Control of a light module to interact with a person outside a vehicle

Abstract

The invention relates to a method for controlling a vehicle lighting module capable of projecting a first pixelated beam in the visible wavelength range and a second pixelated beam in the non-visible range. Upon obtaining information (301) indicating the presence of a person in an environment external to the vehicle, the method comprises determining (302) a first light pattern and a second light pattern and controlling (303.1; 303.2) the formation of the first light pattern in the first projected pixelated beam and the second light pattern in the second projected pixelated beam. The first light pattern comprises at least one interactive light element, and the method comprises obtaining (304.1; 304.2) depth information of the environment determined based on a representative image of the projection of the second pattern.An interaction between the person and an interactive light element of the first light pattern is determined (305) and an action is implemented (306-308) depending on the interactive light element for which the interaction with the person is determined. FIG. 3.

Description

Title of the invention: Control of a light module to interact with a person outside a vehicle

[0001] The present invention relates to the field of controlling at least one lighting module of a motor vehicle. More specifically, the invention concerns a method and a control module for at least one lighting module of a motor vehicle, for interacting with a person outside the motor vehicle.

[0002] Motor vehicles are equipped with several light modules, in particular to perform lighting and / or signaling functions.

[0003] Certain light modules can perform lighting functions by projecting a pixelated light beam, obtained from a light source comprising a plurality of individually activatable light elements. This allows for greater precision in the functions performed, as well as a greater diversity of achievable functions.

[0004] For example, adaptive lighting is permitted by varying the lighting power applied to different areas of the scene facing the vehicle. Other applications allow the projection of an animation, for example in a passenger welcome scenario for the vehicle.

[0005] However, it is not currently possible to interact with a person located outside the vehicle using such vehicle light modules.

[0006] It follows that the use of a light module projecting a pixelated light beam is for the moment limited to lighting or animation functions, but without the possibility of interaction.

[0007] The present invention improves the situation.

[0008] A first aspect of the invention relates to a method for controlling at least one light module for a motor vehicle, the at least one light module comprising at least one pixelated light source and being capable of projecting a first pixelated light beam in a first wavelength range and of projecting a second pixelated light beam in a second wavelength range, the first wavelength range comprising visible wavelengths and the second wavelength range comprising non-visible wavelengths, the method comprising the following steps: - upon obtaining information indicating the presence of a person in an environment external to the motor vehicle, determining a first light pattern to be projected into the external environment in the first wavelength range and a second light pattern to be projected into the environment exterior in the second wavelength range in a common projection area with the first light pattern; - control of at least one light module to form the first light pattern in the first projected pixelated light beam and to form the second light pattern in the second projected pixelated light beam. The first light pattern includes at least one interactive light element and the process further includes the following steps: - obtaining at least one descriptive information of a relative position between the first projected light pattern and the person, said at least one descriptive information including depth information of the external environment determined from a representative image of the projection of the second light pattern into the external environment in the second wavelength range; - determination of an interaction between the person and an interactive light element from said at least one interactive light element of the first light pattern, based on at least one descriptive information of the relative position between the first light pattern and the person; - implementation of a specific action based on the interactive light element for which the interaction with the person is determined. Thus, the invention allows a motor vehicle to interact with a person outside the vehicle by controlling the first pixelated light beam projected to form a first light pattern comprising one or more interactive light elements. The vehicle then implements an action based on the person's interaction with an interactive light element. The functionalities associated with the use of light modules in a vehicle are thus considerably enhanced. Furthermore, the accuracy of determining the person's interaction with the first light pattern is improved because it is based on depth information obtained by projecting a second light pattern in a range of non-visible wavelengths, for example, in the infrared range.

[0009] According to some embodiments, the determination of the first light pattern to be projected can be a function of the position of said person.

[0010] Thus, the accuracy associated with determining the first projected light pattern is improved. Furthermore, the interaction of the person with the interactive light element of the first light pattern is facilitated.

[0011] According to some embodiments, at least one descriptive data point of the relative position may include: - a first image acquired by at least one camera on the vehicle in the first wavelength range and representative of the projection of the first light pattern into the external environment; and - the depth information of the external environment determined from the representative image of the projection of the second light pattern into the external environment in the second wavelength range, called the second image.

[0012] Thus, the joint consideration of the first image representing the projection of the first light pattern, and its interactive elements, and of the depth information derived from the projection of the second light pattern in the non-visible domain, allows a precise determination of the interaction between the person and the first projected light pattern.

[0013] According to embodiments, the action implemented may include a modification of the first light pattern by controlling at least one light module to form the first modified light pattern in the first projected pixelated light beam.

[0014] Thus, the first projected light pattern can evolve dynamically according to the interactions of the person outside the vehicle. This makes it possible to implement a dynamic interactive service, such as a scenario for welcoming a vehicle user, for example.

[0015] In addition, the modification of the first light pattern may include: - a modification of a shape and / or a color of the interactive light element for which the interaction with the person is determined; and - a removal of the interactive light element for which the interaction with the person is determined.

[0016] Thus, the person interacting with the first light pattern has visual feedback indicating that their interactions have been taken into account by the vehicle, which reinforces the precision associated with the interactive service.

[0017] According to embodiments, the action implemented may include adapting a vehicle driving parameter according to the interactive light element for which the interaction with the person is determined. Thus, the interaction of the person outside the vehicle can be used to adapt a driving parameter of the vehicle, which strengthens the safety associated with driving the vehicle, particularly in the case where the motor vehicle is driven autonomously.

[0018] According to some embodiments, the light pattern may include several interactive light elements.

[0019] Thus, a wide variety of interactive services can be implemented from the invention.

[0020] In addition, the first light pattern may include a projected light element indicating a question, and N interactive light elements corresponding to respective answers, N being an integer greater than or equal to 2.

[0021] Such an embodiment makes it possible to establish interactive communication between the vehicle and the person located outside the vehicle.

[0022] According to embodiments, the second light pattern includes dark areas and illuminated areas in the second wavelength range.

[0023] Such a second luminous pattern presenting a plurality of strongly contrasting zones makes it possible to clearly identify the distortion of the second luminous pattern by the scene, and therefore to accurately determine the depth information of the scene.

[0024] In addition, the dark areas and the lit areas may have identical shapes and sizes and may form a checkerboard pattern.

[0025] Such a second light pattern allows for precise determination of depth information of the external environment, while being associated with simple control of at least one light module to project the second pixelated light beam according to the second light pattern.

[0026] According to some embodiments, the second light pattern is a unique predetermined light pattern.

[0027] Thus, the determination of the second light pattern is simplified.

[0028] According to a first variant, the second light pattern can be selected from a set of predetermined second patterns, based on information about the external environment of the vehicle and / or representative of a vehicle driving situation.

[0029] Thus, it is made possible to adapt the second light pattern to the environment or driving situation, while allowing a simple determination of the second light pattern.

[0030] According to a second variant, the second light pattern can be generated from information on the external environment of the vehicle and / or representative of a vehicle driving situation.

[0031] Thus, the second light pattern can be optimized according to the external environment and / or the driving situation, to improve the accuracy of the depth information determined from the projection of the second light pattern into the environment outside the vehicle.

[0032] In addition, the generation of the second light pattern includes applying a first light intensity map generation model to at least one representative image of the external environment to obtain a light intensity map corresponding to the second light pattern, the light intensity map indicating light intensity values ​​to control light elements of a pixelated light source of a vehicle light module capable of projecting the second pixelated light beam in the second wavelength range.

[0033] Thus, a first model is capable of directly generating a light intensity map that facilitates the control of at least one light module to project the second light pattern. Furthermore, depending on the resolution of the second pixelated light beam, it is possible to optimize the second light pattern according to the external environment.

[0034] According to embodiments, depth information can be obtained by applying a second depth information determination model to a representative image of the projection of the second light pattern into the external environment in the second wavelength range.

[0035] Indeed, the distortion induced by the environment in the second projected light pattern makes it possible to determine depth information about the environment. Such a second model can be trained by machine learning, which improves the accuracy associated with determining depth information.

[0036] A second aspect of the invention relates to a computer program comprising instructions for implementing the method according to the first aspect of the invention, when these instructions are executed by a processor.

[0037] A third aspect of the invention relates to a control module for at least one lighting module for a motor vehicle, the at least one lighting module comprising at least one pixelated light source and being capable of projecting a first pixelated light beam in a first wavelength range and of projecting a second pixelated light beam in a second wavelength range, the first wavelength range comprising visible wavelengths and the second wavelength range comprising non-visible wavelengths, the control module comprising a processor configured to: - upon obtaining information indicating the presence of a person in an environment outside the motor vehicle, determine a first light pattern to be projected into the outside environment in the first wavelength range and a second light pattern to be projected into the outside environment in the second wavelength range in a common projection area with the first light pattern; - control at least one light module to form the first light pattern in the first projected pixelated light beam and to form the second light pattern in the second pixelated light beam. The first light pattern includes at least one interactive light element, and the processor is further configured to: - obtain at least one descriptive information of a relative position between the first projected light pattern and the person, said at least one descriptive information including depth information of the external environment determined from a representative image of the projection of the second light pattern into the external environment in the second wavelength range; - determine an interaction between the person and an interactive light element from said at least one interactive light element of the first light pattern, based on at least one descriptive information of the relative position between the first light pattern and the person; - to implement a specific action based on the interactive light element for which the interaction with the person is determined.

[0038] Other features and advantages of the invention will become apparent from an examination of the detailed description below, and the accompanying drawings in which:

[0039] [Fig-1] illustrates a vehicle according to embodiments of the invention;

[0040] [Fig.2a] illustrates a first light module of a vehicle capable of projecting at least a first pixelated light beam in a first range of visible wavelengths, according to embodiments of the invention;

[0041] [Fig.2b] illustrates a second light module of a vehicle capable of projecting a second pixelated light beam in a second range of non-visible wavelengths, according to embodiments of the invention;

[0042] [Fig.3] illustrates a method of controlling at least one light module of a vehicle to interact with a person outside the vehicle, according to embodiments of the invention;

[0043] [Fig.4a] illustrates a projection of a first visible light pattern in a first pixelated light beam to interact with a person outside the vehicle, in a first range of wavelengths, according to a first embodiment of the invention;

[0044] [Fig.4b] illustrates a projection of a second non-visible light pattern into a second pixelated light beam to interact with a person outside the vehicle, in a second range of wavelengths, according to the first embodiment of the invention;

[0045] [Fig.5a] illustrates a projection of a first visible light pattern in a first pixelated light beam to interact with a person outside the vehicle, in the first wavelength range, according to a second embodiment of the invention;

[0046] [Fig.5b] illustrates a projection of a second non-visible light pattern in a second pixelated light beam to interact with a person outside the vehicle, in the second wavelength range, according to the second embodiment of the invention;

[0047] [Fig.6a] illustrates a second non-visible light pattern projected by a lighting module in an environment to a vehicle, according to embodiments of the invention;

[0048] [Fig.6b] illustrates an image acquired by a camera following the projection of a second luminous pattern not visible in an environment outside the vehicle, according to embodiments of the invention;

[0049] [Fig.7] illustrates the structure of a control module for at least one light module according to embodiments of the invention;

[0050] [Fig.8a] illustrates a first example of a pixelated light source of a first light module capable of projecting two pixelated light beams in two distinct wavelength ranges, according to embodiments;

[0051] [Fig.8b] illustrates a second example of a pixelated light source of a first light module capable of projecting two pixelated light beams in two distinct wavelength ranges, according to embodiments;

[0052] [Fig.9] is a training system for a first model of map generation luminous intensity, according to embodiments of the invention;

[0053] [Fig. 10] is a diagram illustrating the steps of a training phase of a first model for generating a light intensity map, according to embodiments of the invention.

[0054] The description focuses on the features that distinguish the vehicle, control module and method from those known in the state of the art.

[0055] Fig. 1 illustrates a vehicle 100 according to embodiments of the invention.

[0056] The vehicle 100 according to the invention comprises at least a first light module 120.1 comprising a first matrix light source capable of forming at least a first visible pixelated light beam projected outwards from the vehicle, for example onto the ground outside the vehicle, in a first range of wavelengths comprising wavelengths in the visible range, therefore comprising wavelengths between 400 and 800 nanometers.

[0057] The vehicle 100 according to the invention may include a second light module 120.2 comprising a second matrix light source capable of forming a second invisible pixelated light beam projected outwards from the vehicle, in the same projection area as the first pixelated light beam, in a second wavelength range.

[0058] According to the invention, the second wavelength range is outside the visible range and therefore does not include any wavelengths within the visible range. No restrictions are attached to the second wavelength range according to the invention, which may be, for example, an infrared or ultraviolet range. By way of example, the second wavelength range may be within the near-infrared (NIR) range, the short-wave infrared (SWIR) range, the medium-wave infrared (MWIR) range, or the long-wave infrared (LWIR) range. It should be noted that the MWIR and LWIR ranges are also referred to as the thermal range.

[0059] In what follows, by way of illustration, and unless otherwise stated, the second range of wavelengths is considered to be in the infrared range.

[0060] According to an embodiment described below, the first light module 120.1 comprises a pixelated light source capable of forming the first pixelated light beam in the first wavelength range and also of forming the second pixelated light beam in the second wavelength range. Alternatively, the first light module 120.1 comprises a first pixelated light source capable of forming the first pixelated light beam in the first wavelength range and a second pixelated light source capable of forming the second pixelated light beam in the second wavelength range. In such embodiments, the second light module 120.2 is optional, and the vehicle 100 may comprise only the first light module 120.1.

[0061] In the example shown in [Fig. 1], the first light module 120.1 and the second light module 120.2 can be integrated into a lighting device 110 such as a front headlight of the vehicle. The lighting device 110 may include at least one lighting module, in addition to the first light module 120.1 and the second light module 120.2 (or the single light module 120.1), capable of performing at least one lighting function, such as a low beam function, also called LB for “Low Beam”, and / or a high beam function, also called HB for “High Beam”.

[0062] Thus, in the example shown, the lighting device 110, in which the first light module 120.1 and the second light module 120.2 are integrated, can be the right front headlight or the left front headlight. Alternatively, a first right light module 120.1 and a second right light module 120.2 are arranged in the front right projector and a first left light module 120.1 and a second left light module 120.2 are arranged in the front left projector of vehicle 100.

[0063] In an unrepresented variant, the first light module 120.1 and the second light module 120.2 are not integrated into a front projector, but can be integrated into a rear light, a rearview mirror, a bodywork element of the vehicle, a bumper, or any other external component of the vehicle 100.

[0064] The vehicle 100 according to the invention further comprises at least one first camera 140.1 comprising at least one sensor capable of acquiring an image, or a series of images, of the external environment in which the first light module 120.1 and the second light module 120.2 are capable of projecting the first and second pixelated light beams. Thus, when the first and second light modules 120.1 and 120.2 are mounted in the front headlight of the vehicle 100, the first and second light modules 120.1 and 120.2 project the first and second pixelated light beams into a scene facing the vehicle: in this case, at least one camera 140.1 is arranged in the vehicle 100 so as to acquire an image, or a series of images, representative of the environment external to the vehicle. For example, as illustrated in [Fig.1], the first camera 140.1 can be arranged in a central position at the top of the windshield of vehicle 100.

[0065] More generally, at least one first camera 140.1 is capable of acquiring images of a scene in which the first and second light modules are capable of projecting the first and second pixelated light beams.

[0066] According to some embodiments, the first camera 140.1 is capable of acquiring images in the first wavelength range, and the vehicle 100 further comprises a second camera 140.2 capable of acquiring images in the second wavelength range. The first camera 140.1 may be an RGB camera, for “Red Green Blue” in English, and the second camera 140.2 may be an infrared camera (in the example where the second wavelength range is an infrared range).

[0067] According to some embodiments, the first camera 140.1 may comprise several sensors, including a first sensor capable of acquiring an image, or a series of images, in the first wavelength range, and a second sensor capable of acquiring an image, or a series of images, in the second wavelength range. In such alternatives, the second camera 140.2 is optional.

[0068] The vehicle 100 further comprises a control device 130 capable of sending control signals to at least one light module of the vehicle 100, for example to the first light module 120.1 and the second light module 120.2 shown on [Fig. 1]. The control signals sent to the first light module 120.1 can be determined based on the images acquired by at least one first camera 140.1. As indicated below, the control device 130 can be suitable for implementing the method according to the invention, illustrated with reference to [Fig. 3].

[0069] According to the invention, the vehicle 100 may include an image processing module 150 capable of processing images in the second wavelength range, acquired by the second camera 140.2 (or by the first camera 140.1 when it includes several sensors), before transmission of the processed images in the second wavelength range to the control device 130 and / or of the result of the processing to the control device 130. The image processing module 150 may also be capable of processing, in addition to the images in the second wavelength range, the images in the first wavelength range acquired by the first camera 140.1.In particular, as described below, the image processing module 150 is capable of determining depth information, specifically a depth map, from an image acquired in the second wavelength range, and optionally based on a second projected light pattern indicated by the control module described below. To this end, the processing module 150 can execute a second depth information determination model capable of generating depth information about the environment from an image in the second wavelength range representative of the environment into which a second light pattern is projected, as described below.

[0070] The control device 130 can determine the control signals of the first light module 120.1 (and optionally of the second light module 120.2) based on the images directly from the camera 140 and / or the processed images, and / or the processing results from the processing module 150. Alternatively, the image processing module 150 is not a separate module, and can be integrated as an internal module in the first camera 140.1 (if it includes two sensors) or in the second camera 140.2, or as an internal module in the control device 130.

[0071] The control device 130 may be a centralized vehicle control device, called a “Body Controller,” capable of implementing a plurality of vehicle functions, including the function of controlling at least one light module, in particular the first light module 120.1 and the second light module 120.2, to interact with a user outside the vehicle. Alternatively, the control device 130 is dedicated solely to generating control signals to operate the first light module 120.1 and the second light module 120.2, for interaction with a user outside the vehicle.

[0072] Fig. 2a illustrates the structure of a first light module 120.1 with a first matrix light source 200.1, according to embodiments of the invention.

[0073] The first light module 120.1 comprises: - a first control unit 201.1, also called the pilot unit or “driver” in English, of the first matrix light source 200.1; - the first matrix light source 200.1 capable of emitting light at least in the first wavelength range; - a first projection optic 202.1 for the light from the first matrix light source 200.1 to project a first pixelated light beam outwards from the vehicle, in particular to project the first pixelated light beam onto the ground in a vehicle environment 100, in the first wavelength range. No restrictions are attached to the first projection optic 202.1, which may comprise any set of optical elements.

[0074] The first matrix light source 200.1 comprises a plurality of first light elements 210.1 individually controllable by the first control unit 201.1. The first control unit 201.1 can thus individually control the first light elements 210.1, and can thus control a first light pattern projected in the first pixelated light beam, in the first wavelength range, by activating some first light elements 210.1 and deactivating other first light elements 210.1.

[0075] No restriction is attached to the number of first light elements 210.1 of the first matrix light source 200.1. Preferably, the first matrix light source 200.1 is a high-definition light source, that is, one that allows the projection of a first light beam comprising more than one hundred pixels, preferably more than 1000 pixels. The first matrix light source 200.1 can allow the projection of a first light beam comprising more than 10,000 pixels, in the first wavelength range, according to embodiments of the invention.

[0076] The first luminous elements 210.1 can be electroluminescent.

[0077] For example, each first luminous element 210.1 can be a light-emitting diode of the LED type, for “Light Emitting Diode”, the plurality of LEDs forming a matrix network of LEDs.

[0078] According to some embodiments, the first matrix light source 200.1 can be monolithic. A so-called "monolithic" source can have a particularly high density of first light elements 210.1, which makes it especially attractive for a plurality of applications. A monolithic source implies a plurality of first semiconductor elements Electroluminescent 210.1 elements of submillimeter dimensions are epitaxially mounted directly onto a common substrate, which is generally silicon. In contrast to sources comprising conventional LED arrays, where each individual light-emitting element is a uniquely produced electronic component mounted on a substrate such as a printed circuit board (PCB), a monolithic source is considered a single electronic component. During its production, multiple areas of electroluminescent semiconductor junctions are generated on a common substrate, forming an array. This production technique allows for the creation of closely spaced electroluminescent areas, each acting as an individual light-emitting element. The gaps between the light-emitting elements can be submillimeter in size.One advantage of this production technique is the high level of pixel density that can be achieved on a single substrate.

[0079] The first individual light elements can be individually controlled by the first control unit 201.1, which is capable of receiving control signals from the control device 130, notably in the form of a first digital image or first light intensity map. Individual control can include controlling the power supply provided to the first individual light element by pulse-width modulation, or PWM. The first control unit 201.1 can, for example, be an ASIC (Application-Specific Integrated Circuit), which has the advantage of being very compact and can be integrated into the monolithic light source.

[0080] Alternatively, the first matrix light source 200.1 comprises a light source and a first matrix 200.1 of micromirrors, each micromirror thus forming a first light element 210.1 by reflection of the light rays from the light source. The first matrix of micromirrors 210.1 is also called DMD for Digital Micromirror Devices, and the micromirrors 210.1 can be individually activated to reflect the light from the light source towards the first projection optic 202.1, thus forming the first pixelated light beam, the first pattern of which is controllable by controlling the micromirrors 210.1.

[0081] Fig. 2b illustrates the structure of a second light module 120.2 with a second matrix light source 200.2, according to embodiments of the invention.

[0082] The second light module 120.2 comprises: - a second control unit 201.2, also called the pilot unit or “driver” in English, of the second matrix light source 200.2; - the second matrix light source 200.2 capable of emitting light in the second wavelength range; - a second projection optic 202.2 for the light from the second matrix light source 200.2 to project a second pixelated light beam outwards from the vehicle, in particular to project the second pixelated light beam onto the ground in the vehicle environment 100, in the second wavelength range. No restrictions are attached to the second projection optic 202.2, which can comprise any set of optical elements.

[0083] The second matrix light source 200.2 comprises a plurality of second light elements 210.2 individually controllable by the second control unit 201.2. The second control unit 201.2 can thus individually control the second light elements 210.2.2, and can thus control a second light pattern projected in the second pixelated light beam, in the second wavelength range, by activating some second light elements 210.2 and deactivating other second light elements 210.2. .

[0084] No restriction is attached to the number of second light elements 210.2 of the second matrix light source 200.2. Preferably, the second matrix light source 200.2 is a high-definition light source, that is, one that allows the projection of a second light beam comprising more than one hundred pixels, preferably more than 1000 pixels. The second matrix light source 200.2 can allow the projection of a second light beam comprising more than 10,000 pixels, in the second wavelength range, according to embodiments of the invention.

[0085] The second luminous elements 210.2 can be electroluminescent.

[0086] For example, each second luminous element 210.2 can be a light-emitting diode of the LED type, for “Light Emitting Diode”, the plurality of LEDs forming a matrix network of LEDs.

[0087] According to some embodiments, the second matrix light source 200.2 can be monolithic. As explained previously, a so-called "monolithic" source can have a particularly high density of second light elements 210.2, making it especially attractive for a wide range of applications. A monolithic source involves a plurality of submillimeter-sized electroluminescent semiconductor second elements 210.2, epitaxially bonded directly onto a common substrate, the substrate generally being made of silicon. Unlike sources comprising conventional LED arrays, in which each elementary second light element is an individually produced electronic component mounted on a substrate such as a printed circuit board (PCB), a monolithic source is to be considered as a component A unique electronic process in which multiple areas of light-emitting semiconductor junctions are generated on a common substrate, forming an array. This production technique allows for the creation of closely spaced light-emitting areas, each acting as a primary light-emitting element. The gaps between the light-emitting elements can be submillimeter in size. One advantage of this production technique is the high pixel density that can be achieved on a single substrate.

[0088] The second individual light elements can be individually controlled by the second control unit 201.2, which is capable of receiving control signals from the control device 130, notably in the form of a second digital image or second light intensity map. Individual control can include controlling the power supply provided to the second individual light element by pulse-width modulation, or PWM. The second control unit 201.2 can, for example, be an ASIC (Application-Specific Integrated Circuit), which has the advantage of being very compact and can be integrated into the monolithic light source.

[0089] Alternatively, the second matrix light source 200.2 comprises a light source and a second micromirror array, each micromirror thus forming a second light element 210.2 by reflecting the light rays from the light source. The second micromirror array 210.2 is also called DMD for Digital Micromirror Devices, and the micromirrors 210.2 can be individually activated to reflect the light from the light source towards the second projection optic 202.2, thus forming the second pixelated light beam, the second pattern of which is controllable by controlling the micromirrors 210.2.

[0090] As previously stated, and as illustrated with reference to Figures 8a and 8b described below, the first light module 120.1 may be capable of forming the first pixelated light beam in the first wavelength range and of forming the second pixelated light beam in the second wavelength range, in which case the vehicle 100 may not include the second light module 120.2.

[0091] Fig. 3 shows the steps of a method for controlling at least one light module (the first light module 120.1, or the first and second light modules 120.1 and 120.2), to interact with a person located outside the vehicle 100, according to embodiments of the invention.

[0092] The method for controlling at least one light module is implemented by a control module which may be the control device 130 capable of transmitting control signals to light modules 120.1 and 120.2, so that the first control unit 201.1 drives the first matrix source 200.1 and so that the second control unit 201.2 drives the second matrix source 200.2, according to the control signals.

[0093] In one variant, the vehicle 100 comprises only the first light module 120.1 which is capable of forming the first light beam and the second light beam, and the control module implementing the invention is the control device 130 or the first control unit 201.1 capable of driving the first matrix source 200.1.

[0094] When the control module is the first control unit 201.1, the control device 130 does not transmit control signals to the first light module 120.1, and the first light module 120.1 determines a first light pattern to be projected based on images acquired by at least one camera and / or based on the outputs of the analysis module 150, and determines a second light pattern as described below (which may be predetermined or dependent on descriptive information about the driving situation or the vehicle environment). Thus, in this embodiment, the control device 130 is optional.

[0095] At a step 300, the control module obtains at least one piece of information indicating the presence of a person in an environment of the vehicle.

[0096] The at least one piece of information indicating the presence of a person may include one or more of the following data: - one or more images acquired by at least one camera in the first wavelength range, for example by the first camera 140.1; - one or more images acquired by at least one camera in the second wavelength range, for example by the second camera 140.2 (or by the first camera 140.1 when it includes several sensors); - an image processed by the processing module 150 and / or a result of the processing from the processing module 150. The result of the processing from the processing module 150 may, for example, indicate whether or not a person is present in the environment of the vehicle, and may also include an identification of the person and / or position data of the person; - a signal received from a vehicle communication interface, not shown in [Fig.1], for example a signal received from a vehicle user key following user input on the key (pressing a button on the key, for example to unlock the vehicle); - information captured by another sensor on the vehicle, for example a proximity sensor, a radar or a lidar.

[0097] Thus, preferably, at least one piece of data relating to the presence of the person indicates: - whether or not a person is detected in the vicinity of the vehicle; - optionally, location data of the detected person; - optionally, identification of the detected person.

[0098] At a step 301, the control module checks whether at least one activation condition of the light module control is met, based on at least one data relating to the presence of the person obtained at step 300.

[0099] If each activation condition is met, then the process proceeds to step 302. Otherwise, if at least one activation condition is not met, the process returns to step 300 until at least one new piece of data relating to the presence of the person is received.

[0100] At least one activation condition includes at least one condition on the presence of the person in the vehicle's environment. Thus, if no person is present in the vehicle's environment, at least one activation condition is not met and the process returns to step 300.

[0101] Another activation condition may be, in certain embodiments, that the identification of the detected person corresponds to a user or owner of the vehicle 100.

[0102] Alternatively, another activation condition may be that the detected person is located at the edge of a road on which the vehicle 100 is traveling. To check such a condition, the processing module 150 can determine a relative position between the road and the detected user in a scene facing the vehicle, from the images captured by at least one camera, in the first wavelength range and / or in the second wavelength range, and can transmit to the control module the relative position between the road and the detected person.

[0103] The vehicle's environment can be defined by a set of positions outside the vehicle, for example, a set of positions located less than M meters from the vehicle. In some embodiments, M may be equal to 20 or 30 meters. In other embodiments, M may be on the order of several hundred meters, for example, 200 or 300 meters.

[0104] As explained previously, in the case where each activation condition is verified in step 301, the process proceeds to step 302.

[0105] In step 302, the control module determines a first light pattern to be projected in the first wavelength range and obtains, or determines, a second light pattern to be projected in the second wavelength range.

[0106] The first light pattern to be projected can be predetermined, or can be determined according to the position of the detected person, as illustrated in particular with reference to figures 4a and 5a described below.

[0107] No restrictions are attached to the way in which the second light pattern is determined or obtained by the control module. The second light pattern can be: - predetermined, i.e. it is a single, predetermined second light pattern, which is fixed; - selected from a set of predetermined second light patterns, the selection being made based on information about the vehicle's environment and / or describing the driving situation. For example, the second light pattern is selected from the set based on the vehicle's current speed; - determined by the control module based on information about the vehicle's environment and / or describing the driving situation. For example, the light pattern can be determined based on objects in the scene identified from images acquired by at least one camera, in the first wavelength range or in the second wavelength range, in particular by applying the first light intensity map generation model described below.

[0108] As described below, particularly with reference to Figures 6a and 6b, the second light pattern is a discontinuous pattern, composed for example of dark areas and lit areas, the distortions induced by the environment in the second light pattern, particularly at the discontinuities which are the boundaries between dark areas and lit areas, allowing the determination of depth information by the processing module 150, as described below.

[0109] At a step 303.1, the control module commands the first matrix light source 200.1 according to the first determined light pattern, so as to form the first light pattern in the pixelated beam projected by the first light module 120.1. When the control module is the control device 130, the step 303.1 includes the transmission of control signals indicating the first determined light pattern, to the first control unit 201.1, for example in the form of a first light intensity map.

[0110] The first light pattern is thus projected into the first pixelated light beam, for example onto the ground in the environment of the vehicle, in the first range of wavelengths which is a visible range.

[0111] Fig. 4a illustrates a projection of a first pattern in a first pixelated light beam to interact with a person 410 outside the vehicle, in the first wavelength range, according to a first embodiment of the invention.

[0112] In the first embodiment, the person 410 is detected (during step 301) near a road 420 on which the vehicle 100 is traveling, in front of the vehicle 100. In this embodiment, the environment of the vehicle can be defined by positions distant from the vehicle, for example more than 50 meters from the vehicle, but less than several hundred meters, so that the person is detectable, in particular from images from at least one camera, in the first wavelength range or in the second wavelength range.

[0113] In the first embodiment, the first light pattern 400 is projected by at least one first light module 120.1 arranged in at least one lighting device among the right front lighting device 110.1 and the left front lighting device 110.2 of the vehicle 100. By way of illustration in the example of [Fig.4a], the first light module 120.1 projecting the first light pattern 400 is integrated into the right front lighting device 110.2 of the vehicle 100.

[0114] The first light pattern 400 can be projected into an area near the person 410 (for example at a distance of less than 5 meters from the person 410), so as to allow interaction between the person 410 and the first projected light pattern 400.

[0115] The first light pattern 400 may include a projected light element 401 displaying information for the attention of the person 410. The information may, for example, be a question to ask the person 400 whether or not they wish to cross the road 420.

[0116] The first light pattern 400 according to the first embodiment further comprises at least one interactive light element. The first light pattern 400 may, in particular, comprise N interactive light elements, N being greater than or equal to 2. In the example of [Fig. 4a], the first light pattern 400 comprises a first interactive light element 402.1 and a second interactive light element 402.2.

[0117] In some embodiments, each of the N interactive light elements corresponds to an answer to a question displayed by the projected light element 401. For example, the first interactive light element 402.1 can display a “yes” answer (indicating that person 410 wants to cross road 420) and a “no” answer (indicating that person 410 does not want to cross road 420).

[0118] No restriction is attached to the number of interactive light elements included in the first light pattern 400.

[0119] According to the invention, the person 410 can interact with an interactive light element from among the first interactive light element 402.1 and the second interactive light element 402.2, and such interaction can be detected by the control module from descriptive information of the relative position between the first motif luminous element 400 and person 410, as described below. Following the detection of the interaction, the control module can implement an action depending on the interactive luminous element with which the interaction is detected.

[0120] Thus, the first embodiment can be implemented during a period of vehicle 100 driving.

[0121] Fig. 5a illustrates a projection of a first light pattern 500 into a first pixelated light beam to interact with a person 510 outside the vehicle 100, in the first wavelength range, according to a second embodiment of the invention.

[0122] The first light pattern 500 according to the second embodiment, can be projected by the first light module 120.1 integrated into the right front lighting device 110.1 or into the left front lighting device 110.2. As in [Fig.4a], [Fig.5a] illustrates an example in which the first light module 120.1 is integrated into the right front lighting device 110.1.

[0123] In the second embodiment, the person 510 can be a user of the vehicle 100, for example the owner of the vehicle. In this second embodiment, the owner can be detected at step 301 by the identification included in the data obtained at step 300, which can be a signal from the vehicle user's key and / or which can be an image processing result from the processing module 150 indicating that the vehicle user is identified, at a given position, in images acquired by at least one camera, in the first wavelength range and / or in the second wavelength range (the image processing module 150 can implement for this purpose a facial recognition algorithm for the vehicle user 100).

[0124] Thus, in the second embodiment, the vehicle environment corresponds to a set of positions at a distance of less than M meters, M being between 10 meters and 30 meters, for example equal to 20 meters.

[0125] The first light pattern 500 according to the second embodiment may include at least one interactive light element. For example, as shown in [Fig. 5a], the first light pattern 500 may include a plurality of interactive light elements 501.1 to 501.4. However, there is no restriction on the number of interactive light elements in the first light pattern 500 according to the second embodiment.

[0126] As shown in [Fig. 5a], the interactive light elements can be identical and can form a path between the person 510 outside the vehicle 100 and the vehicle 100. The user 510 can thus interact sequentially with each of the interactive light elements 501.1 to 501.4 (for example, by walking on each of the interactive light elements 501.1 to 501.4), starting with the interactive light element closest to user 510, namely the interactive light element 501.4.

[0127] Thus, the second embodiment can be implemented during a vehicle parking period 100.

[0128] Referring again to [Fig.3], at a step 303.2, implemented in parallel with step 303.1, the control module commands the second matrix light source 200.2 (or the first matrix light source 200.1 when it is capable of forming the first light beam and the second light beam) according to the second determined light pattern, so as to form the second light pattern in the second pixelated beam projected by the second light module 120.2 (or by the first light module 120.1 when it is capable of projecting the first pixelated light beam and the second pixelated light beam), in the second wavelength range (not visible).

[0129] When the control module is the control device 130, step 303.2 includes the transmission of control signals indicating the second light pattern obtained in step 302, to the second control unit 201.2 (if the second light module 120.2 is capable of projecting the second pixelated light beam) or to the first control unit 201.1 (if the first light module 120.1 is capable of projecting the first and second pixelated light beams).

[0130] The second light pattern is thus projected into the second wavelength range at the same time as the first light pattern in the first wavelength range, in a common projection area.

[0131] [Fig.4b] shows a projection of a second light pattern 430 (not visible) in the second pixelated light beam, in the second wavelength range, according to the first embodiment of the invention described above with reference to [Fig.4a].

[0132] Thus, [Fig. 4b] illustrates the second light pattern 430 projected in the second wavelength range, simultaneously (over a common time period) with the first light pattern 400 projected in the first wavelength range illustrated in [Fig. 4a]. Since the second light pattern 430 is projected into a real environment, its projection is distorted by the real environment within the projection area. The reference numeral 430 thus more accurately designates the distortion of the projected light pattern induced by the environment within the projection area, for example, by the foot of the detected person if it touches the first light pattern 430.

[0133] As illustrated in Figures 4a and 4b, the first light pattern 400 and the second light pattern 430 are projected into the same projection area. The second light pattern 430 can, according to the illustrated embodiment, be a checkerboard of dark and light areas, as described below with reference to figures 6a and 6b.

[0134] [Fig.5b] shows a projection of a second light pattern 530 (not visible) in the second pixelated light beam, in the second wavelength range, according to the second embodiment of the invention described previously with reference to [Fig.5a].

[0135] As illustrated in Figures 5a and 5b, the first light pattern 500 and the second light pattern 530 are projected into the same projection area. The second light pattern can, according to the illustrated embodiment, be a checkerboard with dark and light areas, as described below with reference to Figures 6a and 6b. As with the second projected light pattern 430, since the second light pattern 530 is projected into an environment, its projection is distorted by the real environment, for example, by the foot of the detected person if it touches the first light pattern 530. The reference numeral 530 thus more precisely designates the distortion of the first projected light pattern induced by the scene in [Fig. 5b].

[0136] Referring again to [Fig.3], to a step 304.1 and to a step 304.2, the control module obtains descriptive data of a relative position between the detected person and the first projected light pattern.

[0137] According to the invention, the descriptive data obtained include: - a first image acquired by the first camera 140.1 in the first wavelength range and representative of the projection of the first light pattern in the scene following step 303.1. Such a first image is obtained in step 304.1. According to some embodiments, the first image can be processed by the processing module 150 before being obtained by the control module in step 304.1; and - depth information, such as a depth map, determined by the processing module 150 based on a second image acquired in the second wavelength range and representative of the projection of the second light pattern, following step 303.2.

[0138] Indeed, according to the invention, the projection of the second light pattern into a scene, after acquisition of a second image representative of the external environment in the second wavelength range, makes it possible to determine depth information from the deformations induced by the external environment in the second projected light pattern, as explained with reference to figures 6a and 6b.

[0139] Fig. 6a presents a second light pattern 620 determined by the control module during step 302 described above, according to one embodiment.

[0140] Figure 6a shows in particular the result of a projection of the second light beam according to the second light pattern 620 in a plane perpendicular to a principal beam direction, that is, without distortion induced by a real environment external to the vehicle. Such a distorted projection corresponds to what is shown in [Fig. 6b], and also corresponds to the second light patterns 430 and 530 projected in real scenes described previously.

[0141] The second light pattern 620 is projected by the second light module 120.2 (or the first light module 120.1 if it is capable of projecting the first and second pixelated light beams), in the second wavelength range and in the same projection area as the first light pattern projected in the first wavelength range.

[0142] In the embodiment shown in [Fig.6a], the second light pattern 620 is a checkerboard, showing a regular alternation of dark areas 601 and lit areas 602, with a strong contrast between these areas.

[0143] Each zone 601 or 602 corresponds to a set of at least one pixel, and preferably to a plurality of pixels, for example several tens or hundreds of pixels.

[0144] Thus, a dark area in the second pixelated light beam is created by deactivating the pixels in the area, while an illuminated area is obtained by activating at least some of the pixels in the illuminated area. The activation or deactivation of the pixels is implemented by the second control unit 201.2 (when the second light module 120.2 projects the second pixelated light beam) or by the first control unit 201.1 (when the first light module 120.1 projects the first and second pixelated light beams).

[0145] In the case where the pixelated light source emitting in the second wavelength range is a high-definition source with electroluminescent elements, the control unit 201.1 or 201.2 performs the second light pattern by applying a voltage to the light elements corresponding to the pixels of the illuminated areas 602, and does not apply any voltage to the light elements corresponding to the pixels of the dark areas 601.

[0146] A high-definition pixelated light source thus allows the projection of a second light pattern with a large number of dark and lit areas, which then allows high accuracy in determining depth information implemented by the processing module 150, as described below.

[0147] The checkerboard shown in [Fig. 6a] is given for illustrative purposes. It comprises 9 columns and 8 rows of zones, each zone comprising a plurality of pixels. No restrictions are attached to the shape of the zones, nor to the distribution of light and dark zones. However, a regular checkerboard comprising a large number of zones, for example more than 100 zones, allows in practice a precise determination of a depth map by processing module 150. The shape and size of dark and lit areas may vary, especially when the second light pattern is selected from several or generated by the first light intensity map generation pattern described below.

[0148] Fig. 6b presents a second image 610 acquired, in the second wavelength range, by the second camera 140.2 (or by the first camera 140.1 if it includes two sensors), representative of the projection of the second light pattern 620 in the environment outside the vehicle 100, according to embodiments of the invention.

[0149] The external environment in [Fig. 6b] is intentionally simplified, with few objects, to simplify understanding of the invention. The invention is applicable in other situations, particularly those described above with reference to Figures 4b and 5b.

[0150] Thus, in what follows, the second image 610 can be replaced by the projections 430 and 530 of the second light patterns in the outdoor environments shown in Figures 4b and 5b.

[0151] In [Fig. 6b], the external environment into which the second light pattern 620 is projected includes another vehicle, as well as a vertical wall located behind the other vehicle. The external environment can thus correspond to the interior of a parking lot, a situation given for illustrative purposes only.

[0152] As previously explained, the second light pattern is projected into the external environment in the second wavelength range, and the image 610 captured by the first or second camera is thus a second image in the second wavelength range.

[0153] It can be observed in [Fig.6b]: - that the dark and lit areas 611 projected onto a substantially horizontal plane such as the ground have an elongated shape and are thus lengthened; - that the dark and lit areas 612 projected onto a substantially vertical and close plane, such as the rear of the other vehicle, retain a square format and have a first given size; - that the dark and lit areas 613 projected onto a substantially vertical plane and further away than the areas 612, such as the wall behind the other vehicle, also retain a square format but have a second size greater than the first size; - that the dark and lit areas 614 projected onto irregular objects, with curves, such as the top of the other vehicle, are strongly distorted.

[0154] Thus, the projection of the second light pattern into an external environment provides, through analysis of the deformation of the second projected light pattern and after processing the second image, information on environmental disparity. exterior that allow for the estimation of depth information of the external environment. Thus, a second light pattern comprising a regular repetition of contrasting sub-patterns, such as a repetition of dark and light square areas to form a checkerboard, is particularly suitable for estimating depth information of the external environment.

[0155] Such depth information can be estimated by the processing module 150 and transmitted to the control module 140 during step 304.2 described above, from the second image 610 acquired by the camera. When the person interacts with the first light pattern, they distort the second projected light pattern, which makes it possible to accurately determine the position of the limb of the person interacting with the first light pattern.

[0156] In the scenes shown in Figures 4b and 5b, such depth information allows, in correlation with the first image in the first wavelength range (representative of the 400 or 500 projection of the first light pattern), to determine precisely the relative position between the person 410 or 510, and the first light pattern, in particular with its interactive light elements.

[0157] Referring again to [Fig. 3], following steps 304.1 and 304.2 of obtaining descriptive data on the relative position between the detected person and the first light pattern, the control module determines, in a step 305, whether the detected person has interacted with an interactive element of the first projected light pattern, and, if so, determines which interactive element the person has interacted with, based on the descriptive data obtained in the preceding steps. For example, the person interacts with a given interactive light element by touching (for example, with their foot) an area on the ground into which the given interactive element is projected. The position of the foot is determined precisely from the deformation it induces in the second projected light pattern.

[0158] To implement step 305, the control module can implement a touch detection algorithm from the first image obtained in step 304.1 and depth information (e.g. the depth map) obtained in step 304.2. An example of a touch detection algorithm applicable to a series of images is described, for example, in the document “Virtual Touch Sensor Using a Depth Camera” by Dong-seok Lee and Soon-kak Kwon, Sensors (Basel), February 2019, 19(4):885.

[0159] According to this document, a depth image can be obtained (in the case of the invention, from the processing module 150 capable of determining a depth image from the second image in the second wavelength range, following the projection of the second light pattern). A region of the image to be touched is then defined: in the invention, a portion of the depth image is thus associated with each projected interactive light element identified in the first image. It can then be verified that the depth of a touch object (e.g., a person's foot) crosses a depth initially defined for each part of the image associated with a projected interactive light element in the first wavelength range. Other detection algorithms based on the first image obtained in step 304.1 and on the depth information obtained in step 304.2 can be used within the scope of the invention.

[0160] In the event of no interaction between the detected person and at least one interactive light element of the projected light pattern in step 305, the method can return to steps 304.1 and 304.2 to obtain new descriptive data of the relative position between the detected person and the first projected light pattern, or to step 300 if the descriptive data of the relative position between the detected person and the first projected light pattern indicate that the detected person is moving away from the first projected light pattern, or if the descriptive information of the relative position between the user and the first projected light pattern indicates that the person is no longer detected (for example, when the person has moved out of the field of view of at least one camera).

[0161] If the control module determines, in step 305, that the detected person has interacted with an interactive light element of the first projected light pattern, the process proceeds to a step 306 of implementing at least one action based on the interactive light element for which an interaction with the detected person is determined.

[0162] Two examples of action are described below, with reference to the two embodiments described in Figures 4a and 5a. However, other actions can be implemented within the scope of the invention, by the control module, depending on the interactive light element for which an interaction with the detected person is determined.

[0163] In the first embodiment illustrated in [Fig.4a] and 4b, the control module can determine at step 305 whether the person has interacted with the first interactive light element 402.1 or with the second interactive light element 402.2, for example by walking on an area corresponding to one or the other of the interactive light elements 402.1 and 402.2.

[0164] Depending on the interactive light element with which the person 410 has interacted, the control module (which, in the first embodiment, is preferably the centralized control device 130 of the vehicle 100) can adapt a driving parameter of the vehicle 100, for example, to accelerate, decelerate, or maintain a speed of the vehicle 100. For example, when the person 410 has touched the first If a person 410 touches the interactive light element 402.1 to indicate their intention to cross the road 420, the control device 130 can reduce the speed of the vehicle 100, or display a message to the driver instructing them to reduce the vehicle's speed, at step 307. Conversely, if a person 410 touches the second interactive light element 402.2 to indicate that they do not intend to cross the road 410, the control device 130 can maintain or accelerate the speed, or can indicate to the driver that the speed can be maintained or increased, at step 307.

[0165] In the second embodiment illustrated in Figures 5a and 5b, the control module (which can be the control device 130 or the first control unit 201.1) can determine at a first iteration of step 305 that the person 510 (the user of the vehicle 100, for example) has interacted with the light element interactive 501.4, which is closest to it (for example, by walking in an area where the interactive light element 501.4 is projected). According to the second embodiment, the control module can, at a step 308, modify the interactive light element 501.4 with which the person 510 has interacted, while leaving unchanged the interactive light element(s) with which the person has not interacted. Thus, the control module commands the first light module 120.1 to modify the interactive light element 501.4 in the first projected light pattern 500.

[0166] Such a modification of the interactive light element 501.4 may be: - a change in the color and / or shape of the interactive light element 501.4; - removal of the interactive light element 501.4. Following step 308 of modifying the first light pattern 500, the process returns to steps 304.1 and 304.2, and the control module obtains new descriptive data of a relative position between the detected user and the first light pattern 500 modified during the previous step 308. Thus, at a new iteration of step 305, the control module can determine whether the person 510 has interacted with another interactive light element other than the interactive light element 501.4 with which the person 510 interacted during the first iteration.

[0167] For example, the control module can determine that person 510 has interacted with the interactive light element 501.3 that is closest to the interactive light element 501.4 with which person 510 previously interacted. In a subsequent iteration of step 308, the control module can thus modify the light pattern 500 so as to change the shape and / or color of the interactive light element 501.3 or so as to remove the interactive light element 501.3.

[0168] Thus, an interactive path can be created between vehicle 100 and person 510, with a dynamic evolution depending on interactions with person 510. A such an interactive journey could be a welcome scenario for vehicle 100, or “welcome scenario” in English.

[0169] More generally, in all embodiments of the invention, the control module performs an action at step 306 based on an interactive light element of the first light pattern, with which the detected person interacts. Such an action may be, for example, a modification or maintenance of a vehicle parameter and / or a modification of the first light pattern projected by the light module 120.

[0170] Figure [Fig.7] illustrates the structure of a control module 700 according to embodiments of the invention.

[0171] As described previously, the control module 700 can be the first control unit 201.1 (particularly when the first light module 120.1 is capable of projecting the first pixelated light beam and the second pixelated light beam), can be the control device 130 or can be integrated into the control device 130.

[0172] The control module 700 includes a processor 701 configured to communicate unidirectionally or bidirectionally, via one or more buses or via a direct wired connection, with a memory 702 such as a Random Access Memory (RAM), a Read Only Memory (ROM), or any other type of memory (Flash, EEPROM, etc.). Alternatively, the memory 702 comprises several memories of the aforementioned types.

[0173] The memory 702 is capable of storing, permanently or temporarily, at least some of the data used and / or resulting from the implementation of the steps of the process according to the invention illustrated with reference to [Fig.3].

[0174] The processor 701 is capable of executing instructions, stored in memory 702, for the implementation of steps 300 to 308 described above. Alternatively, the processor 701 can be replaced by a microcontroller designed and configured to perform steps 300 to 308 described above.

[0175] The control module 700 includes a first interface 703 capable of obtaining data during the steps 300, 304.1 and 304.2 described above, in particular from at least one camera and / or from the processing module 150.

[0176] The control module 700 includes a second interface 604 capable of communicating: - with the first light module 120.1 to transmit the control signals during steps 303.1 and 308 (and during step 303.2 when the first light module 120.1 is capable of projecting the first and second pixelated light beams), and with the second light module 120.2 to transmit the control signals during step 303.2. Such transmission of control signals has location when the control module is the control device 130 or is integrated into the control device 130; - with the first matrix source 200.1 to control the projection of the first light pattern into the first projected pixelated light beam and to control the projection of the second light pattern into the second projected pixelated light beam (when the first light module 120.1 is capable of projecting the first and second pixelated light beams), in steps 303.1, 303.2 and 308, when the control module is the first control unit 201.1.

[0177] Note that memory 702 of the control module can store: - control signals corresponding to the second light pattern, when the second light pattern is a unique predetermined second light pattern; - control signals corresponding to predetermined second light patterns, when the second light pattern is selected by the control module from among several predetermined second light patterns, based on information about the vehicle's environment and / or describing the driving situation. For example, the second light pattern is selected based on the vehicle's current speed; - an algorithm for determining the second light pattern based on information about the vehicle's environment and / or describing the driving situation. Memory 701 can notably store initial parameters of a first model for generating a light intensity map corresponding to the second light pattern from information about the vehicle's environment and / or describing the driving situation, for example from an image acquired by at least one camera.

[0178] Such a first model can be trained by machine learning during a training phase, before its implementation in the control module, as described below with reference to Figures 9 and 10. According to embodiments, the first model is trained jointly with a second depth information determination model implemented in the processing module 150.

[0179] The second model is capable of generating depth information, in particular a depth map, from a second image in the second wavelength range representative of the projection of the second light pattern in a scene outside the vehicle.

[0180] Note that the processing module 150 can implement the second depth information determination model, including when the control module does not generate the second light pattern (but selects it from a predetermined set or obtains a predetermined unique second light pattern). Furthermore, the processing module 150 can be implemented in the module of control, in which case memory 702 can also store second descriptive parameters of the second model.

[0181] The first model and / or the second model can be trained by supervised learning, in particular jointly or separately, as described below.

[0182] No restrictions are attached to the second model for determining depth information, which can be constructed from a set of rules for determining depth information from: - prior knowledge of the second light pattern when it is predetermined or selected from a predetermined set, or - from information describing the second light pattern, received as input to the model, when the second light pattern is determined based on descriptive information about the vehicle environment and / or the driving situation, for example when the control module implements the first pattern described previously.

[0183] According to embodiments of the invention, the second depth information determination model is trained during a machine learning training phase, the machine learning enabling the training of the second parameters defining the second model, the second model having a predefined structure. The second model may, for example, have one of the following structures: - a convolutional neural network, such as a U-Net type network for example; - an artificial neural network of the auto-encoder or variational auto-encoder type, also called VAE in English; - a self-aware or transformative model; - any other model structure capable of receiving as input one or more second images representing the projection of a second light pattern in an environment outside the vehicle in the second wavelength range, and optionally information describing the second projected light pattern, and of predicting depth information of the outside environment (of the projection area of ​​the second light pattern in particular), for example in the form of a depth map of the outside environment.

[0184] No restrictions are attached to the machine learning applied to the second depth information determination model to optimize its second parameters. The machine learning can, for example, be supervised, using a training database comprising associations between: - a second image of a projection of a second light pattern in a scene outside a vehicle, in the second wavelength range, and optionally descriptive information of the second projected light pattern; - a depth map of this same scene, indicating the “ground truth”, from which a loss function can be applied to evaluate the difference between the model prediction and the ground truth, which allows optimization of the second parameters during supervised learning.

[0185] The associations stored in the training database can be, for example: - obtained by accumulating second images and associated depth maps, in various real-world driving situations, for example for varied scenes (city driving, motorway driving, rural driving) and in various real-world weather conditions; or - obtained by simulating second images and associated depth maps, in various simulated driving situations, for example for various simulated scenes (in town, on highway, in the countryside) and in various simulated climatic conditions.

[0186] When the second model is capable of receiving input information describing the second light pattern (in the case of a second light pattern determined by the control module based on descriptive information of environmental conditions or the driving situation), the scenes also vary according to the second light patterns that are projected onto it.

[0187] No restrictions are attached to the loss function used to train the second model, which can be any known loss function. The second model thus trained is specific to the second wavelength range in which the second light pattern is projected, as well as to the relative positioning between the camera acquiring the second image and the light module projecting the second light pattern.

[0188] When the second light pattern is predetermined, or when a set of second light patterns is predetermined, the second pattern thus trained can be specific to this second light pattern or to this set of second light patterns.

[0189] Alternatively, when descriptive information of the second light pattern can be received as input to the model, the model is generic and can be applied to a set of second light patterns that is not predefined, in particular when the first model and the second model are jointly trained as described below.

[0190] As for the first model, a training system and a supervised learning training method for the first model are described below with reference to Figures 9 and 10.

[0191] Figure 8a illustrates a first pixelated light source 200.1, according to a first example, capable of forming two light beams in two distinct wavelength ranges. Thus, such a pixelated light source 200.1 can be included in the first light module 120.1, which is thus capable of projecting both the first light pattern and the second light pattern (the second light module 120.2 being therefore optional).

[0192] The first pixelated light source 200.1 according to the first example comprises: - a first set of individually controllable light elements 801, capable of emitting light in the first wavelength range; - a second set of individually controllable 802 light elements, capable of emitting light in the second wavelength range.

[0193] Thus, the first set 801 of light elements is capable of emitting light in the visible range, and may include, in particular, subsets of blue, red, and green light elements, which makes it possible to project a colored beam of light into the scene outside the vehicle, to display the first light pattern. Alternatively, each element of the first set 801 may be capable of emitting white light.

[0194] The light elements of the first set 801 can be controlled to form a first light beam according to the first light pattern described above, with a resolution depending on the number of light elements of the first set 801.

[0195] Simultaneously, the light elements of the second set 802 can be controlled to form a second light beam according to the second light pattern, with a resolution depending on the number of light elements of the second set 802.

[0196] According to an unshown variant, the pixelated light source 200.1 further comprises a third set of light elements capable of projecting light in a third wavelength range. In particular, a third light pattern can be projected in the third wavelength range, and depth information can be determined by the processing module 150 based on deformations induced by the vehicle's external environment on both the second and third light patterns.

[0197] Figure 8b illustrates a pixelated light source 200.1, according to a second example, capable of forming two light beams in at least two distinct wavelength ranges. Thus, such a pixelated light source 200.1 can be included in the first light module 120.1 which is therefore capable of projecting both the first light pattern and the second light pattern (the second light module 120.2 is therefore optional).

[0198] The pixelated light source 200.1 according to the second example comprises two light sources 811.1 and 811.2 and a DMD micromirror array.

[0199] The first light source 811.1 is capable of emitting light rays in the first wavelength range. The second light source 811.2 is capable of emitting light rays in the second wavelength range. According to an alternative embodiment not shown, the pixelated light source 200.1 further comprises a third light source capable of emitting light rays in a third wavelength range, for projecting the third light pattern described above.

[0200] The light sources 811.1 and 811.2 can be arranged on the same support 812, as shown in [Fig.8b], or on two separate supports.

[0201] Only one micromirror 810 is shown in [Fig.8b] to facilitate understanding of the figure. However, in practice, the array comprises a large number of micromirrors, in particular more than 100, or even more than 1000 or more than 10,000 micromirrors individually controllable by the first control unit 201.1 described above.

[0202] Each micro-mirror 810 can be controlled to switch between at least one first position 820.1 and a second position 820.2. In the first position 820.1, the micro-mirror is arranged to reflect a light beam from a source, for example a light beam 813 emitted by the first light source 811.1, towards the first projection system 202.1 so as to contribute to the formation of the first light beam outside the vehicle 100. Thus, by controlling the positions of the micro-mirrors of the matrix by the first control unit 201.1, the first light module 120.1 projects a first light beam according to the first light pattern, in the first wavelength range, when the first light source 811.1 is activated. Similarly, by controlling the positions of the matrix micro-mirrors by the first control unit 201.1, the first light module 120.1 projects a second beam of light according to the second light pattern in the second wavelength range.

[0203] The first control unit 201.1 can thus control the same micromirror array, alternately at a given frequency, to project the first pixelated beam and the second pixelated beam alternately. The switching frequency between the two beams can be greater than 700 Hz, so that the flicker of the first pixelated beam in the visible range is not perceptible to the human eye and therefore does not disturb the driver and other road users. of the road, and not perceptible by the first 140.1 camera, having an exposure time greater than the period associated with the switching frequency.

[0204] Figure 9 illustrates a 900 drive system of a first model of generating a light intensity map corresponding to a second light pattern, according to embodiments of the invention.

[0205] Such a drive system 900 is external to the vehicle 100 and is capable of implementing the drive phase of the process according to the invention, which is described later with reference to [Fig. 10]. The drive phase precedes the current phase described with reference to [Fig. 3], which takes place after the vehicle 100 has been put into service. The drive phase may, in particular, be part of the design and manufacturing process of the control module (and of the processing module 150 when the second model is driven jointly with the first model), before its integration into the vehicle 100.

[0206] The training system 900 includes a training database 901, capable of storing training data associations, each training data association comprising: - a representative training image of a scene in a wavelength range (for example in the first wavelength range or in the second wavelength range); - ground truth which includes reference data corresponding to reference depth information, for example a reference depth map for the training image scene.

[0207] Preferably, the associations stored in the training database are varied, that is to say, they were obtained in outdoor environments varying according to several criteria: scene composition, scene brightness level, weather conditions, etc.

[0208] In addition, the 901 training database includes more than one hundred training data associations, and preferably several thousand or even several tens of thousands of training data associations.

[0209] No restrictions are attached to the manner in which the 901 training database was constructed. The 901 training database can be obtained from images captured by a fleet of vehicles in real-world driving situations, with a depth camera and / or lidar capable of obtaining the reference data corresponding to each captured training image.

[0210] The 900 drive system further comprises the first model, referenced 910, whose initial parameters are to be trained. The first model 910 is structurally capable of generating a light intensity map of a given resolution, a resolution which depends on the second matrix source 200.2 (or the first matrix source 200.1 when it is capable of projecting in the second wavelength range) with which the control module is associated in the vehicle 100 during the current phase, from a representative image of an environment outside a vehicle, an image which may be a training image during the training phase, and which is an image captured in real time by at least one camera during the current phase.

[0211] To this end, the first model may have one of the following structures: - a convolutional neural network, such as a U-Net type network for example; - an artificial neural network of the auto-encoder or variational auto-encoder type, also called VAE in English; - a generating network of a system of generative adversarial networks, also called GAN, for “Generative Adversarial Networks” in English; - a self-aware or transformative model; or - any other model structure capable of receiving as input an image representing an external vehicle environment, and of producing as output a light intensity map in a format suitable for controlling the previously described light module, for projection of the second light pattern in the second wavelength range.

[0212] The training system 900 further comprises a synthesis module 902, capable of producing at least one synthetic image representative of the scene in the training image, illuminated by a second light pattern corresponding to the light intensity map (called the training light intensity map during the training phase) produced at the output of the first model 910, in the second wavelength range. Thus, the synthesis module 902 makes it possible to simulate the capture by a camera of the environment represented by the training image, into which a second light pattern corresponding to the training light intensity map at the output of the first model 910, in the second wavelength range, would be projected. To this end, the synthesis module 902 receives as input: - the training image; - at least one training light intensity map generated by the first 910 model.

[0213] The 900 training system further includes a second training model, referenced 903, which is a model for determining depth information (called training depth information during the training phase).

[0214] The second training model is preferably identical to the second depth information determination model integrated into the module of The first model 910 is specifically trained to improve the performance of the second model, with which it is integrated into the vehicle 100 during the current phase of the process according to the invention, described above.

[0215] According to one embodiment, the second training model 903 is already optimized, i.e. the second parameters of the second model 903 have been optimized during a dedicated training phase of the second model (and implemented before the implementation of the process according to [Fig. 10]), and the second parameters do not vary during the training phase of the first model.

[0216] According to one variant, the second training model 903 is trained at the same time as the first model 910, and the second model 903 from the training phase can be implemented in the vehicle 100 for the current phase in the processing module 150.

[0217] The second 903 training model is suitable for receiving: - at least one computer-generated image from the 902 synthesis module; and - optionally, at least one training light intensity map generated by the first 910 model after receiving the training image.

[0218] Depending on the information received, the second training model 903 is able to generate depth information in the scene represented by the training image, such as a depth map.

[0219] The 900 training system further includes a 904 loss assessment module, which is suitable for: - to assess a loss by comparison between the ground truth reference data from training database 901 and the outputs of the second training model 903 (training depth information); - to modify one or more first parameters of the first model according to the evaluated loss, according to a predefined training strategy, and optionally (when the second model 903 is jointly trained) to modify one or more second parameters of the second training model 903.

[0220] No restrictions are attached to the loss function used, nor to its application to the comparison between the ground truth reference data and the outputs of the second training model 903.

[0221] Such a training system 900 is thus capable of implementing supervised learning of the first model 910, and optionally of the second model jointly.

[0222] Alternatively, the first model and / or the second model may be derived from machine learning other than supervised learning. For example, the first model and / or the second model may be optimized by machine learning. reinforcement, with a reward determined based on the performance of the first or second model.

[0223] The [Fig. 10] is a diagram illustrating the steps of a training phase of the first 910 model implemented in a vehicle control module, according to embodiments of the invention.

[0224] As previously explained, the training phase of the first model 910 can be implemented in the 900 drive system described with reference to [Fig.9].

[0225] In addition, as previously stated, the second model can also be trained during the implementation of the process illustrated with reference to [Fig. 10].

[0226] At a step 1000, the training database 901 obtains a training data association as previously described, comprising a training image representative of an environment outside a vehicle, and reference depth information representative of ground truth.

[0227] At a step 1001, the training image obtained is submitted as input to the first model 910 which generates a training light intensity map as a function of the training image obtained.

[0228] The training light intensity map is transmitted from the first model 910 to the synthesis module 902, which also receives the training image from the training database 901. The first model 910 can further transmit the generated training light intensity map to the second training model 903.

[0229] At a step 1002, the synthesis module 902 determines a synthetic image representative of the scene of the training image, illuminated by a second light pattern corresponding to the light intensity map produced at the output of the generation model 108, in the second wavelength domain.

[0230] The synthetic image is transmitted to the second training model 903, which determines depth information of the scene corresponding to the training image at a step 1003, to obtain depth information.

[0231] The determined depth information is transmitted from the second training model 903 to the loss evaluation module 904.

[0232] At a step 1004, the loss assessment module 904 assesses a loss based on the ground truth reference data received from the training database 901, and based on the depth information received.

[0233] At a step 1005, the loss evaluation module 904 can further determine whether a predefined convergence criterion is met or not, based in particular on the loss evaluated at step 1004, and optionally on losses evaluated during previous iterations of steps 1000 to 1004.

[0234] If the convergence criterion is not met, the loss evaluation module 904 modifies at least one first parameter of the first model 910 at a step 1006, depending on the evaluated loss, according to the predefined training strategy. Such parameter optimization training strategies during supervised learning are well known and are not described further in this description.

[0235] Furthermore, in the embodiment described above in which the second training model 903 is trained jointly with the first model 910, the loss evaluation module 904 can modify at least a second parameter of the second model 903, at a step 1007, as a function of the evaluated loss, according to the predefined training strategy.

[0236] Following step 1006, or following step 1007 when implemented, the process training phase returns to step 1000 to repeat steps 1000 to 1005 on the basis of a new association of training data from training database 901.

[0237] The first model 910 can thus be trained to convergence. In addition, the first model 910 and the second model 903 are both trained to convergence during the same training phase.

[0238] At step 1008, when the loss evaluation module 904 determines that the convergence criterion has been met, the training phase is completed and the first model 903 can be implemented in the control module for implementation of step 302 (for determining the second light pattern) of the current phase of the process according to the invention. Furthermore, when step 1007 is implemented, the second trained model 903 can be implemented in the processing module 150 of the vehicle 100, to determine the depth information transmitted to the control module during step 304.2 described above.

[0239] According to some embodiments, the first model is further capable of taking as input a previous light intensity map when determining the light intensity map, and the supervised learning described with reference to [Fig. 10] can be adapted in one of the following ways: - during step 1000, a random light intensity map is generated, and the synthesis module 902 simulates the projection of a second light pattern corresponding to the random light intensity map onto the training image to obtain a synthetic image, and the synthetic image as well as the random light intensity map are submitted as input to the first model 903 during step 1001; or - the associations of the training base 901 include training images forming a sequence, the training images being submitted sequentially during the iterations of steps 1000 to 1006 (or 1007); or - during step 1000, the synthesis module 902 simulates the projection of a second light pattern corresponding to the previous light intensity map into the newly obtained training image, to obtain a synthesis image, and the synthesis image together with the previous light intensity map are submitted as input to the first model during step 1001.

[0240] The present invention is not limited to the embodiments described above by way of example; it extends to other variants.

Claims

1. Demands Method for controlling at least one light module (120.1; 120.2) for a motor vehicle (100), the at least one light module comprising at least one pixelated light source (200.1; 200.2) and being capable of projecting a first pixelated light beam in a first wavelength range and of projecting a second pixelated light beam in a second wavelength range, the first wavelength range comprising visible wavelengths and the second wavelength range comprising non-visible wavelengths, the method comprising the following steps: - upon obtaining (301) information indicating the presence of a person (410; 510) in an environment outside the motor vehicle determination (302) of a first light pattern (400; 500) to be projected into the external environment in the first wavelength range and of a second light pattern (430; 530; 620) to be projected into the external environment in the second wavelength range in a common projection area with the first light pattern; - control (303.1; 303.2) of at least one light module to form the first light pattern in the first projected pixelated light beam and to form the second light pattern in the second projected pixelated light beam; wherein the first light pattern includes at least one interactive light element (402.1; 402.2; 501.1-501.4) and wherein the method further comprises the following steps: - obtaining (304.1; 304.2) at least one descriptive information of a relative position between the first projected light pattern and the person, said at least one descriptive information including depth information of the external environment determined from an image (610) representative of the projection of the second light pattern into the external environment in the second wavelength range; - determination (305) of an interaction between the person and an interactive light element from said at least one interactive light element of the first light pattern, depending on the less descriptive information of the relative position between the first light pattern and the person; - implementation (306-308) of a determined action based on the interactive light element for which the interaction with the person is determined.

2. Method according to claim 1, wherein the determination (302) of the first luminous pattern (400; 500) to be projected is a function of a position of said person.

3. A method according to claim 1 or 2, wherein at least one relative position data comprises: - a first image acquired by at least one camera (140.1) of the vehicle (100) in the first wavelength range and representative of the projection of the first light pattern (400; 500) into the external environment; and - the depth information of the external environment determined from an image (610) representative of the projection of the second light pattern (430; 530; 620) into the external environment in the second wavelength range, referred to as the second image.

4. A method according to any one of claims 1 to 3, wherein the action carried out comprises a modification (307) of the first light pattern (400; 500) by controlling at least one light module (120.1) to form the first modified light pattern in the first projected pixelated light beam.

5. A method according to claim 4, wherein the modification (307) of the first light pattern (400; 500) comprises: - a modification of a shape and / or a color of the interactive light element for which the interaction with the person is determined; - a deletion of the interactive light element for which the interaction with the person is determined.

6. A method according to any one of claims 1 to 5, wherein the action implemented includes adapting (308) a driving parameter of the vehicle (100) as a function of the interactive light element (402.1; 402.2; 501.1-501.4) for which the interaction with the person (410; 510) is determined.

7. A method according to any one of the preceding claims, wherein the light pattern (400; 500) comprises several interactive light elements (402.1; 402.2; 501.1-501.4).

8. A method according to claim 7, wherein the first light pattern (400; 500) comprises a projected light element (401) indicating a question, and N interactive light elements corresponding to respective answers, N being an integer greater than or equal to 2.

9. A method according to any one of the preceding claims, wherein the second light pattern (430; 530; 620) comprises dark areas (601) and illuminated areas (602) in the second wavelength range.

10. A method according to claim 9, wherein the dark areas (601) and the lit areas (602) have identical shapes and sizes and form a checkerboard pattern.

11. A method according to any one of the preceding claims, wherein the second light pattern (430; 530; 620) is a single predetermined light pattern.

12. A method according to any one of claims 1 to 10, wherein the second light pattern (430; 530; 620) is selected from a set of predetermined second patterns, based on information about the external environment of the vehicle and / or representative of a vehicle driving situation (100).

13. A method according to any one of claims 1 to 10, wherein the second light pattern (430; 530; 620) is generated from information about the external environment of the vehicle and / or representative of a vehicle driving situation (100).

14. A method according to claim 13, wherein the generation of the second light pattern (430; 530; 620) comprises applying a first light intensity map generation model (910) to at least one image representative of the external environment to obtain a light intensity map corresponding to the second light pattern (430; 530; 620), the light intensity map indicating light intensity values ​​to control light elements (210.1; 210.2) of a pixelated light source (200.1; 200.2) of a light module (120.1; 120.2) of the vehicle (100) capable of projecting the second pixelated light beam in the second wavelength range.

15. A method according to any one of the preceding claims, wherein the depth information is obtained by applying a second depth information determination model (903) to an image (610) representative of the projection of the second light pattern (430; 530; 620) into the external environment in the second wavelength range.

16. Computer program comprising instructions for carrying out the method according to any one of claims 1 to 14, when these instructions are executed by a processor (701).

17. Control module (130; 201.1) of at least one light module (120.1; 120.2) for a motor vehicle (100), the at least one light module comprising at least one pixelated light source (200.1; 200.2) and being capable of projecting a first pixelated light beam in a first wavelength range and of projecting a second pixelated light beam in a second wavelength range, the first wavelength range comprising visible wavelengths and the second wavelength range comprising non-visible wavelengths, the control module comprising a processor (701) configured to: - upon obtaining information indicating the presence of a person in the vehicle's environment, determine a first light pattern (400; 500) to be projected into the external environment in the first wavelength range and a second light pattern (430; 530;620) to project into the external environment in the second wavelength range in a common projection area with the first light pattern; - to control at least one light module to form the first light pattern in the first projected pixelated light beam and to form the second light pattern in the second pixelated light beam; wherein the first light pattern includes at least one interactive light element and wherein the processor is further configured to: - obtain at least one descriptive information of a relative position between the first projected light pattern and the person, said at least one descriptive information including depth information of the external environment determined from a representative image of the projection of the second pattern; luminous in the external environment in the second wavelength range; - determine an interaction between the person and an interactive light element from said at least one interactive light element of the first light pattern, based on at least one descriptive information of the relative position between the first light pattern and the person; - to implement a specific action based on the interactive light element for which the interaction with the person is determined.

Images