm can generalize safely. This is based, among other things, on the assumption that if the distance is smaller than a in the input vector space, there will be no double change in the output class. Areas that are classified as safe then have edge lengths that are all smaller than the threshold value a. Areas that are classified as unsafe have at least one side edge that is greater than the threshold value a.
- the threshold a could be fixed for the different positions in the input space. But it would be It is also possible that the threshold is chosen as a function of the position in the input space.
- connection vectors starting from one input vector to the other input vector span an e-dimensional hypervolume of the input vector space if the connection vectors are linearly independent of each other, i.e. H . None of the connection vectors can be formed by a linear combination of the other connection vectors. Mathematically, it is then said that the “D+ 1” positions of the input vectors are in a “general position” relative to one another. It may happen that the connection vectors are nonlinearly independent.
- D+1 input vectors do not yet span a D-dimensional hypervolume
- neighboring points in the input space can be added until a D-dimensional space is spanned. Then this set of points defines a hypervolume for which a generalization can be made and an evaluation of the area can be carried out.
- the triangulation divides the input vector space into hypervolumes (or areas), whereby the boundary of a hypervolume is defined by the area (more precisely: hypersurface) that lies between three points of the triangulation.
- the respective hypervolume i.e. the respective area in the input vector space
- D+l points adjacent to the triangulation which are each in a general position relative to one another in accordance with the specifications for the triangulation.
- Different types of triangulations may be used in the various examples described herein. This means that different algorithms can be used to determine triangulation. Depending on the choice of triangulation, different areas can be obtained; e.g. B. is in FIG. 5B shows how four points can be triangulated differently (solid vs. dashed lines).
- An example of a triangulation is the Delaunay triangulation.
- the Delaunay triangulation can be carried out using the Bowyer-Watson algorithm.
- An example of Delaunay triangulation is shown in FIG. 5C shown.
- the Delaunay triangulation is formed with the so-called circumference condition. The following applies: The circumference of a triangle of the network must not contain any other points from the specified point set. In this way, mathematically speaking, “the smallest interior angle across all triangles is maximized”. This means that the smallest angle in the triangles becomes as large as it can be in this triangular network. However, no statement is made about the other angles fen.
- Another example of a triangulation is the so-called min-max triangulation.
- a classifier can be blocked as a result. Accordingly, if there are several possible expenses, there are two options.
- the classifier can be set to favor either the safest or the most useful output. In the second case, you would of course have to define “usefulness” in terms of prioritized expenditure.
- the triangle in the middle representation allows the safe conclusion that the output is "blue”.
- the ML algorithm can then output the output "blue” because this output is done through a secure interpolation since all edge lengths are smaller than a. This variant increases the availability of the system without compromising security.
- a general triangulation without additional conditions can lead to ambiguity.
- the Delaunay triangulation solves the ambiguity with the Circumference condition: There must not be any other points in the circumference of a triangle. Although this reduces the occurrence of very "acute" triangles, it does not necessarily ensure that the edge length is as small as possible.
- additional training data pairs can be collected. For example, if problems arise when setting up the triangulation (e.g. because examples are not in a suitable position relative to one another), then further neighboring points are added to the set under consideration until the input vectors of this set of examples form a space with the same dimension as that Spanning input space. The positions of the input vectors then define the boundary of a hypervolume for which an output and an assessment of the security of the output can be determined (see TAB. 1).
- D + l input vectors of the training data set are all less than the critical threshold a apart, then these examples B define for the hypervolume H, which is enclosed by the hypersurfaces connecting them, a hypervolume for which a reliable statement about the internal system state is possible ( see TABLE 1).
- the predicted The hidden system state may only assume values that can also be assumed by the hypervolume-limiting examples B in the output; or when estimating the hidden system state, the output value can only take values no more than are away from such examples B.
- the approximate function is continuous in this hypervolume H, and in special embodiments it is even Lipschit continuous. If all "D+1" input vectors belong to the same class K (i.e. have the same ground truth K), then, in accordance with the above statements, all other input vectors that lie within the hypervolume can be safely assigned to this class K.
- the evaluation metric carries out the partitioning of the areas based on their edge length.
- the evaluation metric could evaluate the variety of areas based on a local variation of the ground truths. This means that, for example, it can be checked whether and, if so, how much the basic truths change within or at the edge of the respective area. For example, it could be checked whether there is a double change in an output class for a classifier as a machine-learned algorithm. This is based on the knowledge that a greater variation in the ground truth in a certain area in the input vector space of the input vector can be indicative of a corresponding increased uncertainty in the prediction of the machine-learned algorithm.
- the evaluation metric used in Box 3520 is parameterized. This means that there can be a parameter whose value influences the results of the evaluation.
- the parameter value can then be selected depending on the area of application or situation-specific. For example, a given A threshold value for the side edges (the value a) can be selected based on at least one criterion.
- the criterion could, for example, comprise a minimum distance between the different input vectors of the training dataset that are associated with the different system states. This means that an analysis of the distances between the different input vectors in the training dataset can be performed. In particular, it can be checked what the minimum distance is for input vectors that yield different results, e.g. different classes, for the hidden system state. Then the predetermined threshold could be chosen to be smaller than or equal to this minimum distance.
- FIG. 5E A corresponding example is shown in FIG. 5E shown.
- FIG. 5E A corresponding example is shown in FIG. 5E shown.
- 5E is a two-dimensional representation of the input vector space (generally formulated, the input vector space could also have a higher dimensionality) and shows different regions 691-694 in each of which input vectors are arranged that belong to a specific class.
- the input vectors located in region 691 of the input vector space belong to a first class (e.g., "green")
- the input vectors located in a region 692 of the input vector space belong to another, second class (e.g., "red " ) belong .
- FIG. 5E also shows the minimum distance 699 between two positions in the input vector space that are associated with input vectors of different classes (in the example shown in FIG. 5E, this distance 699 is through two input vectors defined in regions 691 and 693).
- corresponding information could be provided by a histogram-like evaluation.
- the distance between all pairs of input vectors that belong to different classes can be plotted in the histogram.
- the smallest distance between two input vectors that are assigned to different classes can then be determined from the histogram.
- a histogram can also be created showing the distance at which a double class change occurs. To do this, it is determined for every two examples whether there is an example that lies between these examples (that is, it is closer to the first two examples mentioned) and that has a different class to at least one of the first mentioned examples.
- This distance 699 can then be selected as a predetermined threshold value between the smallest distance between two examples that are assigned to different classes and the smallest distance with a double class change.
- FIG. 5E is characterized by good separation of regions 691-694 in the input vector space. This means that the distance 699 and thus the threshold value is comparatively large. This makes it possible to make a reliable prediction of the hidden system state of the technical-physical system in comparatively large parts of the input vector space using the machine-learned algorithm.
- the at least one criterion could be determined based on domain knowledge. For example, if the dimensionality of the input vector space is comparatively low, a minimum distance between different classes could be determined based on domain knowledge.
- a minimum distance between different classes could be determined based on domain knowledge.
- Camera-based route detection for rail vehicles could detect features in images that firstly show parallel rails (if necessary corrected by an optical transfer function of a camera), secondly rail sleepers arranged at a certain distance from one another and thirdly the presence of ballast of a defined size between the sleepers regarding .
- the radius of curvature of the rails can be determined via extended parallel linear structures;
- the frequency of the threshold sequence and the size of the ballast can be determined using a Garbor filter. If in this 3-D input space (radius of curvature of long, slender linear parallel objects, frequency of the thresholds, size of the ballast) three examples differed less than the threshold value determined based on domain knowledge, then a route was recognized. It would then be conceivable that the permissible feature range for recognized rails/tracks could be determined based on this document. man's knowledge is determined and the threshold value is chosen based on this.
- the at least one criterion for determining the threshold value could include, for example, a design specification of a development process of the technical-physical system.
- at least one criterion could specify which security level should be achieved for the operation of the technical-physical system. If, for example, a certain security level - for example a probability of a misclassification or an incorrect control action - is required as part of the development process for the technical-physical system, then depending on this specification a stricter or less strict criterion for the Parameterization of the evaluation metric can be used.
- the areas determined in Box 3515 and optionally evaluated in Box 3520 can optionally be checked for plausibility. This means that it can be checked/validated whether the classification into “safe” and “unsafe”, for example, is accurate or incorrect.
- the threshold value for the edge length (i.e. the parameter a) can be checked for plausibility on the basis of one or more further pairs of training data.
- These one or more further training data pairs can therefore go beyond the training data set and have further input vectors at positions that have a distance to a position of an input vector of the training data pairs of the original training data set that is smaller than the minimum distance.
- These additional pairs of training data can then be used to check whether the output of the machine-learned algorithm corresponds to the corresponding basic truth. ness matches. If this is not the case, the plausibility check may produce a negative result.
- FIG. 51 This is also shown in FIG. 51 shown.
- the circles are input vectors of the training data set in a first class (e.g. "green") and the triangles are input vectors of the training data set in a second class
- TAB. 1 Security level I. Areas classified as unsafe (here TABLE 1: Security level II) - too large due to class change at the corner points and/or edge length - are shown without filling. The peripheral region (TAB. 1: Security level III) is also shown; No reliable prediction can be made for corresponding input vectors (diamonds) because this would correspond to an extrapolation based on input vectors of the training data set.
- the plausibility of the classification into the areas identified as safe is then checked using additional training data pairs, i.e. additional input vectors for which ground truths are known. These additional input vectors are identified with the crosses + circle. An error is detected for one of these additional input vectors (highlighted with an arrow): The ground truth indicates the second class, while this input vector is located in an area classified as safe in relation to the first class.
- the feedback loop could be executed via box 3526.
- the threshold for the Side edges (the value a) are reduced and the areas are reclassified.
- one or more additional positions in the input vector space can optionally be determined at which additional training data pairs should be positioned. In other words, this means that it can be determined where additional training data pairs are helpful in promoting a reliable prediction of the hidden system state by the machine-learned algorithm.
- the training data set can therefore be expanded in a targeted manner.
- the evaluation of the areas within the framework of a corresponding partitioning from Box 3520 can be used.
- areas in the input vector space can be determined that are associated with a large uncertainty with respect to a corresponding evaluation metric. It would then be possible for further pairs of training data to be arranged in such a way that they lead to a reassessment of these areas that are subject to great uncertainty.
- the one or more further positions can be chosen in the vector space so that after a redetermination of the large number of areas, taking into account the training data pairs and the further training data pairs, a total size of the first areas increases to the maximum. This could be determined, for example, as part of an optimization algorithm.
- a human-machine interface could be controlled to obtain an annotation of ground truths for the hidden system state for the additional training data.
- the number of annotations required can be comparatively small.
- a software simulation of the technical/physical system could also be set up appropriately to obtain the ground truths for the hidden system state for the further training data pairs.
- box 3510 and subsequent boxes can be performed again (shown by the feedback dashed arrow in FIG. 5A).
- Box 3535 it is then possible to use the trained machine-learned algorithm obtained by executing Box 3510, as well as the context data 50 obtained from Box 3515 and, if applicable. Box 3520 will be received to save.
- FIG. 5A explains aspects relating to the determination of context data indicating areas. Based on these areas, it can be determined whether a prediction from the machine-learned algorithm is reliable or unreliable. This effect of the areas is explained in more detail below using a practical example according to FIGS. 6-10 (here also shown in two dimensions for illustration purposes; in practical examples there is a higher dimensionality of the corresponding input vector space 71).
- FIG. 6 illustrates an example of positioning input vectors 511-513 and 521-523 in input vector space 71.
- the true hidden system state can be divided into two classes, corresponding to areas 531, 532.
- area 531 could correspond to a properly functioning motor for a track switch; while area 532 could correspond to a stalled engine.
- the input vectors 511-513 and 521-523 could correspond, for example, based on measured values for a current flow through the motor as well as measured data from a vibration sensor. This is just a specific example and other examples are possible.
- the input vectors 511-513 and 521-523 are part of a training data set.
- the training data set therefore comprises several training data pairs, each of which includes a corresponding input vector 511-513 and 521-523, as well as an associated output vector as ground truth, which displays a first output class (e.g. "motor defective”) for the input vectors 511-513 and displays a second output class (e.g. “motor not defective”) for the input vectors 521-523.
- a first output class e.g. "motor defective”
- a second output class e.g. “motor not defective
- FIG. 6 also shows another input vector 550.
- This input vector 550 is not part of the training data set. This means that there is no ground truth for the input vector 550 .
- Input vector 550 actually corresponds to the second output class, as do input vectors 521-523.
- different machine-learned algorithms 89 which are trained based on the input vectors 511-513 and 521-523, can make an incorrect prediction for the input vector 550.
- the input vector 550 is in the immediate vicinity of the input vector 513; which was trained during training as belonging to the first output class. So the nearest neighbor classifier would incorrectly predicts the first class based on the input vector 550.
- FIG. 8 shows how such a machine-learned algorithm 89 - trained based on the input vectors 511-513, 521-523 - predicts the class boundary 565. A discrepancy occurs in the area of the input vector 550, resulting in an incorrect prediction.
- FIG. 9 shows the corresponding class boundary 567.
- FIG. 10A shows areas 571-574 resulting from an interpolation between input vectors 511-513 and 521-524.
- the areas are obtained by triangulation, whereby, as already described above, different triangulations can be used.
- Examples are a min-max triangulation or a Delaunay triangulation.
- FIG. 10A shows the critical edge length 590.
- this critical edge length 590 is determined so that it corresponds to the smallest distance between two input vectors 511-513 and 521-524 of the training data set for which a class change is observed (that is between input vector 512 and input vector 523).
- different distance metrics may be used in the various examples described herein. For example, the Euclidean distance can be used. However, the so-called p-norm could also be used.
- the critical edge length 590 can be chosen so that no double class change for the output of the machine-learned algorithm can occur within the corresponding length in the input vector space. See . TAB. 1: Security level I.
- FIG. 10A it can be seen that the areas 571, 572 and 574 have edge lengths that are not below the critical edge length 590. Accordingly, these areas 571, 572, 574 (cf. TABLE 1: Safety level I) are rated as unsafe.
- Safety level I Safety level I
- area 574 in FIG. 10B it is shown enlarged how a different class can occur along the connecting line between the input vectors 524 and 522 in the region 574--1; there is therefore a double class change.
- An input vector that lies within these areas 571, 573, 574 causes an unsafe prediction of the associated hidden system state by means of the machine-learned algorithm.
- An error indicator can then be output, based on which, for example, a warning or a transfer of the controlled technical-physical system to a safe operating state can take place.
- area 573 A different situation exists for area 573 (see FIG. 10A).
- the area 573 only has edges that are shorter than the critical edge length 590. The area 573 can therefore be assessed as safe, see TAB. 1 Security level I).
- the further input vector 550 does not lie within any of the areas 571-573. Of- Half, no reliable prediction can be made for the further input vector 550 using the machine-learned algorithm (see TAB. 1: Security level III) - which is consistent with the observations presented above according to FIG. 7, FIG. 8 and FIG. 9 covers .
- FIG. IOC Another variant is shown in FIG. IOC shown.
- the input vector space 71 is shown there and positions of input vectors 662 for a first class and of input vectors 663 for a second class, as well as the class boundary 660.
- Inference for an input vector 661 could give an incorrect result (first class). This could - if the threshold value for the edge length of the areas that extend between the positions of the input vectors 662 is chosen to be too large (see TAB. 1: Example I) - be incorrectly classified as safe.
- an area 670 - see FIG. 10D - which surrounds the position of the input vector 663 marked with a cross and delimits a distance b to this position.
- the inference for all input vectors (including input vector 661) within this area 670 can then be recognized as unsafe.
- the interpolated areas 671 could also be recognized as unsafe by the evaluation metric because the distance to the position of the uncertain input vector 663 is smaller than b, i.e. smaller than a distance threshold 679.
- the area 670 could be based on, for example a course of the class limit 660 can be determined. For example, where no ground truth is available in training near a known class boundary, a corresponding area 670 could be arranged.
- FIG. 11 is a flowchart of an example method.
- FIG. 11 can implement Box 3002.
- FIG. 11 concerns inference using a pre-trained machine-learned algorithm based on an input vector for which no ground truth is available.
- the method from FIG. 11 are executed by the processor 92 after it loads and executes a corresponding program code from the memory 93.
- an input vector is obtained in box 3005.
- This can encode status data of the technical-physical system.
- the input vector could be determined based on corresponding state data (see FIG. 1: state data 81) received from the technical-physical system.
- the input vector can be loaded from memory.
- the machine-learned algorithm is loaded into box 3010.
- the corresponding program code can be loaded from memory.
- the machine-learned algorithm has already been trained at this point, which means that parameter values have been set.
- techniques such as those described above in connection with box 3001, for example in FIG. 5A and also in connection with FIG. 4 described can be used.
- the contextual data is associated with the machine-learned algorithm.
- the context data indicates multiple regions in the vector space of the input vector. This was discussed above in connection with FIG. 10A discussed.
- the areas can be determined as part of the parameterization of the machine-learned algorithm, as described above in connection with FIG. 5A: Box 3515 described.
- the context data can index regions in the input vector space.
- the context data could, for example, specify the areas directly, i.e. define corresponding boundaries for each area. It would also be conceivable that the context data indicates the key points; and then the areas are determined from this using a Delaunay triangulation or another triangulation or another scheme.
- the context data can indicate the boundaries between the areas.
- the context data can also be optimized.
- the context data can specify the areas in compressed form. For example, areas that have the same properties can be combined in order to determine the hidden system state as efficiently as possible. Such a combination of areas or other optimization/compression of the context data can be shown in FIG. 5A: box 3515.
- the context data can also display additional information beyond the scopes.
- the areas can be divided into first areas and second areas.
- the first areas (see FIG. 10A: areas 573) may be associated with a low uncertainty of estimating the hidden system state using the machine-learned algorithm; while the second areas are associated with a comparatively high uncertainty in estimating the hidden system state (see FIG. 10A: areas 571, 572).
- the context data it is possible for the context data to already include an indicator that indicates a corresponding partitioning of the areas into the first and second areas.
- a corresponding partitioning of the areas can already be carried out in connection with the parameterization of the machine-learned algorithm and then, as part of the inference, this previously occurred partitioning can be used (corresponding aspects were explained, for example, in connection with box 3520 in FIG. 5A).
- the corresponding partitioning is only determined as part of the inference, namely in box 3020, for example.
- each of the areas indexed by the context data can be evaluated based on a corresponding evaluation metric. Corresponding evaluation metrics have already been described above in connection with Box 3520.
- evaluation metrics relate, for example, to an evaluation of the areas based on an edge length of edges of the areas, whereby this edge length can be compared with a size threshold value.
- the size threshold was discussed, for example, in connection with FIG. 10A: critical length 590 discussed.
- the size threshold can be viewed through the context data.
- a further exemplary component of the evaluation metric could, for example, concern the classification of an area as safe or unsafe based on a distance to positions in the input vector space that are recognized as unsafe. Such positions in the input vector space that are recognized as uncertain could, as described above, be determined based on a course of class boundaries.
- the evaluation metric includes several components, these components can be arranged hierarchically. This can mean that the partitioning treats a first criterion with a higher hierarchy than a second criterion (i.e. the first criterion has a higher hierarchy level than the second criterion). For example, such a criterion could be checked first; and then when it is determined that For example, if a certain area is to be classified as unsafe, it may be unnecessary to assess the area with regard to other criteria of the evaluation metric. For example, it may be particularly desirable to primarily check the distance to positions in the input vector space that are recognized as unsafe with regard to a threshold value (the parameter b; distance 679 in FIG.
- Performing partitioning for inference in box 3020 may have certain advantages over performing partitioning during parameterization of the machine-learned algorithm (see FIG. 5A: box 3520).
- the evaluation metric during inference is adjusted based on the current circumstances of the physical-technical system. This means that, depending on the operating situation of the physical-technical system, a different partitioning can be determined for the several areas, so that depending on the operating situation of the physical-technical system, certain areas can be classified as safe or unsafe. In this way, an adapted security level for the control can be achieved based on the machine-learned algorithm, tailored to the current operating situation of the physical-technical system.
- the evaluation metric is parameterized based on an operating mode of the technical-physical system.
- the operating mode can be associated with the status data. This means that the status data (see FIG. 1: status data 81) are currently provided by the physical-technical system and their values depend on which operating mode is activated.
- the evaluation metric could thus be parameterized based on an operational safety level of the function to be provided by the technical-physical system, which is specific for the operating mode. This is based on the knowledge that certain technical-physical systems can be operated in different operating modes, which require different operational safety levels for automatic control.
- An example would be semi-autonomous versus fully autonomous driving. With semi-autonomous controls (e.g. "Level 2" - partially automated driving), the driver retains access to the vehicle, so that the operational safety level for the control compared to fully autonomous controls (e.g. "Level 5" - autonomous driving) can be reduced. Accordingly, it would be conceivable that for the security of the previous say a different evaluation metric is used to partition the areas depending on whether semi-autonomous or fully autonomous driving is implemented based on the prediction of the machine-learned algorithm.
- the areas are divided based on the edge length in comparison to a predetermined size threshold value (see FIG. 10A: critical length 590), it would be conceivable that this edge length would be chosen to be shorter or longer depending on the operating mode of the technical-physical system .
- a larger size threshold then results in a larger number of areas for which the machine-learned algorithm is assumed to make a confident prediction.
- context data indicates further information beyond the areas; An example in particular was described above in which it is possible for the context data to index the partitioning of the areas. In addition to such further information provided by the context data, other information could also be provided by the context data.
- the context data for at least some areas of the multitude of areas index one or more permitted results for the hidden system state. This means that - in addition to indicating the different areas in the input vector space - it can also be specified which permissible output values the output vector may assume, provided the input vector is positioned in a corresponding area.
- FIG. 12 shows a scenario in which the context data 50 describes several areas 771-776 (the areas form a triangular network and are represented by corresponding triangular areas - in FIG.
- box 3025 in FIG. 11 it is then possible to determine the relative position of the input vector with respect to the areas. For example, it could be checked whether the input vector is in a certain range or outside all ranges. For example, if the context data indexes the regions by the interfaces between the regions, appropriate verification could include forming subspaces in the input vector space based on the corresponding interfaces.
- the error indicator in box 3035 could be output exactly when the input vector does not exist in any condition. is richly arranged (in such a case it may be unnecessary to partition the areas). It would also be conceivable that the error indicator is output even if the input vector lies in a range that is associated with a large uncertainty in estimating the hidden system state, for example as indicated by the corresponding indicator 751 in the context data 750; or as determined by the partitioning in Box 3020. If the error indicator is output in box 3035, this can be taken into account in box 3050 to control the technical-physical system. For example, the technical and physical systems could be transferred to a safe operating state. A warning could be issued to an operator of the technical-physical system. An error log file could be supplemented accordingly.
- an output vector is then obtained as the output of the machine-learned algorithm, which is indicative of the hidden system state of the technical-physical system. In some examples, this output could then be used to control the technical-physical system, Box 3050.
- box 3045 it would also be conceivable in box 3045 to display the hidden system state, which is received as output from the machine-learned algorithm in box 3040, based on permitted results for the system state that are indexed by the context data (see FIG. 12: indicator 752). to adapt. For example, averaging could occur, for example when a regression task is performed by the machine-learned algorithm.
- information regarding a probability distribution of the predicted value can also be output. This probability distribution can depend on the distance to areas with indicators of other values to the position of the input vector in the input space.
- FIG. 13 basically corresponds to FIG. 12.
- the context data 50 is structured in such a way that a corresponding indicator 753 is provided in connection with a corner of the areas 771-776, which provides a respective permitted result for the estimated system state or the output of the machine-learned algorithm is indexed. It would then be possible to determine a distance between the position of the input vector in relation to the different corners of the areas 771-776 and to select the respective permitted result based on these distances (or - for a regression task - the probability distribution).
- the corners of the areas 771-776 correspond to the positions of corresponding input vectors of the underlying training data set; the permitted outcomes could be chosen here based on the corresponding ground truths. For example, the nearest corner could be taken into account and then the permitted result for the output of the machine-learned algorithm could be determined based on the indicator 753 associated with this corner (for the input vector 85 shown in FIG. 13, this would be "blue").
- Box 3040 and Box 3045 therefore correspond to a two-stage approach.
- candidate results are determined using the machine-learned algorithm; these can then be corrected if necessary in Box 3045, based on the context data.
- the techniques described above involve generating contextual data in the context of training a machine-learned algorithm.
- the contextual data is indicative of regions in the input vector space that are associated with confident or uncertain predictions from the machine-learned algorithm.
- the areas can be generated, for example, as a Delaunay triangulation between positions of input vectors of a training data set.
- the areas therefore form hyperspace volumes.
- Each hyperspace volume can be assigned as a marker the set of output classes of the examples that form the corner points of the hyperspace volume.
- I is the distance between all points in the input space smaller than "a”
- the "safe" feature is also set for the marking.
- the attribute safe in the marking means that only the outputs that are present in the examples that limit the hyperspace volume can occur. Other outputs cannot occur in this case because of the assumption in “a” that no double class change occurs.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE102022209903.6A DE102022209903A1 (de) | 2022-09-20 | 2022-09-20 | Sichere steuerung von technisch-physikalischen systemen |
| PCT/EP2023/074279 WO2024061615A1 (de) | 2022-09-20 | 2023-09-05 | Sichere steuerung von technisch-physikalischen systemen |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP4591227A1 true EP4591227A1 (de) | 2025-07-30 |
Family
ID=88016250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP23768194.5A Pending EP4591227A1 (de) | 2022-09-20 | 2023-09-05 | Sichere steuerung von technisch-physikalischen systemen |
Country Status (3)
| Country | Link |
|---|---|
| EP (1) | EP4591227A1 (de) |
| DE (1) | DE102022209903A1 (de) |
| WO (1) | WO2024061615A1 (de) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163224A (zh) * | 2011-04-06 | 2011-08-24 | 中南大学 | 自适应空间聚类方法 |
| US9014905B1 (en) * | 2014-01-28 | 2015-04-21 | Google Inc. | Cyclist hand signal detection by an autonomous vehicle |
| US20210304059A1 (en) * | 2020-03-26 | 2021-09-30 | International Business Machines Corporation | Method for selecting datasets for updating an artificial intelligence module |
| EP3944158B1 (de) * | 2020-07-21 | 2025-10-29 | Volkswagen Aktiengesellschaft | Computer-implementiertes verfahren für maschinelles lernen zum betrieb einer fahrzeugkomponente und verfahren zum betrieb einer fahrzeugkomponente |
| DE102021207613A1 (de) | 2021-05-26 | 2022-12-01 | Siemens Mobility GmbH | Verfahren zur Qualitätssicherung eines Systems |
-
2022
- 2022-09-20 DE DE102022209903.6A patent/DE102022209903A1/de active Pending
-
2023
- 2023-09-05 WO PCT/EP2023/074279 patent/WO2024061615A1/de not_active Ceased
- 2023-09-05 EP EP23768194.5A patent/EP4591227A1/de active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024061615A1 (de) | 2024-03-28 |
| DE102022209903A1 (de) | 2024-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DE112017000787T5 (de) | Verfahren zum Steuern der Bewegung eines Fahrzeugs und Fahrzeugsteuersystem | |
| DE102019215902A1 (de) | Verfahren zur Bestimmung eines Gütegrades von Daten-Sätzen von Sensoren | |
| DE102019209462A1 (de) | Verfahren zur Bestimmung eines Vertrauens-Wertes eines detektierten Objektes | |
| DE102020128978A1 (de) | Trainieren von tiefen neuronalen netzwerken mit synthetischen bildern | |
| DE102021106584B4 (de) | System mit mindestens einem Anlagesystem mit mindestens mehreren Anlageteilen | |
| EP4193135A1 (de) | Computerimplementiertes verfahren zum bereitstellen eines test-verlaufs zu testender verkehrsszenarien | |
| EP3639199A1 (de) | Verfahren zum bewerten eines zustands eines dreidimensionalen prüfobjekts und entsprechendes bewertungssystem | |
| EP3961511A1 (de) | Ml-basierte automatische erkennung von neuen und relevanten datensätzen | |
| DE102021213118A1 (de) | Verfahren und ein system für black-box-universalangriffe mit geringer abfrage | |
| DE102019209463A1 (de) | Verfahren zur Bestimmung eines Vertrauenswertes eines Objektes einer Klasse | |
| EP4591244A1 (de) | Sichere steuerung von technisch-physikalischen systemen | |
| DE102020127051A1 (de) | Verfahren zur Bestimmung von sicherheitskritischen Ausgabewerten mittels einer Datenanalyseeinrichtung für eine technische Entität | |
| DE102021209090B4 (de) | Verfahren zur kontextabhängigen Auswertung eines Fahrzeugzustands und Fahrzeug | |
| WO2024061615A1 (de) | Sichere steuerung von technisch-physikalischen systemen | |
| EP4000011B1 (de) | Komponentenbasierte verarbeitung von eingangsgrössen | |
| DE102021207613A1 (de) | Verfahren zur Qualitätssicherung eines Systems | |
| DE102021114191A1 (de) | Verteiltes System | |
| DE102023135881A1 (de) | Systeme und verfahren zum erklären der verhaltensweise einer künstlichen intelligenz mit kausalen schlussfolgerungen | |
| DE102020104267A1 (de) | Computerimplementiertes Regelungs- oder Folgeregelungsverfahren oder Optimierungsverfahren zur Absicherung von Steuerungsalgorithmen einer Regelung und/oder Regelungsalgorithmen | |
| DE102022107820A1 (de) | Verfahren und Prozessorschaltung zum Überprüfen einer Plausibilität eines Detektionsergebnisses einer Objekterkennung in einem künstlichen neuronalen Netzwerk sowie Kraftfahrzeug | |
| DE102019135022B4 (de) | Verfahren zur rechnergestützten Auswertung von Messungen elektrischer Ströme in einem Hochvolt-Bordnetz eines vorgegebenen elektrisch angetriebenen Kraftfahrzeugs | |
| DE102016206627A1 (de) | Verfahren und Vorrichtung zum Vermessen eines zu testenden Systems | |
| EP4113392B1 (de) | Verfahren zum prüfen der zuverlässigkeit einer ki-basierten objekt-detektion | |
| EP3781996A1 (de) | Verfahren, computersystem und computerprogramm zum steuern eines aktors | |
| DE102023212367A1 (de) | System und Verfahren mit Sequenzmodellierung von Sensordaten für die Fertigung |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
| 17P | Request for examination filed |
Effective date: 20250314 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC ME MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| RAP3 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: SIEMENS MOBILITY GMBH |
|
| DAV | Request for validation of the european patent (deleted) | ||
| DAX | Request for extension of the european patent (deleted) |