# US 12,008,295 B2 · Method and device for creating a model of a technical system from measurements

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,008,295 B2
- Document ID: US12008295B2
- Status: Granted
- Kind: B2
- Application number: 17/157,144
- Filed: 2021-01-25
- Published: 2021-08-26
- Granted: 2024-06-11
- Earliest claimed: 2020-02-24
- Assignees: Robert Bosch GmbH
- Applicants: Robert Bosch GmbH
- Inventors: Andrej Junginger; Holger Ulmer; Jens Stefan Buchner; Patrick Engel; Sebastian Boblest
- Web page: https://patentgrep.com/patent/US12008295B2

## CPC

- G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30)
- B60R 16/0231: {Circuits relating to the driving or the functioning of the vehicle (monitoring tyres B60C23/00; indicating overspeed B60K31/00; for dash boards B60K37/00, B60Q3/10; for indicating emergencies B60Q1/52; brake control systems B60T; registering or indicating the working of vehicles G07C5/00; measuring distance G01C, e.g. combinations of speed and distance G01C23/00; engine indicators G01L; measuring speed or acceleration G01P)}
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06F 2111/10: Numerical modelling
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- B60R: VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR

## Abstract

A method for creating a model of a technical system as a function of measured sensor data of the technical system. The method includes the following steps: initializing a symbolic regression problem. A list of mathematical functions is established, including at least one linear and/or non-linear function and/or at least a one-dimensional parameterizable characteristic curve. The at least one-dimensional characteristic curve is implemented by a Smoothed Grid Regression (SGR) model. Solving the symbolic regression problem with the aid of a genetic algorithm.

## Claims

### Claim 1

A computer-implemented method for creating a model of a technical system as a function of measured sensor data of the technical system, the method comprising the following steps:
providing the measured sensor data, which have been detected using different sensors;
initializing a model generator;
providing a list of mathematical functions including at least one linear and/or non-linear function and an at least one-dimensional parameterizable characteristic curve, the at least one-dimensional characteristic curve being implemented by a Smoothed Grid Regression Model (SGR);
combining, using a genetic algorithm, multiple models with the mathematical functions from the list;
optimizing each of the combined models in such a way that the models are able to calculate one of the measured sensor data as a function of others of the measured sensor data;
assessing, by the genetic algorithm, each of the optimized models as a function of a fitness function, the fitness function being a function of an efficiency of the model to be assessed and a function of a complexity of the model; and
outputting best ones of the models with respect to the fitness function as a function of their complexity;
storing one of the output models on a control system including one or more processors;
controlling the technical system to perform a physical function by activating an actuator of the technical system using the stored model, the technical system performing the physical function under control of the activated actuator.

### Claim 2 (depends on 1)

The method as recited in claim 1, wherein the technical system is an electric motor or an internal combustion engine.

### Claim 3 (depends on 1)

The method as recited in claim 1, wherein a local second-order optimizer is used to optimize the combined models with respect to their efficiency.

### Claim 4 (depends on 1)

The method as recited in claim 1, wherein a model is selected from the output best models by a statistical analysis of data residues.

### Claim 5 (depends on 1)

The method as recited in claim 1, wherein a user selects one of the output best models as a function of the fitness function and of the complexity.

### Claim 6 (depends on 1)

The method as recited in claim 1, wherein an input value range of the combined models is adapted during the optimization.

### Claim 7 (depends on 1)

The method as recited in claim 1, wherein an input value range of the combined models, for each iteration step, is adapted during the optimization.

### Claim 8 (depends on 1)

The method as recited in claim 1, wherein a portion of the mathematical functions of the list is predefined by a user as a function of a priori knowledge about the technical system and/or about the measured sensor variables.

### Claim 9 (depends on 1)

The method as recited in claim 1, wherein the control system ascertains a control variable for the technical system using the stored model.

### Claim 10

A non-transitory machine-readable memory element on which is stored a computer program for creating a model of a technical system as a function of measured sensor data of the technical system, the computer program, when executed by a computer, causing the computer to perform the following steps:
providing the measured sensor data, which have been detected using different sensors;
initializing a model generator;
providing a list of mathematical functions including at least one linear and/or non-linear function and an at least one-dimensional parameterizable characteristic curve, the at least one-dimensional characteristic curve being implemented by a Smoothed Grid Regression Model (SGR);
combining, using a genetic algorithm, multiple models with the mathematical functions from the list;
optimizing each of the combined models in such a way that the models are able to calculate one of the measured sensor data as a function of others of the measured sensor data;
assessing, by the genetic algorithm, each of the optimized models as a function of a fitness function, the fitness function being a function of an efficiency of the model to be assessed and a function of a complexity of the model;
outputting best ones of the models with respect to the fitness function as a function of their complexity;
storing one of the output models on a control system including one or more processors;
controlling the technical system to perform a physical function by activating an actuator of the technical system using the stored model, the technical system performing the physical function under control of the activated actuator.

### Claim 11

A system comprising a control system including one or more processors, and a device configured to create a model of a technical system as a function of measured sensor data of the technical system, the device configured to:
provide the measured sensor data, which have been detected using different sensors;
initialize a model generator;
provide a list of mathematical functions including at least one linear and/or non-linear function and an at least one-dimensional parameterizable characteristic curve, the at least one-dimensional characteristic curve being implemented by a Smoothed Grid Regression Model (SGR);
combine, using a genetic algorithm, multiple models with the mathematical functions from the list;
optimize each of the combined models in such a way that the models are able to calculate one of the measured sensor data as a function of others of the measured sensor data;
assess, by the genetic algorithm, each of the optimized models as a function of a fitness function, the fitness function being a function of an efficiency of the model to be assessed and a function of a complexity of the model; and
output best ones of the models with respect to the fitness function as a function of their complexity;
store one of the output models on the control system;
wherein the control system is configured to control the technical system to perform a physical function by activating an actuator of the technical system using the stored model, the technical system performing the physical function under control of the activated actuator.

### Claim 12 (depends on 1)

The method as recited in claim 1, wherein the technical system is an at least semi-autonomous robot or a manufacturing machine or a medical imaging system.

### Claim 13 (depends on 1)

The method as recited in claim 1, wherein the control system further includes a sensor.