Simulator and method for simulating an object acquisition sensor

Abstract

The test scenario tests the functionality of the control unit in a virtual environment using a simulation environment.SOLUTION: The simulator includes a visualization module, a mapping module, a combination module, and an output module. Wherein the visualization module generates object data in dependence on the sensor for a virtual object located in the virtual environment, and the mapping module determines at least two parameters of the virtual object from the object data, the virtual object is displayed in a first allocation map (24) as a function of the two parameters, the combination module transfers the first allocation map, taking into account a characteristic function (46) of the sensor, to a second allocation map (26), in which the virtual object is displayed as a function of the two parameters, and the output module converts the second allocation map into synthesized sensor data and provides the synthesized sensor data for output.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present application relates to a simulator and method and computer program product for simulating a sensor for capturing an object. [Background technology]

[0002] A device for performing open-loop and / or closed-loop control tasks in a vehicle is also called a control device. A control device in a vehicle, in particular an automobile, can have a computing unit, a memory, an interface, and possibly further components necessary for processing input signals containing input data for the control device and for generating control signals containing output data. The interface is used to receive the input signals or to output the control signals.

[0003] Whether for Advanced Driver Assistance Systems (ADAS) or autonomous or semi-autonomous driving, the control equipment for driving functions can receive sensor data as input from various sensors, such as object acquisition sensors.

[0004] One way to test a control device that evaluates sensor data from a sensor is to test the control device with the corresponding sensor in an installed state, for example, during a test drive in a vehicle. This is time-consuming and cost-intensive, and many situations occur only in extreme cases, such as an accident, and therefore cannot be tested in a real environment. Therefore, the corresponding control device is tested in an artificial environment, for example, during a bench test. A further frequent test scenario here is to test the functionality of the control device in a so-called virtual environment using a simulated environment. This environment simulation can include the simulation of real sensors using a sensor simulation. The sensor simulation can generate synthetic sensor data that simulates the virtual environment of the control device.

[0005] From the paper "Ray Tracing for Range-Doppler Simulation of 77 GHz Automotive Scenarios, 13th European Conference on Antennas and Propagation (2019)" by Stefan O. Wald, Frank Weinmann et al., it is known that so-called ray tracing is used to determine range-Doppler maps. Here, in a virtual surrounding environment, propagation paths are analytically calculated using the positions of the transmitter, scattering centers, and receivers. This can then be reused to determine synthetic range-Doppler maps. Summary of the Invention [Means for solving the problem]

[0006] A simulator for simulating an object capture sensor is configured to generate synthesized sensor data of the object capture sensor using the simulation, the simulator including a visualization module, a mapping module, a combination module, and an output module.

[0007] The visualization module is configured to generate sensor-dependent object data for at least one virtual object present within the virtual surroundings.

[0008] The mapping module is configured to determine at least two parameters of the virtual object from the object data and to display the at least one virtual object in the first allocation map depending on the at least two parameters.

[0009] The combination module is configured to transform the first assignment map into a second assignment map taking into account a characteristic function of the sensor, wherein in the second assignment map at least one virtual object is displayed depending on at least two parameters.

[0010] An output module is configured to convert the second assignment map into combined sensor data and provide the combined sensor data for output.

[0011] A method for simulating a sensor for object acquisition generates synthesized sensor data for the sensor, the method comprising the steps of: generating object data for at least one virtual object present in the virtual environment, the generation of the object data being dependent on an object acquisition sensor; - determining at least two parameters of the virtual object from the object data and displaying at least one virtual object in a first allocation map depending on the at least two parameters; - a step of transferring the first allocation map into a second allocation map taking into account a characteristic function of the sensor, in which at least one virtual object is displayed in the second allocation map depending on at least two parameters; converting the second assignment map into a combined sensor data and providing the combined sensor data for output; Includes.

[0012] The computer program product contains instructions that, when the program is executed by a computer, cause the computer to perform the steps of the described method.

[0013] By using a describing simulator and a describing device, the synthesized sensor data that is generated can be better adapted to the requirements placed on the synthesized sensor data.

[0014] The simulator for simulating a sensor for capturing an object and the method for simulating a sensor for capturing an object can calculate partially processed sensor data in real time. The synthesized sensor data provided for output can represent synthesized sensor data that simulates data that exists as partially processed sensor data within the sensor. This allows such partially processed sensor data to be provided as part of the synthesized sensor data for further processing within a simulation that uses the synthesized sensor data.

[0015] The simulator and the method can dispense with the need to simulate specific technical details of the sensor. It is only necessary to provide a characteristic function that characterizes the sensor. However, this can be measured, for example, and the respective sensor manufacturer does not need to reveal their know-how regarding the specific technical details of the sensor's structure.

[0016] The simulator for simulating an object acquisition sensor can avoid the use of a real object acquisition sensor, which can reduce testing efforts. The object acquisition sensor can include, for example, a radar sensor, a lidar sensor, a camera, and / or an ultrasonic sensor. The simulator simulates such sensors as intended.

[0017] The simulator generates synthesized sensor data based on a virtual environment with at least one virtual object, which can then be used for further processing. This eliminates the need for an inspection structure with sensors and an object simulator; instead, the sensors and the synthesized sensor data generated by the sensors can be simulated on one or more computers based on a virtual environment with virtual objects. Therefore, the simulator comprises one or more computers and / or processors with corresponding memories, input interfaces, and output interfaces.

[0018] The synthesized sensor data generated by the simulation of the sensor should ideally be indistinguishable from sensor data generated by a sensor, such as a radar sensor, etc. The synthesized sensor data can be further processed, for example, by a control device and / or a simulation of the control device.

[0019] The visualization module generates object data for a virtual environment having at least one virtual object that would have been generated by sensors in a real environment having real objects. The visualization module may therefore be configured as a software function. The visualization module may also run on one or more processors, such as graphics processors, assigned to the software function.

[0020] The mapping module may also be configured as a software module that generates a first assignment map having at least two parameters of the virtual object from the object data. When two parameters are used, the first assignment map can be understood as a two-dimensional diagram that maps the two parameters to function values. When more than two parameters are used, a correspondingly higher-dimensional first assignment map exists.

[0021] The first assignment map is a representation of how it was determined from the object data by the first assignment map. The sensor characteristics are still missing from the first assignment map. These are inserted by a combination module, which may also be configured as a software function. This combination module uses the first assignment map and the sensor characteristic functions to determine the second assignment map.

[0022] The sensor characteristic function reflects the sensor's characteristics. That is, it can be determined from the sensor's output data based on the supplied test data. Therefore, the second assignment map provides a more realistic representation of at least two parameters of at least one virtual object. This allows for, for example, consideration of edge effects, mapping artifacts, or other effects caused by the sensor's realistic characteristics.

[0023] The output module may be a software function that transfers the second assignment map to the combined sensor data and provides it for output, which can then be used to perform further processing steps, such as testing a control device.

[0024] The same applies correspondingly to the method and computer program product.

[0025] In one embodiment, the first assignment map can include a first range-Doppler map, and the second assignment map can include a second range-Doppler map. The range-Doppler map plots distance from an object against its relative velocity to the sensor on a two-dimensional graph. That is, the two parameters are distance and relative velocity. That is, the first range-Doppler map is transformed into the second range-Doppler map by the combination module.

[0026] Measuring the Doppler shift provides the possibility for direct velocity measurement of a sensor signal, such as a radar or lidar echo, reflected by an object. For example, if an object moves toward the sensor, the frequency of the object's echo signal will be slightly higher than the transmitted signal. Conversely, the lowest frequency indicates that the object is moving away from the sensor. The Doppler shift increases the faster the object moves relative to the sensor, allowing for direct inference of the object's relative radial velocity.

[0027] One example of a sensor for target acquisition is a frequency-modulated continuous wave (FMCW) radar. Such an FMCW radar transmits a continuous radar signal, the probe signal. This signal is an uninterrupted, periodic, and consecutive signal segments with a varying frequency (chirp). The signal reflected by the target is a correspondingly time-shifted chirp. Due to the time offset and the varying frequency, a small frequency difference occurs between the probe signal and the echo signal. In the radar's mixer, the transmitted signal, the probe signal, is superimposed with the echo signal. This results in a beat with a beat frequency for each echo signal. The farther the target is from the sensor, the higher the beat frequency. From this beat, the sensor generates a Fourier transform. In the Fourier transform, each echo signal, i.e., each reflected target, is visible as a peak corresponding to the respective beat frequency. The distance of the target can be directly inferred from its position on the frequency axis.

[0028] A Fourier transform is determined for each individual chirp by the sensor. The individual results of these Fourier transforms are combined. This results in a two-dimensional distance-time map with distance on the vertical axis and time on the horizontal axis. On the vertical axis (distance), an individual Fourier transform is plotted. Then, along the horizontal axis, multiple Fourier transforms are plotted in time order.

[0029] The sensor then performs a second Fourier transform, the range-time map, along the time axis. The result is the so-called range-Doppler map. In each row of the range-time diagram, the phase of the values ​​changes along the time axis. The horizontal position in the range-Doppler map indicates the frequency of the object's phase path, i.e., the phase change over time. This phase change is the frequency shift of the radar echo due to the object's relative movement. Each peak in the range-Doppler map therefore corresponds to an object detected by the radar. The vertical position of the peak allows the object's distance to be read, while the horizontal position allows the relative radial velocity to be read. Such a range-Doppler map can also be used, for example, in the case of a lidar sensor, provided that the lidar sensor also operates according to the FMCW principle.

[0030] In particular, the characteristic function can comprise a point spread function of the sensor, and the combination module can be configured to transform the first assignment map into a second assignment map under a convolution function with the point spread function. In one embodiment, the characteristic function is a point spread function of the sensor, and the first range-Doppler map is transformed into the second range-Doppler map by convolution with the point spread function.

[0031] The point spread function of a sensor describes the effect of band-limiting influence factors such as diffraction phenomena at the iris, mapping errors or the influence of the sensor surface or aperture, for example in high frequency technology, when the sensor is a radar or lidar in optical systems, and also in image processing.

[0032] The point spread function describes how an idealized point-like object would be mapped by the system (i.e., the sensor). In many cases, the shape of the response does not depend on the original location of the ideal point-like object. In this case, referred to as a linear system, the overall system response can be calculated as a sum over the point responses of the object divided into its points. Convolution causes each point-like maximum in the first mapping map to be replaced by a function progression corresponding to the point spread function.

[0033] In one embodiment, the sensor for object acquisition includes an active sensor configured to generate raw sensor data by transmitting probe signals and receiving echo signals, and to determine the second assignment map from the raw sensor data. An active sensor is a sensor that transmits a signal, also called a probe signal, to acquire objects in its surrounding environment, which is reflected from objects present in the surrounding environment via echo signals. Through the echo signals, the objects can then be detected and their relative motion with respect to the sensor can be observed over time. Thus, active sensors include radar, lidar, or acoustic sensors that output probe signals and then evaluate the corresponding echo signals.

[0034] In one embodiment, an inverse calculation module, which may be configured as a software function, is provided, where the inverse calculation module determines synthesized raw sensor data from the second assignment map, where determining the synthesized raw sensor data depends on the target acquisition sensor, and access to the synthesized raw sensor data may make sense in the future for further processing of the synthesized sensor data during sensor simulation.

[0035] That is, by using the simulator or the method, the second assignment map can be determined from the object data. That is, the synthesized sensor data having the object data and the second assignment map can be provided for output. Determining the synthesized raw sensor data is optional, but not required. Eliminating the calculation of the synthesized raw sensor data allows for efficient and fast calculations and cost-effective implementation.

[0036] In one embodiment, determining the combined raw sensor data by the inverse calculation module comprises an inverse Fourier transform, from which the second allocation map is determined by the Fourier transform, and therefore, an inverse Fourier transform can be used for the inverse calculation.

[0037] In one embodiment, simulated sensors within the virtual surroundings are assigned virtual positions and object data for at least one virtual object is periodically generated, for example by a visualization module, taking into account the respective virtual positions.

[0038] For this purpose, for example, so-called ray tracing can be used, in which the transmission of a virtual probe signal in a virtual surrounding environment and the transmission of a respective virtual echo signal on a virtual object are calculated, and the object data for each virtual object can comprise a representation of the respective virtual echo signal of the respective virtual probe signal on the respective virtual object.

[0039] The synthesized sensor data is configured in embodiments of the simulator and method such that object data is derivable from sensor-synthesized sensor data, i.e., the synthesized sensor data is of such quality that it could be real sensor data and may have been processed by a real sensor.

[0040] The drawings show embodiments of the invention, which are explained in more detail in the following description. [Brief explanation of the drawings]

[0041] [Figure 1] FIG. 1 is a block diagram showing a simulation environment. [Figure 2] FIG. 10 illustrates the conversion of sensor-object data into a second assignment map. [Figure 3] FIG. 1 illustrates a virtual surroundings with virtual objects. [Figure 4] FIG. 10 shows a range-Doppler map from object data. [Figure 5] FIG. 10 is a diagram showing a characteristic function. [Figure 6] FIG. 10 is a diagram illustrating a second allocation map. [Figure 7] FIG. 10 illustrates the generation of a second allocation map. [Figure 8] 1 is a flow chart illustrating the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0042] In the drawings, the same reference numerals are used for the same or similar elements. The depictions in the drawings may not be to scale.

[0043] 1 shows a simulator 10, a virtual environment 30, and a simulated environment 40 having an application 36. The simulator 10 is configured to simulate an object acquisition sensor, such as a radar or lidar sensor.

[0044] The application 36 may be, for example, an automated control unit of a vehicle in which the simulated sensor 38 is to be located. However, other driver assistance functions may also be such applications 36 or apps. The application 36 may also be at least partially configured as hardware, for example, as a control device or at least part of a control device.

[0045] This structure of the simulated environment 40 according to Fig. 1 may be purely virtual and may be executed, for example, on one or more computers, making it possible to simulate signal processing and / or test applications 36, such as driving functions, without time-consuming and costly test structures.

[0046] The structure of the simulated environment 40 according to FIG. 1 may be implemented partly in hardware and partly in software.

[0047] In the virtual surroundings 30, a desired scenario can be displayed using one or more virtual objects O1-O6. The movement of these virtual objects O1-O6 relative to a simulated sensor 38 can also be realistically simulated. In particular, an application 36 can interact with the virtual surroundings 30 via, for example, application data and synthesized sensor data 32.

[0048] The simulator 10 includes a visualization module 12 that generates object data 22 for at least one virtual object O1-O6 in a virtual surrounding environment 30, relying on sensors. The object data 22 includes information about the presence, distance, and orientation of the virtual objects O1-O6. The object data 22 can be obtained, for example, by ray tracing.

[0049] The object data 22 are output by the visualization module 12 to the mapping module 14 on the one hand and to the output module 20 on the other hand. These modules 12, 14, and 20 are preferably implemented as software functions. In particular, in the case of computationally intensive functions, such as the calculation of a Fourier transform, it is possible that dedicated hardware may be provided for the execution of these functions. For this purpose, for example, a graphics processor may be used.

[0050] The mapping module 14 determines at least two parameters of the virtual objects O1-O6 from the object data 22 and displays the at least two parameters in a first allocation map 24. The first allocation map 24 is transmitted to the combining module 16.

[0051] The combination module 16 transforms the first assignment map 24, taking into account the characteristic function 46 of the sensor, into a second assignment map 26, in which at least one virtual object O1-O6 is displayed in dependence on at least two parameters. The second assignment map 26 is then output by the combination module 16 on the one hand to an inverse calculation module 18 and on the other hand to an output module 20.

[0052] The inverse calculation module 18 determines synthesized raw sensor data 28 from the second assignment map 26, which the inverse calculation module 18 forwards to the output module 20. To determine the raw sensor data 28 from the second assignment map 26, the inverse calculation module 18 may use an inverse Fourier transform.

[0053] The output module 20 provides synthesized sensor data 32 for output, for example, to an application 36. These synthesized sensor data 32 are generated from the second assignment map 26 and here contain, among other things, information from the second assignment map 26. Here, the object data 22 and / or the synthesized raw sensor data 28 may also be taken into account. The synthesized sensor data 32 are generated by the output module 20 in particular to simulate as accurately as possible the sensor data of the real sensor being simulated.

[0054] An application 36 that relies on the synthesized sensor data 32 to determine, for example, a trajectory for a vehicle, can transmit such application data 34 to the virtual surroundings 30, thereby taking into account the relative motion of the simulated sensors 38 over time with respect to virtual objects O1-O6 in the virtual surroundings 30.

[0055] The second allocation map is here configured such that its information corresponds to the processing steps within the real sensor whose information is simulated by the virtual sensor 38. The characteristic functions 46 described above, in particular the point spread function, make it possible to carry out a realistic simulation of the signal processing chain and also to output the intermediate product of the second allocation map 26 by the simulator 10. For example, the point spread function allows each sensor to characterize its own mapping properties, so that this point spread function can be used to back-calculate from the object data the range-Doppler map as would exist in the sensor to be simulated.

[0056] 2 shows how various virtual objects O1-O6 simulated in a virtual surroundings 30 are reflected in the second allocation map 26. As shown in the left image of Fig. 2, a simulated sensor 38 transmits a probe signal to the virtual surroundings 30 containing the objects O1-O6. For this purpose, so-called ray tracing can preferably be used.

[0057] Driver assistance systems that can actively observe the vehicle's environment and control the vehicle's operation based on the observations are regularly developed and tested in virtual environments, where the respective sensors, e.g., radar sensors, are completely or partially replaced by virtual sensors 38, which may also be referred to as free cuts.

[0058] The virtual sensors 38 emulate the functionality of real physical sensors in a realistic manner and generate synthesized sensor data 32 that is adapted to match the virtual test environment 30. As described with reference to Figure 1, the synthesized sensor data 32 may be generated from the second assignment map 26. Optionally, the synthesized sensor data 32 may additionally be generated from the object data 22.

[0059] The synthesized sensor data 32 is provided to an application 36, such as an application for emergency braking assistance, pedestrian recognition and / or fully autonomous driving. The synthesized sensor data 32 can be used to test whether the application 36 reacts as expected to situations simulated in the virtual test environment 30.

[0060] The virtual test environment 30 should realistically simulate a real-world test of the application 36. This means that the entire simulation, including the generation of the synthesized sensor data 32, must be performed in real time. For example, if the application 36 expects new sensor data every 30 ms, the simulation must provide a complete, current set of synthesized sensor data 32 every 30 ms.

[0061] In Fig. 2, virtual objects O1-O6 are visible in the virtual surroundings 30, from which a simulated probe signal 42 emitted by a simulated sensor 38 is reflected (see Fig. 3). The corresponding wavefront of the simulated echo signal 44 (see Fig. 3) is received by the simulated sensor 38 and determines the generation of object data 22, from which the second allocation map 26 is then determined, as described with respect to Fig. 1.

[0062] A second allocation map 26 is depicted on the right side of Figure 2. The positions of objects O1-O6 in the second allocation map 26 are marked by arrows between the image on the left and the second allocation map 26.

[0063] The horizontal axis of this second assignment map 26 plots the relative velocity between the simulated sensor 38 and each of the virtual objects O1-O6, and the vertical axis plots the distance between the simulated sensor 38 and each of the objects O1-O6. The brightness of the second assignment map 26 can represent a third dimension, specifically, how strong the echo signal of each object is. This is also called a peak. This can indicate the size of each of the objects O1-O6, but at least how strongly each of the objects O1-O6 reflects the signal transmitted from the sensor 38. To the right of the second assignment map 26, the assignment of brightness to a numerical value is shown.

[0064] FIG. 3 shows an exemplary ray tracing method used to generate object data 22 within a virtual environment.

[0065] Ray tracing is a known method for realistically representing optical phenomena in computer graphics. To calculate the image perceived by a virtual observer, a bundle of geometric rays is projected starting from the observer's eye into the virtual surroundings 30. By further tracing the individual rays based on the laws of reflection or refraction, it is possible to calculate what the observer sees at reflective and / or refractive surfaces. The use of graphics maps allows the calculation of ray tracing images in real time.

[0066] In this example, the observer is a virtual sensor 38, and ray tracing is applied to radar waves, i.e., electromagnetic waves in the radio frequency range. Polygons, i.e., objects O1-O6, visible to the virtual sensor 38 are illuminated within the virtual test environment 30 by virtual geometric beams as simulated probe signals 42. Each beam is reflected and traced as a simulated echo signal 44 according to the laws of geometric optics. Based on the characteristics of electromagnetic waves in the radar range, this simulation is performed so that each contact of the radar beam as the simulated probe signal 42 with a polygon generates a local wavefront as the simulated echo signal 44. Each wavefront is simulated as a plane wave directed toward the virtual sensor 38. The strength of this plane wave depends on the spatial orientation distance and the material of each polygon, i.e., object O1-O6. The echo measured by the virtual sensor 38 is calculated as the sum of all plane waves in the simulated echo signal 44.

[0067] 3 shows a schematic representation of a simulated sensor 38 with a simulated probe signal 42 and a simulated echo signal 44 as a wavefront, where in this example only fictitious objects O1 and O3 and a sidewall are shown, which may be one of the objects O1-O6.

[0068] The probe signal emitted from the center of the sensor 38 does not cause any reflections, but the probe signal emitted on the objects O1, O3 and on the right side wall causes echo signals 44 as plane wavefronts (far-field approximation). In the sensor 38, the echo signals 44 are superimposed, and then the simulated sensor 38 generates the object data 22 from these echo signals 44 using the visualization module 12. In this way, the laws of optical reflection and refraction are used to determine the course of the probe signal 42 and the echo signals 44.

[0069] 4 shows the first assignment map 24 configured as a first range-Doppler map 24, which shows Doppler velocity in meters per second on the horizontal axis and the distance in meters between the sensor 38 and each object O1-O6. The third dimension is the intensity of the objects O1-O6 shown, and in this case the relationship between this intensity and a numerical value is shown further to the right.

[0070] The first assignment map 24 contains information determined by the mapping module 14 from the object data 22. Only the local maximum values ​​are plotted from this first range-Doppler map. This first range-Doppler map can be quickly generated because all the information needed for this purpose is readily available and retrievable within the object data 22.

[0071] FIG. 5 shows a characteristic function 46 for a sensor, configured as a point spread function. A point spread function is the sensor's response to ideal point-like objects. Such a characteristic function 46 can be measured and / or analytically described for sensors such as radar sensors or lidar sensors. The point spread function here describes how each point-like object is mapped by the sensor.

[0072] The second range-Doppler map 26 of Figure 6 is determined by the combination module 16 by convolving the characteristic function 46 of Figure 5 with the first range-Doppler map 24 of Figure 4. Smearing is then added to the first range-Doppler map 24 (Figure 4), thereby generating the second range-Doppler map 26 (Figure 6).

[0073] From the first range-Doppler map 24, a second assignment map 26 is created by convolution with a characteristic function 46, and this second assignment map 26 is realistic because it takes into account the mapping characteristics of the sensor to be simulated.

[0074] FIG. 6 shows the second assignment map 26 with the desired smearing of the plotted signals. The right side again shows the conversion of the plotted signals' brightness, peaks, to numerical values. The recognized objects O1-O6 can be recognized as brightly illuminated peaks. Each object O1-O6 is flanked by cross-shaped smears along both axes. These smears do not represent the physical reality of the sensor's environment, but are artifacts of the sensor 38 due to the time-limited analysis interval of the raw data when performing a Fourier transform.

[0075] 7 shows a schematic diagram of how the convolution is performed, where the first allocation map 24 is convolved with the characteristic function 46 to produce the second allocation map 26 with the corresponding surface. Smearing around the two dark fields is clearly visible, but this is due to the characteristics of the sensor 38.

[0076] FIG. 8 shows a flow chart of the method according to the present application.

[0077] In a method step 80, object data 22 for at least one virtual object O1-O6 present in the virtual surroundings 30 is generated as a function of the sensor 38. For this purpose, for example, the ray tracing method described with reference to Figures 2 and 3 can be used. This method step 80 is, for example, executed by the visualization module 12.

[0078] In method step 82, at least two parameters of at least one virtual object O1-O6 are determined, in particular from the object data 22. The virtual object O1-O6 is then displayed in the first allocation map 24 depending on the at least two parameters. Method step 82 is executed, for example, by the mapping module 14.

[0079] In method step 84, the first assignment map 24 is transferred to a second assignment map 26 taking into account the characteristic function 46 of the sensor 38, in which at least one virtual object O1-O6 is displayed in the second assignment map 26 depending on at least two parameters. Method step 84 is executed, for example, by the combination module 16.

[0080] A method step 86 involves converting the second assignment map 26 into combined sensor data 32 and providing the combined sensor data 32 for output. This method step 86 is performed, for example, by the output module 20. [Explanation of symbols]

[0081] 10 Simulator 12 Visualization Module 14 Mapping Module 16 Combined Modules 18 Inverse calculation module 20 Output Module 22 Object Data 24 First Allocation Map 26 Second Allocation Map 28 Combined raw sensor data 30 Virtual Surroundings 32 Synthesized Sensor Data 34 Application Data 36 Applications 38 Simulated Sensor 40 Simulated Surrounding Environment 42 Simulated search signal 44 Simulated echo signal as a wavefront 46 Characteristic Functions 80-86 Method Steps O1-O6 Virtual Objects

Claims

1. A simulator (10) for simulating a sensor for capturing an object, comprising: The sensor comprises an active sensor configured to generate raw sensor data by transmitting a probe signal and receiving an echo signal; the simulator (10) is configured to generate synthesized sensor data (32) for the sensor using a simulation; The simulator (10) comprises a visualization module (12), a mapping module (14), a combination module (16), an inverse calculation module (18), and an output module (20); the visualization module (12) is configured to generate sensor-dependent object data (22) for at least one virtual object (O1-O6) present in a virtual surrounding environment (30); the mapping module (14) is configured to determine at least two parameters of the virtual object (O1-O6) from the object data (22) and to display the at least one virtual object (O1-O6) in a first allocation map (24) depending on the at least two parameters; the combining module (16) is configured to convert the first assignment map (24) into a second assignment map (26) taking into account a characteristic function (46) of the sensor, in which the at least one virtual object (O1 to O6) is displayed depending on at least two parameters, and the object acquisition sensor is configured to determine the second assignment map (26) from raw sensor data; the inverse calculation module (18) is configured to determine synthesized raw sensor data (28) from the second assignment map (26), and determining the synthesized raw sensor data (28) is dependent on the object capture sensor and includes an inverse Fourier transform; the output module (20) is configured to provide the combined raw sensor data for output; Simulator.

2. The first allocation map (24) comprises a first range-Doppler map; The second allocation map (26) comprises a second range-Doppler map. The simulator of claim 1.

3. the characteristic function (46) comprises a point spread function of the sensor, and the combination module (16) is configured to transfer the first assignment map (24) to the second assignment map (26) under execution of a convolution function with the point spread function (46), the convolution function in particular comprising a convolution; The simulator according to claim 1 or 2.

4. a simulated sensor (38) in the virtual surroundings (30) is assigned a virtual position, and the visualization module (12) is configured to periodically generate object data (22) for the at least one virtual object (O1-O6) taking into account the respective virtual position; The simulator according to any one of claims 1 to 3.

5. The object data (22) for each virtual object (O1-O6) comprises a representation of a respective virtual echo signal (44) of a respective virtual probe signal (42) in said each virtual object (O1-O6). The simulator according to claim 4.

6. the combined sensor data (32) is configured such that the object data (22) can be derived therefrom by the sensors; A simulator according to any one of claims 1 to 5.

7. 1. A method for simulating a sensor for capturing an object, comprising: The sensor comprises an active sensor configured to generate raw sensor data by transmitting a probe signal and receiving an echo signal; the simulation generates synthesized sensor data (32) for the sensors; The method comprises the following method steps: generating object data (22) for at least one virtual object (O1-O6) present in a virtual surrounding environment (30) in dependence on said sensors; - determining at least two parameters of the virtual objects (O1 to O6) from the object data (22) and displaying at least one of the virtual objects (O1 to O6) in a first allocation map (24) depending on the at least two parameters; - transforming said first allocation map (24) into a second allocation map (26) taking into account a characteristic function (46) of said sensor, in which said at least one virtual object (O1 to O6) is displayed in dependence on at least two parameters; - determining a combined raw sensor data (28) from the second allocation map, the combined raw sensor data (28) being dependent on the object acquisition sensor and comprising an inverse Fourier transform; - providing said combined raw sensor data (32) for output; A method comprising:

8. The first allocation map (24) comprises a first range-Doppler map; The second allocation map (26) comprises a second range-Doppler map. The method of claim 7.

9. the characteristic function (46) comprises a point spread function of the sensor, and the first allocation map (24) is transferred to the second allocation map (26) under the execution of a convolution function with the point spread function, the convolution function in particular comprising a convolution; 9. The method according to claim 7 or 8.

10. The object capturing sensor includes an active sensor configured to generate sensor raw data by transmitting a probe signal and receiving an echo signal, and to determine the second assignment map from the sensor raw data.

10. The method according to any one of claims 7 to 9.

11. a simulated sensor (38) in the virtual environment (30) is assigned a virtual position, and object data (22) for the at least one virtual object (O1-O6) is periodically generated taking into account the respective virtual position; 11. The method according to any one of claims 7 to 10.

12. 1. A computer program product comprising: The computer program product comprises instructions which, when executed by a computer, cause the computer to carry out the steps of the method according to any one of claims 7 to 11. Computer program products.

Images