DE102020132384A1 - System, method and computer program product for configuring an optimized sensor set for a moving object - Google Patents

Abstract

The invention relates to a method for configuring an optimized sensor set for a moving object, comprising: - defining at least one requirement for a sensor set with sensors to fulfill a task function of the moving object; - initializing a population of selected individuals of sensor sets, each individual at least one parameter is assigned;- generating synthetic sensor data for each sensor set within a simulation;- evaluating the quality of the synthetic sensor data of the individuals using an evaluation function for all individuals of the population;- determining whether the algorithm can be terminated;- terminating the algorithm , if at least one defined termination criterion applies;- selecting one or more individuals from sensor sets based on an evaluation result;- combining the parameters of one or more of the selected individuals with one another and generating one or more Ind next generation individuals; - varying the parameters of the new generation individuals by a defined amount and creating a new individual of the new population; - submitting the new population to the initialization module.

Description

The invention relates to a system, method and computer program product for configuring an optimized sensor set for a moving object, in particular a motor vehicle.

The development of highly automated driving (HAD) is accompanied by an increase in the complexity of driving functions, which leads to an increase in the demands on vehicle sensor systems. The perception of the environment through various sensor technologies such as LiDAR, radar, camera and ultrasound plays a decisive role for highly automated driving functions and in the field of industrial robotics.

In addition to the perception of the environment, self-motion and system monitoring sensors such as force/torque sensors, acceleration sensors, temperature sensors or power management sensors provide important information in order to be able to carry out a task correctly and/or to avoid collisions and damage .

In order to ensure that the tasks of a moving object, e.g. a robot or an automated motor vehicle, are successful and safe as the tasks of a moving object increase in complexity, reliable environmental perception as well as self-movement and system monitoring are required. In order to achieve this goal, the number of sensor elements required in a vehicle or robot can increase significantly. In addition, some sensor technologies have advantages and disadvantages that can be compensated for by other sensor technologies. Ultrasonic sensors, for example, have a limited range of around one centimeter to a few meters and a maximum directional cone of around 30 degrees. In addition, they offer a high level of accuracy at close range, which, however, decreases with increasing distance and measuring angle. In addition, they are susceptible to temperature and pressure fluctuations, as well as interference from neighboring robots or automobiles that use ultrasonic sensors with the same operating frequency. Nevertheless, in combination with other presence sensors, they can generate reliable position information. Therefore, the combination of different sensor technologies within the framework of a sensor fusion is crucial, for example the combination of sensor information from a 2D/3D camera, LiDAR sensors, radar sensors and ultrasonic sensors, in order to enable a motor vehicle or a robot to spatially perceive their surroundings to perceive

In order to ensure the required functional quality and safety with this increase in the required sensor elements and the diverse possible combinations of sensor fusion of different sensor technologies while at the same time bearable costs, an optimization of the design of the sensor set, consisting of at least one sensor element, is desirable. Optimization can include the selection of the sensor elements used, their technology, positioning, orientation in a vehicle or robot, as well as their resolution, sampling rate or monitoring area. A genetic algorithm is aimed at as an optimization method.

Sensors in a vehicle or robot are currently designed largely on the basis of empirical values and expert knowledge, so that a general optimization method has not yet been given.

the CN 109917290A describes a temperature determination method for motor vehicle batteries, with temperature measuring points being determined and optimized using genetic algorithms.

the U.S. 600 6604 A describes a method for determining an optimized arrangement of liquid surface sensors for measuring the amount of liquid in a container using genetic algorithms.

the US 789 5021 B1 describes a method for placing a sensor in an environment where optimal values are determined with a genetic algorithm over a set of fitness function parameters.

The object on which the invention is based is now to create a method, a system and a computer program product for configuring an optimized sensor set for a moving object, in particular for a motor vehicle, which is characterized by high reliability and accuracy with simultaneously realizable optimization effort , excellent.

According to the present invention, a method, a system and a computer program product is proposed by which the selection, positioning, orientation of the individual sensors and their resolution, sampling rate and monitoring range for the environmental and/or self-monitoring and/or system monitoring of a moving object is optimized in order to thereby create the basis for a reliable and accurate sensor perception and thus a reliable and accurate functioning of the task fulfillment of the object.

This object is achieved according to the invention with regard to a method by the features of patent claim 1, with regard to a system by the features of patent claim 6, and with regard to a computer program product by the features of patent claim 11. The further claims relate to preferred embodiments of the invention.

• - Definieren von mindestens einer Anforderung an ein Sensorset, bestehend aus einem oder mehreren Sensoren, zur Erfüllung einer Aufgabenfunktion des sich bewegenden Objekts;
• - Initialisieren einer Population von zufällig oder nach bestimmten Kriterien ausgewählten Individuen von Sensorsets, wobei jedem Individuum zumindest ein Parameter zugeordnet wird;
• - Generieren von synthetischen Sensordaten und Referenzinformationen durch eine virtuelle Simulation für alle Individuen der Population;
• - Bewerten der Qualität der vorliegenden synthetischen Sensordaten mittels einer Bewertungsfunktion, in Abhängigkeit der Anforderungen an die zu erfüllende Aufgabe, für alle Individuen der Population;
• - Beenden des Algorithmus, falls zumindest ein definiertes Abschlusskriterium zutrifft;
• - Auswählen von ein oder mehreren Individuen basierend auf dem Bewertungsergebnis der Bewertungsfunktion;
• - Kombinieren der Parameter von einem oder mehreren der ausgewählten Individuen miteinander und Erzeugen von ein oder mehreren Individuen der nächsten Generation;
• - Variieren der Parameter der Individuen der neuen Generation um ein definierbares Maß und Erzeugen eines neuen Individuums der neuen Generation;
• - Übermitteln der neuen Population an das Initialisierungsmodul und Durchführen der vorangehenden Verfahrensschritte mit der neuen Population.

According to a first aspect, the invention provides a method for configuring an optimized sensor set for a moving object, in particular a motor vehicle. The procedure comprises the following procedural steps:

• - defining at least one requirement for a sensor set consisting of one or more sensors to fulfill a task function of the moving object;
• - Initializing a population of individuals from sensor sets selected randomly or according to specific criteria, with each individual being assigned at least one parameter;
• - Generation of synthetic sensor data and reference information through a virtual simulation for all individuals of the population;
• - Assess the quality of the available synthetic sensor data by means of an assessment function, depending on the requirements of the task to be performed, for all individuals in the population;
• - Termination of the algorithm if at least one defined termination criterion applies;
• - selecting one or more individuals based on the evaluation result of the evaluation function;
• - combining the parameters of one or more of the selected individuals together and generating one or more next generation individuals;
• - varying the parameters of the new generation individuals by a definable amount and creating a new new generation individual;
• - Sending the new population to the initialization module and performing the previous method steps on the new population.

A further development provides that a parameter is a property such as technology, installation position, orientation, resolution, sampling rate or monitoring range of a sensor element.

In particular, the sensors are 2D/3D cameras and/or LiDAR sensors and/or radar sensors and/or ultrasonic sensors and/or force/torque sensors and/or acceleration sensors and/or pressure sensors and /or temperature sensors and/or humidity sensors and/or power management sensors.

In one embodiment it is provided that one or more of the following conditions constitute a termination criterion: the evaluation result has reached a target value which is defined by the quality of the synthetic sensor data of an individual being above a threshold value; and/or the individuals of a population reach a convergence value; and/or a given maximum number of iterations has been reached.

In particular, the moving object is a motor vehicle, a watercraft, an aircraft or a robot.

According to a second aspect, the invention provides a system for configuring an optimized sensor set for a moving object, in particular a motor vehicle. The system includes a requirement module which is designed to define at least one requirement for a sensor set consisting of at least one sensor element for fulfilling a task function of the moving object; an initialization module that is designed to initialize a population of individuals from sensor sets selected randomly or according to specific criteria, wherein each individual is assigned at least one sensor element with at least one variable parameter; a simulation module that is designed to generate synthetic sensor data for one or more individuals of a population within a catalog of defined simulation scenarios; an evaluation module that is designed to evaluate the quality of the available synthetic sensor data by means of an evaluation function of one or more individuals in a population; a termination module that is designed to determine whether the algorithm can be terminated and to terminate the algorithm if at least one defined termination criterion applies; a selection module configured to select one or more individuals from a population based on an evaluation result of the evaluation function; a recombination module configured to combine the parameters of at least two of the selected individuals and to generate one or more individuals of the next generation; a mutation module designed to vary the parameters of the new generation individuals by a defined amount and generate a new individual of the new population and the new ones Submit population to the initialization module.

A further development provides that a parameter is a property such as technology, installation position, orientation, resolution, sampling rate or monitoring range of a sensor element.

One embodiment provides that the sensors are 2D/3D cameras and/or LiDAR sensors and/or radar sensors and/or ultrasonic sensors and/or force/torque sensors and/or acceleration sensors and/or pressure sensors and/or temperature sensors and/or humidity sensors and/or power management sensors.

In a further embodiment it is provided that one or more of the following conditions represent a conclusion criterion: the evaluation result has reached a target value which is defined by the fact that the quality of the synthetic sensor data of an individual is above a threshold value; and/or the individuals of a population reach a convergence value; and/or a given maximum number of iterations has been reached.

In particular, the moving object is a motor vehicle, a watercraft, an aircraft or a robot.

According to a third aspect, the invention provides a computer program product comprising executable program code configured, when executed, to carry out the method according to the first aspect.

The invention is explained in more detail below with reference to exemplary embodiments illustrated in the drawing.

• 1 eine schematische Darstellung eines Sensorsets für ein sich bewegendes Objekt;
• 2 ein Blockdiagram zur Erläuterung eines Ausführungsbeispiels des erfindungsgemäßen Systems;
• 3 ein Flussdiagramm zur Erläuterung der einzelnen Verfahrensschritte des erfindungsgemäßen Verfahrens;
• 4 eine schematische Darstellung eines Computerprogrammprodukts gemäß einer Ausführungsform des dritten Aspekts der Erfindung.

It shows:

• 1 a schematic representation of a sensor set for a moving object;
• 2 a block diagram to explain an embodiment of the system according to the invention;
• 3 a flowchart to explain the individual steps of the method according to the invention;
• 4 a schematic representation of a computer program product according to an embodiment of the third aspect of the invention.

Additional features, aspects and advantages of the invention or embodiments thereof will become apparent from the detailed description coupled with the claims.

1 shows a sensor set 10 for a moving object 20. A sensor set 10 is a group of one or more sensor elements a ₁ , a ₂ , ..., _ak ; b ₁ , b ₂ , ..., b ₁ ; ...; n ₁ , n ₂ , ..., n _m one or more sensor types S ₁ , S ₂ , ..., S _n for a moving object 20 for determining sensor data. The moving object 20 can be a motor vehicle, an aircraft such as drones or a watercraft, or a robot, for example in a production facility or a logistics operation.

The various sensors a ₁ , a ₂ , ..., a _k ; b ₁ , b ₂ , ..., b ₁ ; ...; _{n 1} , n ₂ , ), resolution (in number of measuring points), sampling rate (in Hz) or monitoring range (in angles). The sensor elements, which consist of the sensor set 10 and the other electrical and electronic components connected thereto, such as actuators, processors and/or storage elements, are used to acquire measurement data in order to fulfill specific task functions of the object 20 . A task function can be, for example, an autopilot in a vehicle, object recognition or localization or movement sequences of a robot. The sensor set used must therefore be designed and configured in such a way that the sensor elements a ₁ , a ₂ , . . . , a _k ; b ₁ , b ₂ , ..., b ₁ ; ...; n ₁ , n ₂ , ..., n _m , all relevant information required for the respective task function is recorded. It may also be necessary for several task functions to be fulfilled simultaneously, such as monitoring the surroundings of the object 20 and coordinating a movement of the object 20.

The _sensor types S ₁ , S ₂ , , pressure sensors, temperature sensors, humidity sensors or power management sensors that are used for various task functions. Several sensors from different sensor technologies can also be combined with one another in order to fulfill one or more task functions.

According to the invention, to optimize the sensor set used, a ₁ , a ₂ , . . . , a _k ; _b1 , _b2 , ..., b _l ; ...; n ₁ , n ₂ , ..., n _m evolutionary algorithms and in particular genetic algorithms based on a population approach are used to solve an optimization problem using a heuristic search. The system 100 shown in 2 , includes various modules, some of which are equipped with processors and memory units.

In the context of the invention, a “processor” can be understood to mean, for example, a machine or an electronic circuit. A processor can in particular be a central processing unit (CPU), a microprocessor or a microcontroller, for example an application-specific integrated circuit or a digital signal processor, possibly in combination with a memory unit for storing program instructions, etc . A processor can also be understood to mean a virtualized processor, a virtual machine or a soft CPU. It can also be a programmable processor, for example, which is equipped with configuration steps for executing the mentioned method according to the invention or is configured with configuration steps in such a way that the programmable processor has the inventive features of the method, the component, the modules, or other aspects and/or or implemented partial aspects of the invention.

In the context of the invention, a “memory unit” or “memory module” and the like can mean, for example, a volatile memory in the form of a random-access memory (RAM) or a permanent memory such as a hard disk or a data carrier or e.g. B. be understood as a removable memory module. However, the storage module can also be a cloud-based storage solution.

In connection with the invention, a “module” can be understood to mean, for example, a processor and/or a memory unit for storing program instructions. For example, a module is specially set up to execute the program instructions in such a way that the processor and/or the memory unit executes functions in order to implement or realize the method according to the invention or a step of the method according to the invention.

As in 2 shown, the requirements for a sensor set 10 consisting of the sensors a ₁ , a ₂ , . . . _ak ; b ₁ , b ₂ , ..., b ₁ ; ...; n ₁ , n ₂ , ..., n _m defined to fulfill a certain task function. The requirements module 110 can have user interfaces in order to enter the requirements or read them in automatically from a requirements catalogue.

In an initialization module 120, an initial population Ω _t =0 of randomly selected individuals X ₁ , X ₂ , . . . , X _k from sensor sets 10 is generated. Each individual X _i corresponds to a sensor set 10. The individual individuals have a genetic code that can be represented, for example, as a character string and/or as real digits and/or in a tree structure. An example of a character string is a column vector with a number m of elements Ψ ₁ , Ψ ₂ , ..., Ψ _p . The individual elements of a character string represent the sensor parameters, which depict the properties of a sensor element, such as a specific installation position, orientation, resolution, sampling rate or monitoring area.

A population Ω can also be generated in another module and transmitted to the initialization module 120 . An individual X _i thus represents an individual characteristic of a sensor set 10 with certain properties of individual sensors a ₁ , a ₂ , . . . _ak ; b ₁ , b ₂ ,..., _bl ; ...; n ₁ , _{n 2} ,..., nm in terms of technology, installation position, resolution and/or field of view, etc. The initial individuals X ₁ , X ₂ ,..., X _k can be a random selection or represent a current series version and do not have to be optimal.

Then the population Ω is transferred to a simulation module 130 . Each individual X _i runs through a catalog of defined simulation scenarios that best represent the task function. The simulation module 130 generates synthetic sensor data for each individual X _i .

Based on requirements of the task function, an evaluation function F(X _i ) is determined in order to assess the quality of a specific individual X _i of a sensor set 10 consisting of the various sensors a ₁ , a ₂ , ..., _ak ; b ₁ , b ₂ , ..., b ₁ ; ...; n ₁ , n ₂ ,..., n _m . The evaluation function F(X _i ) consists, for example, of weighted properties e _i that correspond to the criteria that are required for an optimized selection and arrangement of an individual X _i of the sensor set 10 . The goal of the algorithm of the system 100 according to the invention is to generate individuals X _i from sensor sets 10 that have a high evaluation result in order to fulfill the respective task function well with regard to the evaluation criteria.

mit
F(X_i): Bewertungsfunktion
w_i: Gewicht
e_i: EigenschaftIn an evaluation module 140, an evaluation is now carried out for each individual X _i of the population Ω using the evaluation function F(X _i ). It is evaluated how well the synthetic sensor data of the individual X _i defined the able to meet the requirements of the task function. Other criteria such as costs, weight, bus utilization, transmittable data rate, etc. of the sensor set X _i are also included in the evaluation. All evaluation criteria are combined into the evaluation function F(X _i ). For example, the evaluation function F(X _i ) can have a linear combination of properties e _i and their weighting w _i : $$f\left(X_i\right)=w_1\ast e_1+w_2\ast e_2+\cdot \cdot \cdot +w_n\ast e_n$$

With
F(X _i ): evaluation function
w _i : weight
_ei : property

A high value for the evaluation function F(X _i ) is a measure of the quality of the respective sensor set X _i . An evaluation result is calculated for each individual X _i of the population Ω using the evaluation function F(X _i ).

One or more individuals X _i from sensor sets 10 are selected in a selection module 150 based on the evaluation result. Various approaches are conceivable here, such as selecting the n-best individuals X _i or all individuals X _i with an evaluation result that is greater than a threshold value according to: $$f\left(X_i\right)>ScHwellenwe\mathrm{right}t$$

The selected individuals X _i are now transferred to a recombination module 160 . In the recombination module 160, the parameters Ψ _i of all selected individuals X _i are combined with one another and one or more individuals Y _i of the next generation are generated. Thus, the selected individuals X _i inherit their parameters Ψ _i to the next generation. Several approaches are conceivable here as to how these parameters Ψ _i can be inherited, for example by combining the parameters of at least two parents. For example, a parent's string Ψ ₁ , Ψ ₂ , ..., Ψ _p may be split at a random location and padded with the parameter values of the other parent. This creates a new individual Y _i that has characteristics of both parents.

Now, in a mutation module 170, the parameters Ψ _i of the individuals Y _i of the new generation are varied by a defined amount. A new individual Y' _i thus arises. The individuals Y' _i now represent the new population Ω _t+1 for a new run through the individual method steps.

The method steps in the initialization module 120, the simulation module 130, the evaluation module 140, the selection module 150, the recombination module 160 and the mutation module 170 are carried out iteratively until a termination criterion defined in a final module 180 is met. For this purpose, several completion criteria can be selected, for example that the evaluation result has reached a target value, which is defined by the fact that the deviation is below a threshold value. Another criterion can be that the individuals of the previous population and the individuals of the next population approach and thus reach a convergence value. Another criterion is that a given maximum number of iterations has been reached. If one or more of these criteria are met, the termination module 180 ends the algorithm.

However, in the event that a closure criterion is not met, a new population Ω _t+1 is created. The new population Ω _t+1 is now transmitted to the initialization module 120 and the method steps are carried out iteratively until a termination criterion is met.

The method according to the invention can be used to optimize entire sensor sets 10, but can also be used to optimize an individual sensor.

It is also conceivable not to provide the method step in the recombination module 160 or the method step in the mutation module 170 .

In 3 the method steps for configuring an optimized sensor set 10 for a moving object 20 are shown.

In a step S10, in a request module 110, at least one request is made to a sensor set 10 with one or more sensors a1, a2, . . . ak; b1, b2,..., _bl ; ...; n1,n2,...,n _m are defined to fulfill a task function of a moving object.

In a step _S20 , a population _Ω of individuals X ₁ , _{X 2} , .

In a step S30, each individual X _i runs through a catalog of defined simulation scenarios in a simulation module 130, which represent the task function in the best possible way. The simulation module 130 generates synthetic sensor data for each individual X _i .

In a step S40, the quality of the synthetic sensor data is evaluated in an evaluation module 140 for each individual X _i of the population Ω using an evaluation function F(X _i ) with regard to the requirements placed on the sensor set for fulfilling a specific task.

In a step S50, a termination module 180 determines whether the algorithm can be terminated.

In a step S60, the algorithm is terminated if at least one termination criterion defined in the termination module 180 applies.

In a step S70, one or more individuals X _i from sensor sets 10 are selected in a selection module 150 based on the evaluation result.

In a step S80, the parameters Ψ _i of one or more of the selected individuals X _i are combined with one another in a recombination module 160 and one or more individuals Y _i of the next generation are generated.

In a step S90, the parameters Ψ _i of the individuals Y _i of the new generation are varied by a defined amount in a mutation module 170 and a new individual Y′ _i of the new population Ω _t+1 is generated.

The population Ω _t+1 of the new generation is transmitted to the initialization module 120 in a step S100. In this case, the new generation Ω _t+1 is transferred to the current population Ω _t and then steps S10-S50 are carried out.

4 Figure 12 schematically illustrates a computer program product 700 comprising executable program code 750 configured to perform the method according to the first aspect of the present invention when executed.

With the method according to the present invention, time and costs for configuring suitable sensor sets 10 can be reduced and the safety of the moving object 20, in particular a motor vehicle, can be improved. The use of an evolutionary algorithm enables the parameters Ψ _i that have an influence on the data quality of a sensor set 10 to be adapted autonomously and automatically.

Reference List

10

sensor set

20

object

100

System for configuring a sensor set

110

requirements module

120

initialization module

130

simulation module

140

rating module

150

selection module

160

recombination module

170

mutation module

180

final module

700

computer program product

750

program code

a1, a2, ..., ak; b1, b2, ..., bl; ...; n1, n2, ..., nm

sensors

X1, X2, ..., Xi, ..., Xn

individuals of sensor sets

Y1, Y2, ..., Yi, ..., Yn

Individuals of next-generation sensor sets

Ωt

Current generation population

Ωt+1

next generation population

Ψ1, Ψ2, ..., Ψi, ..., Ψp

parameter string

F(Xi)

rating function

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• CN 109917290A [0007]
• US6006604A [0008]
• US7895021B1 [0009]

Claims (11)

A method for configuring an optimized sensor set (10) for a moving object (20), in particular a motor vehicle, comprising: - defining (S10) at least one requirement for a sensor set (10) with one or more sensors (a ₁ , a ₂ , ..., _ak ; b ₁ , b ₂ , ..., b _l ; ...; n ₁ , n ₂ , ..., n _m ) for fulfilling a task function of the moving object (20); - Initializing (S20) a population (Ω) of individuals (X ₁ , X ₂ , ..., X _n ) of sensor sets (10) selected randomly or according to certain criteria, wherein each individual (X _i ) has at least one parameter (Ψ _i ) is assigned; - Generating (S30) synthetic sensor data and reference information through the simulation for all individuals (X _i ) of the population (Ω); - Evaluation (S40) of the quality of the synthetic sensor data by means of an evaluation function (F(X _i )) for the synthetic sensor data and/or further data for one or more individuals (X _i ) of the population; - determining (S50) whether the algorithm can be terminated; - Terminating (S60) the algorithm if at least one defined termination criterion applies; - selecting (S70) one or more individuals (X _i ) from sensor sets (10) based on an evaluation result of the evaluation function (F(X _i )); - combining (S80) the parameters (Ψ _i ) of all selected individuals (X _i ) with one another and generating one or more individuals (Y _i ) of the next generation; - varying (S90) the parameters (Ψ _i ) of the individuals (Y _i ) of the new generation by a defined amount and generating a new individual (Y' _i ) of the new population (Ω _t+1 ); - Transmission (S100) of the new population (Ω _t+1 ) to the initialization module (120) and carrying out the method steps (S10-S50) with the new population (Ω _t+1 → Ω). procedure after claim 1 , where a parameter (Ψ _i ) is a specific property of a sensor element, such as the positioning, orientation in a vehicle or robot, and its resolution, sampling rate or monitoring range. Method according to one of the preceding claims, wherein the sensors (a ₁ , a ₂ , ..., _ak ; b ₁ , b ₂ , ..., b _l ; ...; n ₁ , n ₂ , ..., n _m ) are 2D/3D cameras, LiDAR sensors, radar sensors, ultrasonic sensors, force/torque sensors, acceleration sensors, pressure sensors, temperature sensors, humidity sensors or power management sensors. Method according to one of the preceding claims, in which one or more of the following conditions constitute a termination criterion: the evaluation result has reached a target value which is defined by the deviation being below a threshold value, and/or the individuals (X ₁ , X ₂ , ..., X _n ) of the population (Ω) are approaching and the population (Ω) is reaching a convergence value and/or a given maximum number of iterations has been reached. Method according to one of the preceding claims, in which the moving object (20) is an automobile, a watercraft, an aircraft or a robot. System (100) for configuring an optimized sensor set (10) for a moving object (20), in particular a motor vehicle, comprising a requirement module (110) which is designed to make at least one requirement for a sensor set (10) with one or more sensors (a ₁ , a ₂ , ..., a _k ; b ₁ , b ₂ , ..., b _l ; ...; n ₁ , n ₂ , ..., n _m ) to fulfill a task function of the itself to define moving object (20); an initialization module (120) that is designed to initialize a population (Ω) of individuals (X ₁ , X ₂ , ..., X _n ) of sensor sets (10) selected randomly or according to specific criteria, with each individual (X _i ) at least one parameter (Ψ _i ) is assigned; a simulation module (130) which is designed to generate synthetic sensor data and/or reference information for all individuals (X _i ) of the population (Ω) within a catalog of defined simulation scenarios; an evaluation module (140) which is designed to evaluate the quality of the individuals (X _i ) by means of an evaluation function (F(X _i )) for the synthetic sensor data for all individuals (X _i ) of the population; a termination module (180) configured to determine whether the algorithm can be terminated and to terminate the algorithm if at least one defined termination criterion applies; a selection module (150) which is designed to select one or more individuals (X _i ) from sensor sets (10) based on an evaluation result of the evaluation function (F(Xi)); a recombination module (160) configured to combine the parameters (Ψ _i ) of one or more of the selected individuals (X _i ) with one another and to generate one or more individuals (Y _i ) of the next generation; a mutation module (170) which is designed to vary the parameters (Ψ _i ) of the individuals (Y _i ) of the new generation by a defined amount and add a new individual (Y' _i ) to the new population (Ω _t+1 ). generate and transmit the new population (Ω _t+1 ) to the initialization module (120). system (100) after claim 6 , where a parameter (Ψ _i ) is a particular eigen shank of a sensor element such as the positioning, orientation in a vehicle or robot, as well as its resolution, sampling rate or monitoring area. System (100) according to one of Claims 6 - 8th , where the sensors (a ₁ , a ₂ , ..., a _k ; b ₁ , b ₂ , ..., b _l ; ...; n ₁ , n ₂ , ..., n _m ) are 2D/3D cameras, LiDAR sensors, radar sensors, ultrasonic sensors, force/torque sensors, acceleration sensors, pressure sensors, temperature sensors, humidity sensors or power management sensors. System (100) according to one of Claims 6 - 9 , where one or more of the following conditions constitute a termination criterion: the evaluation result has reached a target value, which is defined by the deviation being below a threshold value, and/or the individuals (X ₁ , X ₂ , ..., X _n ) of the population (Ω) are approaching and the population (Ω) is reaching a convergence value, and/or a given maximum number of iterations has been reached. System (100) according to one of Claims 6 - 10 , wherein the moving object (20) is a motor vehicle, a watercraft, an aircraft or a robot. Computer program product (700), comprising an executable program code (750), which is configured so that when it is executed it performs the method according to any one of Claims 1 until 5 executes

Images