Method for simulating remote sensing of a real environment using a lidar sensor

Abstract

One aspect of the invention relates to a method (100) for simulating remote sensing of a real environment performed by a lidar sensor, the method comprising the steps of: Generating (120) a set of particles, each particle of the set of particles corresponding to a laser shot performed by the lidar sensor in the real environment and each particle of the set of particles having a time value; Calculating (130), for each particle of the set of particles, a direction of the laser shot represented by said particle, the direction of the laser shot being calculated using the time value of the current particle; and Tracing (140), for each particle of the set of particles, a corresponding segment, the segment being defined by a point having the position of the digital representation of the lidar sensor.a length equal to a predetermined maximum range of the lidar sensor and a direction equal to the laser firing direction calculated for the current particle, and Calculation (150), for each particle in the particle set, of a corresponding collision distance, the collision distance being the distance between the point defining said corresponding segment and a collision position of the corresponding segment with the 3D digital scene. Figure to be published with the abbreviation: Figure 1,

Description

Title of the invention: Method for simulating remote sensing of a real environment using a lidar sensor. TECHNICAL FIELD OF THE INVENTION

[0001] The technical field of the invention is that of laser remote sensing.

[0002] The present invention relates to a method for simulating remote sensing of a real environment carried out by a lidar sensor. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0003] Laser remote sensing, or lidar (light detection and ranging), is a remote measurement technique based on analyzing the properties of a light beam reflected back to its emitter. A lidar sensor is therefore a device that scans its environment and produces a 3D point cloud. To do this, the lidar sensor contains a laser that is fired at high frequency in given directions. The laser beam bounces off an object in the environment and returns to the lidar sensor. The lidar sensor then measures the travel time of the laser beam to deduce the distance to the object struck. Lidar sensors have applications in many fields, particularly in assisting with the guidance of land vehicles or aircraft.

[0004] In order, for example, to design an aircraft guidance system incorporating one or more lidar sensors, or to develop algorithms that exploit the data produced by the lidar sensor, it is useful to simulate the behavior of a lidar sensor under future operating conditions. For instance, it is useful to simulate the behavior of one or more lidar sensors mounted on the landing gear of an airliner in order to provide the pilot with better perception of the aircraft's surroundings during ground maneuvers. To be able to conduct tests without waiting for the completion of the physical prototype, digital models of an airport, an aircraft, and the future lidar sensor(s) are used to simulate the behavior of the lidar sensor(s) during ground maneuvers when they are positioned on the landing gear.

[0005] The simulation of the behavior of a lidar sensor is based on: • the generation of a point cloud seen by the lidar sensor, the generation being carried out at a high frequency and each point of the point cloud corresponding to a laser shot bouncing off a surface of a 3D object in the real environment and returning to the lidar sensor, and • the simulation of the real environment, making it possible in particular to faithfully reproduce the movements of the different 3D objects of the real environment between two generations of the point cloud.

[0006] Various 3D environment simulation software exist and for some years now, video game engines, such as Unreal Engine, have also been used in industry and for engineering.

[0007] A first category of methods for simulating the behavior of a lidar sensor is based on the use of "shaders." A shader is a computer program, used in computer graphics, to parameterize part of the rendering process performed by a graphics card or rendering engine. For example, it is possible to use custom shaders to render a material. Through the use of shaders, it is possible to access intermediate images generated by the rendering engine, such as the depth of the scene as seen by a camera. These intermediate images are then manipulated and sampled to imitate the operation of a lidar sensor. However, the methods in the first category require generating dense data from the real environment, i.e., images, in order to then subsample them into a point cloud.

[0008] A second category of methods for simulating the behavior of a lidar sensor is based on ray tracing, which consists of simulating the reverse path of light by calculating the illumination from the camera to the objects and then back to the lights. This solution exploits the power of graphics processing units (GPUs) dedicated to this type of calculation in order to achieve very high performance. However, methods in the second category require modifying the implementation of the sequence of operations generally performed by a graphics card necessary for rendering a batch of data, called a "pipeline." For example, it is necessary to add calculation steps to the rendering pipeline. This is sometimes impossible to implement, for example, if the pipeline owner does not allow it. Moreover, it can require considerable development work.Finally, the hardware acceleration functions of ray tracing are specific to each operating system and not cross-platform.

[0009] A third category of methods for simulating the behavior of a lidar sensor is based on a tool, for example called "LineTrace" or "RayCast" or "LineCast" in English, found in video game engines, which allows a line segment to be drawn between two given positions and the digital environment to be queried to check if a geometric shape intersects the segment. However, the main drawback of the methods in this third category is that the segments are drawn iteratively. Thus, the performance of these methods in the third category may be limited and may not allow for the simulation of a lidar sensor. having a high operating frequency. Indeed, one of the main challenges in simulating the behavior of a lidar sensor is that the lidar sensor fires lasers and produces 3D points at a very high frequency, i.e., on the order of several hundred thousand, or even millions, of shots per second, while the overall simulation of the environment must be in real time, and therefore updated at a minimum frequency of 30 hertz. In simulation, time is discretized, and all the calculations necessary to update the real environment must be performed in a sufficiently short time to maintain a regular refresh rate of the overall simulation. If all the calculations for simulating the behavior of the lidar sensor are not performed efficiently, either the frequency of updating the digital environment decreases, or some laser shots will not be simulated.When the frequency of updates to the digital environment decreases, it becomes less and less possible to accurately capture rapid movements, and the simulation therefore loses its appeal.

[0010] There is therefore a need to provide a method for simulating the behavior of a lidar sensor which limits, at least in part, the problems associated with the use of the aforementioned prior art methods. Summary of the invention

[0011] The invention offers a solution to the problems mentioned above, by allowing the simulation of remote sensing of a real environment carried out by a lidar sensor by generating a set of particles, each particle thus allowing the simulation of a laser shot carried out by the lidar sensor.

[0012] One aspect of the invention relates to a computer-implemented method for simulating remote sensing of a real environment performed by a lidar sensor, the simulation being carried out at a simulation frequency f, the method comprising the steps of: • Generation of a 3D digital scene representing the real environment and the lidar sensor, the generation of the 3D digital scene being carried out for a time t, • Generation of a set of particles, each particle in the set corresponding to a laser shot performed by the lidar sensor in the real environment and each particle in the set having a temporal value tparticle with: If tparticle — t + A t • With : • At, a duration of time equal to 2, • Calculation, for each particle in the set of particles, of a laser firing direction represented by said particle, the laser firing direction being calculated using the time value of the current particle, and • Tracing, for each particle in the set of particles, of a corresponding segment, the segment being defined by a point having the position of the digital representation of the lidar sensor, a length equal to a predetermined maximum range of the lidar sensor and a direction equal to the direction of the laser firing calculated for the current particle, • Calculation, for each particle in the set of particles, of a corresponding collision distance, the collision distance being the distance between the point defining said corresponding segment and a collision position of the corresponding segment with the 3D digital scene, and • Storing a simulation result, the simulation result including, for each particle in the set of particles: • the direction of the corresponding segment, and • the corresponding collision distance.

[0013] The method according to the invention makes it possible to implement a simulation of a Lidar sensor that is simple in terms of development cost while achieving significant performance in terms of computation time; in particular, the simulation can be performed in real time while reproducing the operating frequency of a Lidar sensor. Finally, the method according to the invention is compatible with different hardware and operating systems such as Windows, notably using the DirectX graphics library, or Linux, using the Vulkan graphics library.

[0014] In addition to the characteristics mentioned in the preceding paragraph, the method according to one aspect of the invention may have one or more additional characteristics from among the following, considered individually or in all technically possible combinations: • The method further comprising a step, carried out after the calculation of the collision distance and before the storage of the simulation result, of encoding the simulation result into an image, comprising a number P of pixels greater than a number N of particles in the particle set, a pixel of the image comprising at least three channels intended to store respectively: • an azimuth of the direction of the corresponding segment, • an elevation of the direction of the corresponding segment, and • the corresponding collision distance, • The process further comprising at least one final step from among: Displaying an image of a rendered 3D digital scene and all the particles located within the 3D scene, the location of each particle in the set of particles being obtained from the simulation result, Modification, based on the simulation results, of the lidar sensor's position in the real environment. Modification, based on the simulation results, of at least one parameter of the lidar sensor, and Modification, based on the result of the simulation, of at least one position and / or at least one volumetric shape of an object in the real environment. • The temporal value tpartiad of each particle in the set of particles is equal to: tparticle ~ t + (îdparticle* N ) • With: • time to generate the 3D digital scene, expressed in milliseconds from a predetermined initial time • îdparticw[e, a unique identifier of the current particle, the identifier being an integer between 0 and N, • At a time interval between the generation of the current 3D digital scene, performed at time t, and the next generation of the 3D digital scene, performed at time t+A, • the particle set comprises a number N of particles equal to: N = f,.r * A / haar • With : * J Hdae 'a lidar frequency expressed in hertz, i.e. the number of laser shots made in one second by the lidar sensor, the calculation of the laser shot direction is performed using a function f defined by: laser — particle^ • With : • dirlaser, the direction of the laser beam, and • / , the function specifying the evolution over time of the direction of the Laser firing from the lidar sensor • The real environment is an airport in which the lidar sensor is attached to the landing gear of an aircraft moving on the ground in the airport.

[0015] Another aspect of the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, lead the computer to implement the process according to the invention.

[0016] An additional aspect of the invention relates to a computer-readable recording medium comprising instructions which, when executed by a computer, cause the computer to implement the process according to the invention.

[0017] A final aspect of the invention relates to a system comprising the means adapted to carry out the process according to the invention.

[0018] The invention and its various applications will be better understood upon reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0019] The figures are presented for illustrative purposes only and are in no way limiting of the invention. • Fig. 1 is a synoptic diagram illustrating the steps of an example of the simulation process according to the invention. • Fig. 2 is an example of an image that can be obtained from a simulation result of the simulation process according to the invention. DETAILED DESCRIPTION

[0020] Unless otherwise specified, the same element appearing on different figures has a unique reference.

[0021] Figure 1 is a block diagram illustrating the steps of an example of process 100 according to the invention. The mandatory steps of the example of process 100 are indicated by a solid rectangle and the optional steps are indicated by a dashed rectangle.

[0022] The method 100 is computer-implemented. By "computer-implemented," it is meant that the steps, or virtually all of the steps, are executed by at least one computer or processor or other similar system. Thus, steps are carried out by the computer, possibly fully automatically, or semi-automatically. In examples, the triggering of at least some of the steps of these methods can be achieved through user-computer interaction. The level of user-computer interaction required may depend on the level of automation intended and be balanced against the need to implement the user's wishes. In examples, this level may be user-defined and / or predefined.

[0023] A typical example of a computer implementation of process 100 consists of running process 100 with a system adapted for this purpose. The system may include a processor coupled to memory and a graphical user interface (GUI), the memory having stored a computer program comprising Instructions for implementing the process. Memory can also store a database. Memory is any hardware suitable for such storage, possibly comprising several distinct physical parts.

[0024] Method 100 is a method for simulating a lidar sensor in operation. In other words, the method simulates remote sensing, with remote sensing being performed in a real environment by a lidar sensor. Thus, with method 100, the remote sensing of a real environment performed by a lidar sensor is simulated. The lidar sensor is located in the real environment at all times and can be moved. Remote sensing is a method that makes it possible to obtain information about a real environment by collecting and analyzing data without direct contact between the instrument used and the environment being analyzed.

[0025] In one example, the real environment is: • An airport including an aircraft on which the lidar sensor is attached to the landing gear, • A road or parking lot containing a car to which the lidar sensor is attached, or • A factory comprising an object such as a moving robot on which the lidar sensor is attached.

[0026] Steps 110 to 170 of process 100 can be performed iteratively at a predetermined frequency f. The predetermined frequency f can be fixed, for example, greater than 30, 60, or 120 hertz. The frequency f can also be variable, for example, depending on the time required to perform all the calculations. When the frequency f is variable, it can preferably remain greater than 30 hertz. The time interval At corresponds to the time interval between two iterations of process 100. The time interval At is therefore equal to:

[0027] &t=j

[0028] A step 110 of the process 100 includes the generation of a digital scene 3D. The 3D digital scene faithfully represents the real environment. Alternatively, the 3D digital scene may faithfully reproduce only certain features of the real environment. Furthermore, the real environment may be one that does not yet exist at the time of the simulation but will be constructed or obtained as a result of modifications to a real environment. In this application, the term "faithful" means that the 3D digital scene reproduces the geometry of the real environment exactly, with an error of less than 1, 5, or 10 centimeters. The real environment may include objects that are also faithfully represented in the 3D digital scene by digital models. These objects may, in particular, move. in the real-world environment. When an object moves within the scene, such as the lidar sensor or any object to which the lidar sensor is attached, the generation of the 3D digital scene involves updating the position and orientation of the digital model representing that object. It should be noted that the digital model(s) of the real-world environment, as well as the objects within it, can be obtained beforehand by the computer. These digital models can then be stored in memory, and the generation of the digital scene consists of updating their position and orientation at a given time t.

[0029] The generation 110 of the 3D digital scene is performed for a time t. In other words, the generation 110 of the 3D digital scene is performed for an iteration i out of a set of simulation iterations carried out at frequency f. When the simulation frequency is fixed, the time t can be calculated according to the following formula:

[0030] t= t0 + P Af

[0031] With: • ^o, the time of a start of the simulation.

[0032] Step 110 can be performed using a video game engine such as Unreal Engine. In order to obtain the generation 110 of the 3D digital scene for a given time t, it is possible to provide the video game engine with various input elements such as: • the 3D digital scene, • physical data on one or more digital models of the 3D digital scene, such as, for example, their current position, velocity and acceleration, and • a simulation time step.

[0033] A step 120 of the process 100 includes the generation of a set of particles. The set of particles can comprise a number N of particles, for example between 500 and 500,000. The number N of particles can be obtained using the following formula:

[0034] N= f... * MLJ lidar

[0035] With: * f lidar' 'a lidar frequency expressed in hertz, i.e., the number of laser shots performed per second by the lidar sensor during remote sensing.

[0036] Each particle in the set of particles corresponds to a laser shot performed by the lidar sensor in the real environment. Indeed, in the following steps of method 100, the location of each particle will numerically represent a collision between a laser shot performed by the lidar sensor and the real environment. Thus, the set of localized particles is a result of the simulation of the remote sensing of the real environment performed by the lidar sensor, since it allows for the reproduction

[0037]

[0038]

[0039] faithfully represents the result of remote sensing performed by the lidar sensor in the real environment. It is of course possible, when steps 110 to 170 are iterated, to group the sets of particles from each iteration into a single set of particles, thus allowing the simulation of remote sensing performed for a longer duration than the time interval A t. Each particle in the particle set has a time value tparticle. This time value corresponds to the time at which, in the real world, the laser firing would be performed by the lidar sensor. The time value tparticle is between the time t of the current iteration, i.e., the time t at which generation 110 of the 3D digital scene was performed, and the time t + At of the next iteration, i.e., the next generation 110 of the 3D digital scene. In an example, consistent with the previous examples, the time value tparticule of each particle is calculated according to the following formula: tparticle ~ t +

[0040] With:

[0041] idparticule, a unique identifier of the current particle, the identifier being a number integer between 0 and N.

[0042] The generation of the particle set can be performed using a visual effects tool. This type of tool allows for the creation of a wide variety of visual effects, such as flames, debris, dust, and any other visual effect from the world of video games. This tool allows for the efficient manipulation of a large number of particles. The main advantage of this tool, which enables this innovation, is that it offers the option of creating a visual effect whose calculations are performed in parallel for each particle using graphics processing units (GPUs). This makes the tool even more efficient for managing a very large number of particles. An example of a visual effects tool compatible with the invention is the tool called Niagara. The Niagara tool allows, in particular, the creation and management of particles within a 3D digital scene generated by the Unreal Engine video game engine.One advantage of this tool is its compatibility with different operating systems such as Windows or Linux, thus reducing the development time of a process using this tool.

[0043] Steps 130 to 150 can be performed by parallelizing the calculations carried out for each particle in the particle set. In other words, the calculations of at least one of the steps 130 to 150 can be performed in parallel in different processes. Indeed, the calculations for each particle in the particle set are independent of each other during these steps 130 to 150. Thus, the computation time of steps 130 to 150 can be greatly reduced in order to reduce The time duration At. For example, process 100 can be performed in a time duration At of less than 33 milliseconds, or even 16 milliseconds or 8 milliseconds. In order to parallelize the calculations of at least one step among steps 130 to 150, the computer implementing process 100 may include a graphics processing unit, commonly called a GPU.

[0044] A step 130 of the process 100 comprises calculating, for each particle in the particle set, a direction of the laser beam represented by said particle. Thus, this step determines the direction in which the particle will be sent in order to simulate the corresponding laser beam. The direction of the laser beam, or equivalently, the direction of particle emission, is calculated using the temporal value of the current particle. Indeed, a Lidar sensor, during remote sensing, modifies the orientation of the laser beam as the remote sensing progresses; this step therefore aims to reproduce this behavior of the Lidar sensor.

[0045] In an example, consistent with the preceding examples, the calculation of the laser firing direction is performed using a function f defined by:

[0046] director luser =

[0047] With:

[0048] due to the laser - the direction of the laser firing, and

[0049] f, the function specifying the evolution over time of the direction of the laser firing of the lidar sensor.

[0050] The function f may be specific to the simulated lidar sensor. For example, the function f may correspond to a circular band scan scheme used by many known lidar sensors.

[0051] A step 140 of the process 100 comprises tracing, for each particle in the set of particles, a corresponding segment. This segment can be traced using a "LineTrace" tool from a video game engine, for example. The segment is defined by: • a point having the position of the digital representation of the lidar sensor, • a length equal to a predetermined maximum range of the lidar sensor, And • a direction equal to the direction of the laser beam calculated for the current particle,

[0052] In an example consistent with the preceding examples, the direction of the laser beam is defined by two angles: azimuth and elevation. These two angles are relative to the orientation of the lidar sensor, or more precisely to the digital model representing the lidar sensor. The orientation, in the 3D digital scene, of the digital model of the The lidar sensor is known at each time t of the simulation, for example at the time of generation 110 of the 3D digital scene. The orientation of the digital model of the lidar sensor can also be considered constant throughout the entire duration At for a given time t.

[0053] A step 150 of the method 100 comprises calculating, for each particle in the particle set, a corresponding collision distance. The collision corresponds to the intersection of the segment plotted in step 140 for said particle and the 3D digital scene. The collision distance is the distance between the point defining said corresponding segment, i.e., the position of the lidar sensor at time t, and a collision position of the corresponding segment with the 3D digital scene.

[0054] An optional step 160 of the method 100 includes encoding the simulation result into an image. This step 160 can be performed after step 150, which calculates the collision distance, and before step 170, which stores the simulation result. The image used can include a number P of pixels greater than the number N of particles in the particle set. Preferably, when the time interval At is variable, the number P of pixels is, for example, equal to P = a*N with 2 <a< 10. chaque pixel de l’image comprend les informations permettant définir la position particule. par exemple, un peut comprendre au moins trois canaux destinés à stocker respectivement : • an azimuth of the direction of the segment corresponding to a particle, • an elevation of the direction of the segment corresponding to a particle, and • the collision distance corresponding to a particle.

[0055] Encoding 160 of the simulation result in an image is a practical and efficient way to release the information from the tool in charge of generating and managing particles while leaving it on the graphics card of the computer implementing the method 100.

[0056] A step 170 of the process 100 includes storing the result of the remote sensing simulation of the real environment performed by the lidar sensor. The simulation result includes, for each particle in the particle set, information enabling the position of said particle to be defined, i.e., equivalently, the position of the collision between the traced segment 150 and the 3D digital scene. Thus, the simulation result includes, for each particle in the particle set: • the direction of the corresponding segment, and • the corresponding collision distance.

[0057] Equivalently, the result could include another dataset allowing the position of each particle to be defined.

[0058] Steps 180 to 210 are examples of using the simulation result obtained using method 100. Other uses of the simulation result are also possible, such as using the simulation result to develop algorithms that exploit the results of remote sensing performed by a lidar sensor. This use is particularly useful when the lidar sensor is still under development or when implementing lidar remote sensing of a real environment is complex, time-consuming, and expensive due to, for example, the characteristics of the real environment. For instance, it is possible to modify the position of a lidar sensor attached to the landing gear of an aircraft. The simulation results can also be sent via messages, for example, UDP (User Datagram Protocol) messages, to be read by a third-party application.Thus, the simulation results are processed by the third-party application in the same way as remote sensing results provided by the lidar sensor.

[0059] An optional step 180 of the method 100 includes displaying an image of a rendering of the 3D digital scene and of all the particles located in the 3D scene. An example of such an image is provided in [Fig. 2]. The location of each particle in the set of particles can be obtained from the simulation result stored in step 170.

[0060] An optional step 190 of the method 100 includes changing the position of the lidar sensor in the real environment. For example, when the lidar sensor is attached to an object in the real environment, it is possible to change the position of the attachment.

[0061] The modification of the sensor's position in the real environment is based on the simulation results. For example, the simulation results may show that an object, or part of an object, partially obscures the lidar sensor, thus limiting the angle at which remote sensing is performed. The term "partially obscures" means that the object generates a collision with at least one laser pulse from the lidar sensor. Typically, this can be caused by an object that is very close to the lidar sensor, for example, less than one meter, resulting in unwanted collisions during remote sensing of the real environment.

[0062] An optional step 200 of the method 100 includes modifying at least one parameter of the lidar sensor. This step may, for example, include modifying the angular range of the lidar sensor. Indeed, as with step 190, the simulation may show that an object or part of an object partially obscures the lidar sensor, and therefore it is necessary to modify the angular range of the lidar sensor.

[0063] An optional step 210 of the method 100 includes modifying at least one position and / or at least one volumetric shape of an object in the real environment. This step may, for example, include modifying the position of an object that at least partially obscures the lidar sensor in order to reduce or even eliminate the occlusion generated by that object. This step may also, for example, include modifying the volumetric shape of an object that at least partially obscures the lidar sensor in order to reduce or even eliminate the occlusion generated by that object.

Claims

1. Demands A computer-implemented method (100) for simulating remote sensing of a real environment using a lidar sensor, the simulation being carried out at a simulation frequency f, the method comprising the steps of: - Generation (110) of a 3D digital scene representing the real environment and the lidar sensor, the generation of the 3D digital scene being carried out for a time t, - Generation (120) of a set of particles, each particle in the set of particles corresponding to a laser shot performed by the lidar sensor in the real environment and each particle in the set of particles having a time value tpartwule with: particle — t + A t With : • At, a duration of time equal to -1, • Calculation (130), for each particle in the set of particles, of a laser firing direction represented by said particle, the laser firing direction being calculated using the time value tparticule of the current particle, and • Tracing (140), for each particle in the particle set, of a corresponding segment, the segment being defined by a point having the position of the digital representation of the lidar sensor, a length equal to a predetermined maximum range of the lidar sensor and a direction equal to the direction of the laser shot calculated for the current particle, • Calculation (150), for each particle in the particle set, of a corresponding collision distance, the collision distance being the distance between the point defining said corresponding segment and a collision position of the corresponding segment with the 3D digital scene, and • Storage (170) of a simulation result, the simulation result including for each particle in the set of particles: • the direction of the corresponding segment, and • the corresponding collision distance.

2. A simulation method (100) according to claim 1 further comprising a step (160), carried out after the calculation (150) of the collision distance and before the storage (170) of the simulation result, of encoding the simulation result in an image, comprising a number P of pixels greater than a number N of particles in the particle set, a pixel of the image comprising at least three channels intended to store respectively: - an azimuth of the direction of the corresponding segment, - an elevation of the direction of the corresponding segment, and - the corresponding collision distance.

3. A simulation method (100) according to claim 1 or 2 further comprising at least one final step among: - Displaying (180) an image of a rendering of the 3D digital scene and of all the particles located in the 3D scene, the location of each particle of the set of particles being obtained from the result of the simulation, - Modifying (190), according to the result of the simulation, the position of the lidar sensor in the real environment, - Modifying (200), according to the result of the simulation, at least one parameter of the lidar sensor, and - Modifying (210), according to the result of the simulation, at least one position and / or at least one volumetric shape of an object in the real environment.

4. A simulation method (100) according to any one of the preceding claims, wherein the time value ^particle of each particle in the set of particles is equal to: tparticle ~ 14- (idparticle N ) With: - *, time of the generation of the 3D digital scene, expressed in milliseconds from a predetermined initial time ^o, - idparticule, a unique identifier of the current particle, the identifier being an integer between 0 and N, - A f, a time interval between the generation of the current 3D digital scene, carried out at time t, and the next generation of the 3D digital scene, carried out at a time t+ A t.

5. A simulation method (100) according to any one of the preceding claims in which the particle set comprises a number N of particles equal to: N = f,., * Af lidar With: - fudar, 'the lidar frequency expressed in hertz, i.e. the number of laser shots made in one second by the lidar sensor.

6. A simulation method (100) according to any one of the preceding claims wherein the calculation (130) of the laser firing direction is carried out using a function f defined by: laser particle) With: - ^laser, the laser firing direction, and - f, the function specifying the time evolution of the laser firing direction of the lidar sensor.

7. Method (100) of simulation according to any one of the preceding claims wherein the real environment is an airport and wherein the lidar sensor is fixed to a landing gear of an aircraft moving on the ground in the airport.

8. Product computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method according to any one of the preceding claims.

9. A computer-readable recording medium comprising instructions which, when executed by a computer, 17 lead him to implement the process according to any one of claims 1 to 7.

10. System comprising means adapted to carry out the process according to any one of claims 1 to 7.

Images