JPWO2021004626A5 - - Google Patents

Description

The present invention relates to a method for simulating sensor data of a continuous wave (CW) light detection and ranging (LiDAR) sensor (CW lidar sensor). The invention further relates to an apparatus for simulating sensor data of a CW lidar sensor.

Lidar is a laser-based technique for measuring the distance to a target, which can provide a high-resolution, three-dimensional representation of its surroundings. Lidar System Services applications also provide so-called environment models that are used in automatic driver assistance systems (ADAS) for parking assistance, collision warning and autonomous driving applications. Lidar systems are widely used in a wide variety of fields such as geodesy, seismology, airborne laser swath mapping (ALSM), and laser altimeters.

There are various types of lidar sensors. For example, a pulsed lidar sensor emits a short laser pulse and measures the time of flight from the emission of the laser pulse to its reflection off the target and back to the lidar sensor. Using this measured time-of-flight, the distance to the reflective target can be calculated.

Another type of lidar sensor uses continuous wave (CW) signals. In this case the sensor emits light continuously. and modulating the light source either in amplitude with an amplitude modulated continuous wave (AMCW) or in frequency with a frequency modulated continuous wave (FMCW). can be modulated. For many applications, the CW method is preferred over the pulse lidar sensor method due to its accuracy and robustness.

DE 10 2016 100416 A1 EP-A-3260875 U.S. Patent Application Publication No. 2016/005209

A. Kim et al., "Simulating full-waveform lidar", Proc. SPIE 7684, Laser Radar Technology and Applications XV, 2010 A. Kim, "Simulating full-waveform LIDAR", Diss. Monterey, California. Naval Postgraduate School, 2009

Sensor data obtained by lidar systems require high accuracy for many applications. As such, new development of lidar applications may require extensive testing of lidar sensors, including physical sensors. Therefore, in order to reduce manufacturing costs for this purpose, simulation data that can reproduce the output of the lidar sensor under development as accurately as possible is required. Lidar simulation data are also needed in a variety of other areas, including signal processing algorithm development, neural network training, virtual verification, and so on.

Providing lidar sensor data that models pulsed lidar sensors is discussed in Non-Patent Documents 1 and 2.

Also, US Pat. No. 6,200,000 deals with testing virtual sensors in a virtual environment. This can include various sensor types, including lidar.

Patent Document 2 discloses a vehicle test method. The method includes acquiring radar sensor data responsive to radar excitation signals.

US Pat. No. 6,200,000 deals with photorealistic imaging and image generation, with a focus on lighting design.

A known method of providing lidar sensor data is to simulate an ideal point cloud using a single simulated ray. Ray propagation is calculated to the nearest surface intersection using ray tracing methods, and only specular reflections from that surface are considered. Therefore, only one distance is calculated per ray. In addition, beams containing multiple rays can be used to provide multiple points per beam. Such simulations provide only point data as output and do not produce information about the resulting signal. In contrast, real lidar sensors generate time-varying electrical signals that must be processed to generate points in the simulated scene. The time-varying signal contains a lot of additional information about the simulated scene, including object shape, object class, multiple reflections, and so on. Such information is inevitably lost if only the point cloud is provided as output.

In addition, the time-varying signal contains features resulting from the physical properties of the lidar sensor and the environment, such as beam shape, movement of objects between beams, and the effects of weather on the signal. These features introduce artifacts into lidar data and must be taken into account when developing algorithms based on sensors and their outputs. Therefore, point clouds may not be sufficient to simulate lidar sensors under development, and more realistic data may be needed.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide a method and apparatus for providing realistic continuous wave lidar data.

This problem is solved by the subject matter of the independent claims. Advantageous embodiments are described in the dependent claims.

According to a first aspect, the present invention provides a method of simulating (modeling) sensor data of a continuous wave (CW) light detection and ranging (LIDAR) sensor according to claim 1 . According to a second aspect, the invention provides a device for simulating sensor data of a CW lidar sensor as claimed in claim 10 . According to a third aspect, the invention provides a computer program product as claimed in claim 14. According to a fourth aspect, the invention provides a non-transitory computer-readable storage medium as claimed in claim 15.

According to a first aspect, the present invention provides a method for simulating sensor data of a continuous wave (CW) light detection and ranging (LIDAR) sensor, wherein based on the CW signal, at least A ray set containing one ray is generated. Each ray in the ray set has an emission start time and an emission duration. For each ray in the ray set, the ray is propagated through a simulated scene containing at least one object. For each ray in the ray set, the signal contribution of the propagated ray is calculated at the detection location in the simulated scene. An output signal is generated based on mixing the CW signal with the calculated signal contributions of the rays in the ray set. The method further provides for storing and/or outputting the output signal.

According to a second aspect, the invention provides an apparatus for simulating sensor data of a CW lidar sensor. The apparatus comprises a processing portion and at least one of a storage portion and an output portion. A processor generates a ray set including at least one ray based on the CW signal. Each ray of the ray set has an emission start time and an emission duration. A processor propagates the ray through a simulated scene containing at least one object for each ray in the ray set. The processing unit computes, for each ray in the ray set, the signal contribution of the propagated ray at the detected location in the simulated scene. The processor further produces an output signal based on mixing the CW signal with the calculated signal contribution of the rays in the ray set. The storage section stores the output signal, and the output section outputs the output signal.

According to a third aspect, the invention provides a computer program product. The computer program product includes executable program code configured to perform a method for simulating sensor data of a continuous wave sensor when executed by a computer.

According to a fourth aspect, the invention provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium contains executable program code configured to perform a method for simulating sensor data of a CW lidar sensor when executed by a computer.

The present invention provides simulated data corresponding to the full waveform signal of a CW lidar sensor. Here the entire signal information is preserved rather than providing only a “complete” point cloud type representation of the simulated scene. As such, providing sensor data, including output signals, also allows additional information to be determined, such as object shape and object class, as well as object orientation. Furthermore, sensor data also contains artifacts present in actual lidar data. Such artifacts, such as multiple reflections of the lidar beam, come from the physical properties of the lidar sensor and the environment and are not present in point cloud type data. Therefore, sensor development can be improved by also taking such artifacts into account.

The present invention accurately reproduces the output of a real CW lidar sensor because it accurately simulates how a real CW lidar sensor works. In addition, the high-quality data provided is complete and accurate for the real world, and can also be of high value in the development of signal processing algorithms, neural network learning, virtual verification, and so on.

In an embodiment method, the ray set contains only one ray. However, it is generally preferred that a ray set includes multiple rays. For example, a ray set can include at least 2, preferably at least 100, more preferably at least 500, and most preferably at least 1000 rays. The upper number of rays may also be restricted. For example, a ray set may include at most 100,000 rays, preferably at most 10,000 rays, more preferably at most 5000 rays, and most preferably at most 2000 rays.

According to the preferred embodiment method, each ray in the ray set includes a spatial origin in the simulated scene and a radial direction in the simulated scene. Rays are thus sampled in both space and time. Sampling in space involves determining the spatial origin and direction of emission, and sampling in time involves assigning portions of the CW signal to rays. A radiation direction corresponds to a vector in the simulated scene and can indicate in which direction the rays are emitted.

According to the method of the preferred embodiment, the step of propagating the ray is based on the direction of radiation of the ray, based on the reflection of the ray on objects in the simulated scene, and further on the spatial distribution of the ray. Including determining the optical path of the ray based on the origin. In addition, the throughput of rays along the optical path to the detection location is calculated. A signal contribution calculation step is performed based on the calculated throughput. According to one embodiment, only single reflections can be considered. However, in general, multiple reflections of rays from objects can also be considered. The number of reflections may be limited, ie only reflections below a predetermined threshold may be considered.

According to the preferred embodiment method, for each ray in the ray set, the mixing of the CW signal with the ray's calculated signal contribution is based on the signal offset between the signal contribution and the CW signal. . A signal offset is determined based on the radiation start time of the beam and the optical path of the beam.

According to the method of the preferred embodiment, the radiation direction is randomly chosen for each ray. Also, the direction of light emission may be uniformly selected. In general, the direction of emission is chosen to allow for the finite extent of the laser beam emitted by the lidar sensor. In general, a laser beam can be described by a Gaussian beam with waist w_0. However, the laser beam can also have any shape. In this case, the radial direction of the entire beam can be described by a solid angle. Additionally, the radiation direction of the lidar sensor may be adjustable. For example, a lidar sensor may comprise micromirrors adapted to deflect an emitted laser, thereby scanning a region of space.

According to the method of the preferred embodiment, the ray set comprises a plurality of rays. The radiation start time of the light beam is randomly selected. The light emission time may be chosen uniformly. In general, the ray emission time is chosen such that all points in time are sufficiently sampled to produce an accurate output signal. The contributions of different rays may be weighted according to an importance sampling method. For example, the signal contributions of rays may be weighted with different weights.

According to a preferred embodiment of the method, the CW signal is a frequency modulated continuous wave (FMCW) signal. The CW signal may also be an amplitude modulated continuous wave (AMCW) signal.

According to a preferred embodiment of the method, generating the ray set includes, for each ray, assigning a portion of the CW signal to the ray. A ray corresponds to a portion of the infinite-time CW signal that begins at the emission start time and has a length equal to the emission duration. In other words, the CW signal is an infinite time signal, ie the laser emits continuously, each beam corresponding to a section of the CW signal at some time interval. The time dependence of the beam amplitude is equal to the time dependence of the corresponding section of the CW signal.

According to a preferred embodiment of the method, generating the output signal comprises, for each ray in the ray set, calculating a mixed signal contribution obtained by mixing the CW signal with the ray's calculated signal contribution . including doing Additionally, the output signal is generated by summing the mixed signal contributions of all rays in the ray set. Therefore, the output signal is produced after mixing all signal contributions with the CW signal.

According to a preferred embodiment of the device, the ray set comprises a plurality of rays and the processing unit is adapted to determine the emission start times of the rays in a random or uniform manner.

According to a preferred embodiment of the device, the processing unit is adapted to generate the ray set by assigning parts of the CW signal to the rays. The portion of the CW signal assigned to the beam begins at the emission start time and extends over the emission duration.

According to a preferred embodiment of the apparatus, the processing unit provides, for each ray in the ray set, a mixed signal contribution by mixing the CW signal with the calculated signal contribution of the ray and all is configured to generate the output signal by adding the mixed signal contributions of the rays of the .

In addition to computing devices, some or all components of a system can include hardware and software components. The hardware components may include at least one of a microcontroller, central processing unit (CPU), graphics processing unit (GPU), memory, and storage.

The invention will be explained in more detail with reference to exemplary embodiments shown in the accompanying drawings.

The accompanying drawings are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

Other embodiments of the present invention and many of the intended advantages of the present invention will be readily appreciated as it becomes better understood by reference to the following detailed description. The steps in the method are numbered for ease of reference, but such numbering does not necessarily imply the steps performed in that order unless expressly or implicitly stated otherwise. It should be understood that it is not. In particular, steps may be performed in a different order than indicated by their numbering. Some steps may be performed concurrently or with overlap.

FIG. 1 schematically shows a block diagram illustrating an apparatus for simulating sensor data of a CW lidar sensor according to one embodiment of the invention. FIG. 2 schematically shows the state of the lidar sensor in operation. FIG. 3 schematically shows a CW signal with sections assigned to rays. FIG. 4 schematically shows how light rays propagate from a transmitter to a receiver. FIG. 5 schematically shows how light rays propagate from the receiver to the transmitter. FIG. 6 schematically illustrates the generation of the output signal based on mixing the CW signal with the signal contribution of the beam. FIG. 7 shows a flow diagram of a method for simulating sensor data of a CW lidar sensor according to one embodiment of the invention. FIG. 8 schematically depicts a block diagram illustrating a computer program product according to one embodiment of the invention. FIG. 9 schematically illustrates a block diagram illustrating a non-transitory computer-readable storage medium according to one embodiment of the invention.

FIG. 1 schematically shows an apparatus 1 for simulating sensor data of a CW lidar sensor. Before describing the components of device 1 in more detail, the principle of operation of a CW lidar sensor will be described with reference to FIG.

As shown in FIG. 2, the lidar sensor's laser produces a continuous wave (CW) signal 101 . The laser is controlled such that the amplitude varies as a function of time. In this case, the lidar system is prepared for the amplitude modulated continuous wave (AMCW) method. The signals shown below are of varying amplitude, however, the LIDAR system can also be adapted for use with frequency modulated continuous wave (FMCW) methods. In this case, the frequency of the CW signal changes as a function of time. The transmitter Tx of the lidar system emits a CW signal. The CW signal is reflected by one or more objects in the scene 102 and is at least partially received by the lidar system's receiver Rx. The received signal is mixed with the original CW signal by the mixing section 103 of the lidar system. The signal thus obtained is radiated as output signal 104 .

In the following, the components of the device 1 for simulating sensor data of a CW lidar sensor are described in more detail.

Device 1 comprises an interface 4 adapted to receive data from and transmit data to external devices. Therefore, the interface 4 can be arranged as both an input and an output and can be used by other systems (e.g. WLAN, Bluetooth®, ZigBee®, Profibus, ETHERNET, etc.) or users (displays, It can be any kind of port or link or interface capable of communicating information to a printer, speaker, etc.).

The device 1 further comprises a processing unit 2 arranged to process data received from the interface 4 . The processing unit 2 includes a microcontroller (μC), an integrated circuit (IC), an application specific integrated circuit (ASIC), an application specific standard product (ASSP), a digital signal processor (DSP), a field programmable gate array ( It can be a central processing unit (CPU) or a graphics processing unit (GPU), such as an FPGA).

The processing unit 2 includes a ray set generation unit 21 in communication with the interface 4, a ray propagation unit 22 in communication with the ray set generation unit 21, and a signal contribution calculation unit 23 in communication with the ray propagation unit 22. , and an output signal generator 24 in communication with the signal contribution calculator 23 . These modules 21 to 24 may be part of the processing unit 2, may be mounted on the processing unit 2, or may be mounted on separate units communicably connected to the processing unit 2. may have been

The device 1 further comprises a storage unit 3 in communication with the signal generator 24 . The storage unit 3 can be a data storage device such as a magnetic core memory, a magnetic tape, a magnetic card, a hard disk drive, a floppy disk, or a magnetic storage device or memory such as a removable storage device. , or may comprise them. The storage unit 3 also stores, for example, holographic memory, optical tape, laser disc, phase writer, phase writer dual (PD), compact disc (CD), digital video disc (DVD), high definition DVD (HD DVD), Blu-ray It may be or comprise an optical storage device or memory such as a disc (BD) or Ultra Density Optical (UDO). The storage unit 3 further includes a magneto-optical storage device or memory such as a minidisk or a magneto-optical disk (MO-Disk), such as a random access memory (RAM), a dynamic RAM (DRAM), or a static RAM (SRAM). volatile semiconductor or solid-state memories, such as read-only memory (ROM), programmable ROM (PROM), erasable PROM (EPROM), electrical EPROM (EEPROM), flash EEPROM and, for example, USB sticks, ferroelectric RAM (FRAM (registered trademark)), magnetoresistive RAM (MRAM), or phase change RAM, or a data carrier/medium.

The device 1 receives certain input parameters via an interface 4, for example the waveform of the time-dependent CW signal, or other parameters describing the CW signal, for example parameters related to the phase or amplitude of the CW signal. can do. The input parameters can also include the maximum or minimum number of rays generated by the ray generator 21 . Furthermore, the device 1 can receive information about the simulated scene, such as the number, orientation and properties of objects in the simulated scene. The simulated scene corresponds to the artificial environment of the simulated lidar sensor. A laser beam emitted by a lidar sensor is reflected by objects placed in the simulated scene.

The ray generator 21 is adapted to generate a ray set containing at least one, preferably a plurality of rays.

In the following, the situation in which multiple rays are generated will be described in more detail. However, the invention is in principle also applicable to ray sets containing only one ray.

The ray generator 21 defines for each ray in the ray set the emission start time, the emission duration, the spatial origin in the simulated scene, and the emission direction. All rays can be omitted from the same spatial origin. However, the spatial origin may be different for different rays. The emission start time may be randomly selected. Therefore, the ray generator 21 can comprise a (pseudo-)random number generator. However, the ray generator 21 can also select the emission start times deterministically or according to a predefined distribution.

The light beam generator 21 assigns a given section of the CW signal to each light beam. The interval begins at the emission start time and spans the emission duration.

A ray propagator 22 then propagates all rays in the ray set through the simulated scene. The ray propagating portion 22 is adapted to use ray tracing methods known in the art. Ray tracing is a method of sampling geometry in a virtual scene known in computer graphics. In computer graphics, ray tracing is used to create images by shooting light rays from a camera and accumulating light on sensor pixels instantaneously, i.e. without taking into account the finite propagation time. be done. In contrast, ray tracing according to the present invention also takes into account finite propagation times.

Each ray is propagated through the simulated scene by computing the intersection of the current ray, i.e. the original emitted ray or an already scattered ray, and the nearest object in the simulated scene. , and then use a suitable physical model to compute the parameters of the reflection based on the properties of the object. Ray Propagator 22 calculates the path of each ray in the simulated scene by determining the (possibly multiple) reflections for each ray. Reflection of rays can have the additional effect that only a portion of the energy is received at the detection location. Therefore, the ray propagator 22 calculates the throughput of the ray to the detection position along the optical path in addition to the optical path itself.

The signal contribution calculator 23 is adapted to calculate the signal contribution of each propagating ray of the ray set at the detected position in the simulated scene. The signal contribution calculator 23 calculates the total length of the optical path of the ray from the spatial starting point of the ray to the detection position. The signal contributor 23 further calculates, for each ray, the travel time, ie the propagation time or flight time, from the total length of the corresponding optical path, based on the speed of light c as a conversion factor.

A signal contribution calculator 23 calculates the signal contribution of each ray based on the portion of the CW signal assigned to the ray. Here, the portion of the CW signal assigned to the ray is phase-shifted with respect to the original CW signal according to the calculated travel time of the optical path corresponding to the ray. A phase shift introduces a signal offset between the signal contribution and the CW signal. Additionally, the amplitude of the signal contribution of the beam can be adjusted according to the calculated throughput.

The output signal generator 24 computes the mixed signal contribution for each ray in the ray set by blending the original CW signal with the ray's computed signal contribution . The output signal generator 24 also generates the output signal by summing the mixed signal contributions of all rays in the ray set.

Output signal generator 24 may be configured to provide an output signal to a user via interface 4 . Additionally or alternatively, the output signal may be stored in the storage unit 3 .

Some aspects of the apparatus 1 for simulating sensor data of a CW lidar sensor are described in more detail with reference to FIGS. 3-6.

FIG. 3 shows an exemplary CW signal 5, which converts the output signal to produce a set of rays and by mixing the CW signal 5 with the calculated signal contributions of the rays. used to generate. As shown in FIG. 3, the CW signal is amplitude modulated, ie the CW signal 5 is an amplitude modulated continuous wave (AMCW) signal. According to different embodiments, the CW signal 5 may also be frequency modulated, ie a frequency modulated continuous wave (FMCW) signal.

For the rays illustrated in the ray set, the emission start time t_0 is set, for example 46 ns, measured against a predetermined initial time of 0 ns. A radiation duration T is also set, starting at the radiation start time t_0 and ending at the radiation end time t_1. The corresponding section or part of the CW signal 5 between the radiation start time t_0 and the radiation end time t_1 is assigned to the ray.

As shown in FIG. 4, a ray can start from the simulated lidar sensor transmitter Tx in the simulated scene 6 and propagate through the simulated scene. The transmitter Tx is then placed at the spatial origin of the ray in the simulated scene 6 . A ray is reflected from the first object 61 and then from the second object 62 to reach the detection position corresponding to the receiver Rx of the simulated lidar sensor.

Ray propagation may also occur in the opposite direction, as shown in FIG. That is, the ray is first reflected from the second object 62 and then from the first object 61, from the spatial origin of the ray located at the position of the receiver Rx of the simulated lidar sensor, and the ray can be traced until it reaches a detection position corresponding to the position of the transmitter Tx of the simulated lidar sensor in the simulated scene.

Referring to FIG. 6, the ray set generator 21 assigns portions of the CW signal 5 to rays as described in more detail above with reference to FIG. And after the ray propagator 22 propagates the ray through the simulated scene, the signal contribution calculator 23 calculates the corresponding signal contribution 71 of the ray. The output signal generator 24 comprises a mixer 72 that mixes the signal contribution 71 of the ray and the portion 51 of the CW signal starting at time t_2 when the ray was received at the detection location. The time t_2 corresponds to the sum of the light emission start time and the light travel time. The portion 51 of the CW signal 5 then generally differs from the portion of the CW signal assigned to the beam (starting at the emission start time) by a phase shift corresponding to the travel time of the beam. Additionally, the actual signal contribution of a ray can also be affected by the throughput of the ray along the optical path.

Output signal generator 24 generates mixed signal contribution 73 by mixing the CW signal with the calculated signal contribution of the rays. The output signal generator 24 will generate an output signal by adding the mixed signal contributions 73 of all the rays in the ray set for the plurality of rays in the ray set.

FIG. 7 shows a flow diagram of a method for simulating sensor data for a CW lidar sensor.

In a first method step S1 a ray set is generated comprising at least one ray, preferably a plurality of rays. The number of rays in a ray set may be fixed. The number of rays in the ray set may be randomly selected. The number of rays in the ray set can be selected to be greater than a predetermined minimum number. For example, a ray set can include at least 2, preferably at least 100, more preferably at least 500, and most preferably at least 1000 rays. Also, the number of rays in the ray set is selected to be less than a predetermined maximum number. For example, a ray set may include at most 100,000, preferably at most 10,000, more preferably at most 5000, and most preferably at most 2000 rays.

For each ray, the emission start time is determined relative to some predetermined time origin, such as 0 ns. Furthermore, the emission duration of the rays is determined. The emission duration may be the same for all rays, but may vary for different rays. The radiation duration of the rays can also follow a predetermined distribution. A spatial origin in the simulated scene is also determined for each ray. The spatial origin (corresponding to the point of emission of the ray) can be the same for all rays in the ray set. However, different rays may also contain different spatial origins. Also, the emission direction of each light beam is determined.

Radiation direction and/or radiation start time and/or radiation duration may be randomly selected or uniformly selected or sampled. The emission start time is preferably sampled in such a way that an accurate output signal is produced. In particular, the radiation start times are selected such that the portion of the CW signal assigned to the beam covers the entirety of at least one phase of the CW signal. The contribution of each ray can be adjusted using sampling theory, eg importance sampling.

In method step S2, each ray in the ray set propagates through the simulated scene. A simulated scene contains a plurality of objects having predetermined geometric shapes and physical properties. The positions and/or physical properties of objects in the simulated scene may be fixed or may vary over time. Ray propagation is performed using ray tracing algorithms known in the art. For each ray, the ray path is determined based on the ray's spatial origin, the ray's radial direction, and the ray's reflections on objects in the simulated scene. In addition, the throughput of rays along the optical path to the detection location is calculated.

In method step S3, for each ray in the ray set, the signal contribution of the propagated ray at the detection location in the scene is calculated. A signal contribution is calculated based on the calculated throughput and based on the travel time of the ray. The travel time of the ray can be calculated from the length of the optical path of the ray. The final travel time of the ray leads to a signal offset between the signal contribution and the CW signal. A signal offset is determined based on the start time of emission of the light beam and the travel time of the light beam along the optical path.

In method step S4, each calculated signal contribution is mixed with the CW signal to calculate the mixed signal contribution of the corresponding ray. The mixing of the CW signal with the calculated signal contribution is based on the signal offset. The mixed signal contributions of all rays are added to generate the output signal.

In method step S5, the output signal is stored in the memory store 3. FIG. Additionally or alternatively, the output signal is output to an output unit 4, such as a display or printer.

FIG. 8 schematically shows a block diagram showing a computer program product P containing executable program code PC. The executable program code PC, when executed (eg, by a computing device), is configured to perform a method for simulating sensor data of a CW lidar sensor according to the present invention.

FIG. 9 schematically illustrates a block diagram illustrating a non-transitory computer-readable storage medium M that, when executed (eg, by a computing device), implements the present invention. It comprises executable program code MC configured to perform a method for simulating sensor data of such a CW lidar sensor.

All of the advantageous options, differences in modifications, and the foregoing description of embodiments of an apparatus for simulating sensor data of a CW lidar sensor described herein have been described for simulating sensor data of a CW lidar sensor. can equally be applied to embodiments of the method of and vice versa.

In the foregoing Detailed Description, various features are grouped together in one or more examples for the purpose of streamlining the disclosure. It should be understood that the above description is intended to be illustrative and not restrictive. It is intended to cover alternatives, modifications and equivalents. Many other examples will be apparent to those of skill in the art upon reviewing the above specification.

Although particular embodiments have been illustrated and described herein, it will be appreciated by those skilled in the art that various alternatives and/or equivalent implementations exist. It should be understood that the exemplary embodiment or exemplary embodiments are illustrative only and are not intended to limit the scope, applicability, or configuration in any way. Rather, the foregoing summary and detailed description provide those skilled in the art with a convenient road map for implementing at least one exemplary embodiment, and the scope of the appended claims and their legal implications. It is to be understood that various changes may be made in the function and arrangement of elements described in the illustrative embodiments without departing from the scope described in equivalents. In general, this application is intended to cover any adaptations or variations of the specific embodiments discussed herein.

Certain terminology used in the foregoing specification is used to provide a thorough understanding of the present invention. However, it will be apparent to one skilled in the art that the specific details are not required to practice the present invention in light of the specification provided herein. Accordingly, the foregoing descriptions of specific embodiments of the invention are presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed, and obviously many modifications and variations are possible in light of the above teachings. be. This embodiment was chosen and described in order to best explain the principles of the invention and its practical application, thereby enabling those skilled in the art to adapt it to the particular application contemplated. The present invention and various embodiments may be best utilized with various modifications. Throughout this specification, the terms "including" and "in which" are the Plain English equivalents of the Simple English terms "comprising" and "wherein" used respectively as Furthermore, the terms "first," "second," and "third," etc. are used merely as labels to impose numerical requirements on their subject, or It is not intended to establish any particular ranking of the importance of those subjects. In the context of this specification and claims, the conjunction "or" is an inclusive disjunction ("and/or") and not an exclusive disjunction ("... or shall be understood as either...or").

1: device 2: processing unit 21: ray generation unit (ray set generation unit)
22: light beam propagation unit 23: signal contribution calculation unit 24: output signal generation unit 3: storage unit 4: output unit (interface)
5: CW signal 51: Part 6: Scene 61: Object 62: Object 71: Signal contribution
72: Mixer 73: Mixed signal contribution
101: Signal 102: Scene 103: Mixer 104: Output signal M: Computer readable storage medium MC: Executable program code P: Computer program product PC: Program code Rx: Receiver Tx: Transmitter T: Radiation duration t_0: Radiation start time t_1: Radiation end time

Claims (15)

A method for simulating sensor data for continuous wave optical detection and ranging (CW lidar sensor), comprising:
generating, based on the CW signal (5), a ray set comprising at least one ray, each ray in the ray set having a radiation start time t_0 and a radiation duration T;
for each ray in said ray set, propagating said ray through a simulated scene (6) containing at least one object;
calculating, for each ray in said ray set, the signal contribution (71) of said ray propagated at a detection location in said simulated scene (6);
generating an output signal (104) based on mixing the CW signal with the calculated signal contribution of the rays in the ray set, wherein the CW signal is typically the differ from the calculated signal contribution (71) by a phase shift corresponding to the time traveled by the ray; and
performing at least one of storing and outputting said output signal (104);
A method, including each ray in the ray set includes a spatial origin in the simulated scene (6) and a radial direction in the simulated scene (6);
The method of claim 1. Propagating the light beam comprises:
determining the path of the ray based on the spatial origin of the ray, the radial direction of the ray and the reflection of the ray on the objects (61, 62) in the simulated scene (6) a step of determining;
calculating the throughput of the beam along the optical path to the detection location;
including
calculating the signal contribution (71) is based on the calculated throughput;
3. The method of claim 2, wherein the throughput of the light beam relates to the fraction of energy received at the detection location due to reflection of the light beam along the optical path. For each ray in the ray set, mixing the CW signal (5) with the signal contribution (71) of the calculated ray is the difference between the signal contribution (71) and the CW signal (5). Based on signal offset,
the signal offset is determined based on the radiation start time t_0 of the light beam and the optical path of the light beam;
4. The method of claim 3. the radial direction is randomly selected or uniformly selected;
The method according to any one of claims 2-4. the ray set includes a plurality of rays;
the emission start time t_0 of the light beam is randomly selected or uniformly selected;
The method according to any one of claims 1-5. The CW signal (5) is a frequency modulated continuous wave (FMCW) signal or an amplitude modulated continuous wave (AMCW) signal,
The method according to any one of claims 1-6. generating a set of rays comprises, for each ray, assigning a portion (51) of said CW signal to said ray;
said portion (51) of said CW signal begins at said radiation start time t_0 and has a length equal to said radiation duration T;
The method according to any one of claims 1-7. The step of generating the output signal (104) comprises:
calculating, for each ray in said ray set, a mixed signal contribution (73) by mixing said CW signal (5) with said calculated signal contribution (71) of said ray;
generating said output signal (104) by summing said mixed signal contributions (73) of all rays in said ray set;
The method according to any one of claims 1 to 8, comprising An apparatus (1) for simulating sensor data for continuous wave optical detection and ranging (CW lidar sensor), comprising:
generating, based on the CW signal (5), a ray set comprising at least one ray, each ray in the ray set having a radiation start time t_0 and a radiation duration T;
for each ray in said ray set, propagating said ray through a simulated scene (6) containing at least one object;
calculating, for each ray in said ray set, the propagated signal contribution (71) of said ray at a detection location in said simulated scene (6); and
generating an output signal (104) based on mixing said CW signal (5) with said calculated signal contribution (71) of said rays in said ray set, wherein said CW signal is typically said differs from the calculated signal contribution (71) of the ray in the ray set by a phase shift corresponding to the time traveled by the ray;
a processing unit (2) adapted to
at least one of a storage unit (3) adapted to store said output signal (104) and an output unit (4) adapted to output said output signal (104);
A device comprising: the ray set includes a plurality of rays;
said processing unit (2) is adapted to determine said radiation start time t_0 of said light beam in a random or uniform manner;
11. Apparatus according to claim 10. said processing unit (2) is adapted to generate said ray set by assigning, for each ray, a portion (51) of said CW signal to said ray,
12. Apparatus according to claim 10 or 11, wherein said part (51) of said CW signal starts at said radiation start time t_0 and has a length equal to said radiation duration T. The processing unit (2) is
calculating, for each ray in said ray set, a mixed signal contribution (73) by mixing said CW signal (5) with said calculated signal contribution (71) of said ray, and all in said ray set; generating said output signal (104) by summing said mixed signal contributions (73) of rays of
adapted to generate said output signal (104) by
A device according to any one of claims 10-12. A computer program product (P ). Non-transitory computer comprising executable program code (MC) configured to perform the method according to any one of claims 1 to 9 when executed by a computer device (1) A readable storage medium (M).